2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy...
-
Upload
luis-carrasco -
Category
Documents
-
view
218 -
download
0
Transcript of 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy...
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 111
A hierarchical methodology for the integral net energy designof small-scale hybrid renewable energy systems
David Hernaacutendez-Torres an Alberto J Urdaneta Urdaneta b Paulo De Oliveira-De Jesus c
a Laboratoire Plasma et Conversion d Energie (LAPLACE) UMR5213-CNRS-INPT-UPS Toulouse Franceb Fundacioacuten para el Desarrollo del Servicio Eleacutectrico Caracas Bolivarian Republic of Venezuelac Department of Conversion and Energy Delivery Universidad Simoacuten Boliacutevar Caracas Bolivarian Republic of Venezuela
a r t i c l e i n f o
Article historyReceived 16 March 2015
Received in revised form
20 May 2015
Accepted 6 July 2015
Keywords
Alternative renewable energy
Power generation
Levelized costs
Analytic hierarchy process
a b s t r a c t
In this paper a two level hierarchical methodology is presented for the integral design of hybrid powergeneration systems based on alternative renewable energy (ARE) systems using the net energy concept
and considering technical economical societal as well as environmental aspects Results are presented
for the design of a small-scale hybrid renewable energy system using the proposed methodology in
Margarita Island Venezuela The proposed methodology applies the integrated analytical hierarchy
process (AHP) methodology to the selection of the best hybrid renewable energy system using the
concept of net energy and is divided in two phases a classic optimization process using levelized costs
minimization and an AHP implementation for decision making problems Under this methodology
technicalndasheconomical aspects are considered as quantitative parameters while socialndashenvironmental
aspects depend largely on the criteria of the system planning engineers and future users of the system
Technicalndasheconomical aspects are considered using specialized software used to optimize and compute
the best con1047297guration of an ARE project Socialndasheconomical aspects are de1047297ned as a series of parameters
that should be considered by the planning team a meeting of experts or a community consensus on the
project site Several diesel price scenarios (low intermediate and high) are considered The results show
the importance of the proposed tools for decision making problems
amp 2015 Elsevier Ltd All rights reserved
Contents
1 Introduction 100
2 Proposed methodology 102
21 Analytic hierarchy process (AHP) 102
22 The AHP method and the ARE sources 103
23 Energy return on energy invested (EROEI) 103
3 Study case 104
31 Low diesel price scenario 105
32 High diesel price scenario 107
33 Intermediate diesel prices scenarios 108
4 Conclusion 109
Acknowledgements 109
Appendix A Pair-wise matrices for low diesel price case study scenario 109Appendix B Pair-wise matrices for high diesel price case study scenario 109
References 110
1 Introduction
The future of the energy exploitation in our planet is bounded
to many challenges such as the achievement of the diversi1047297cation
of the primary energy sources Although recently oil prices have
Contents lists available at ScienceDirect
jo ur nal ho me pag e wwwelseviercomlocaterser
Renewable and Sustainable Energy Reviews
httpdxdoiorg101016jrser201507008
1364-0321amp 2015 Elsevier Ltd All rights reserved
n Corresponding author
E-mail addresses dhernandezlaplaceuniv-tlsefr (D Hernaacutendez-Torres)
aurdanetafundelecgobve (AJ Urdaneta Urdaneta)
pdeoliveirausbve (P De Oliveira-De Jesus)
Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 211
had an important decrease the alternative renewable energy
(ARE) sources are signaled to play an increasingly more important
role in this diversi1047297cation process Renewable energy sources have
grown from 01 of total of primary energy in 1973 (6106 MToe) to
11 of 13 371 MToe in 2013 [1] This implies an increase of 2400
in 40 years according to International Energy Agency [1]
Several reasons can explain the emergence of renewable
energy as demography conventional energy scarcity equilibrium
of planetary resource use Climate change is a fact Traditionallythe role of ARE sources has been referred to as part of the solution
to global warming [2] Also 80 of global energy resources are
used by 20 of the population living in developed countries [3]
ARE sources are an instrument to address the development of
world population instead of using conventional energy resources
Some advantages of ARE source are their participation on
energy diversi1047297cation their modularity and the capability of
decentralized operation They may contribute to equitable geo-
graphical energy distribution in rural countries and of course ARE
source are environmental friendly when compared to some con-
ventional sources as fossil fuel In a regional context ARE sources
are expected to play a more important leading role since wind
energy sources show a signi1047297cant development potential [4] A
very interesting comparison includes the concept of net energy
where the energy ef 1047297ciency of the given sources is computed
namely a focus is given to ldquohow much energy is needed to produce
energyrdquo [5] In this net energy analysis ARE sources are well
placed when compared to hydrocarbon sources
The analysis of the important role that ARE sources may play
includes the following factors
Their participation in the energy diversi1047297cation process Inevitable participation in the national energy approach as
conventional source are becoming scarce (as big scale hydro-
power) oil pro1047297tability should be maximized through exports
or due to immediate unavailability of some resources at big
scale as in the case of natural gas Identi1047297cation of the politicalndashterritorial importance of ARE
installations on remote and isolated sites with special interestin strengthening the nationals developing axes
Active role of ARE source in network stability when considered
as distributed energy sources with migration towards smart
grids networks In1047298uence on the electricity balance and reduction of power
transmission limits with installation of ARE sources near
energy consumption centers
The development of ARE resources are strongly dependent on
energy decision makers preferences These decision makers are
generally energy stakeholders that participate in the energy
market as investors and operators In general ARE projects are
strongly encouraged by national or regional regulatory boards
with strong incentives in order to achieve some economy of scalein alternative generating technologies Within this context when a
energy decision maker prospects the implementation and use of
ARE sources an ef 1047297cient methodology for a technicalndasheconomical
selection is required
There exists in literature several contributions about how to select
an appropriate electrical generating scheme based upon ARE sources
A comprehensive review can be found in [6] with a survey of
methodologies and computational tools to address this problem We
will highlight two of them First the RETscreen platform [7] provides
an option for an initial analysis of technical and 1047297nancial viability of potential renewable energy Another well developed tool is HOMER
[8] which is able to simulate and 1047297nd the best selected option for
energy source con1047297gurations in a given project In this paper it is
considered that the HOMER tool is known by the reader however a
detailed user manual is given in [8]
However despite ongoing ARE evaluation methodologies have
strong strengths on system optimization of technical and eco-
nomic variables It is observed that social and environmental
aspects are hard to treat and valuate in these proposals Quantify-
ing and evaluating the social and environmental viability of a
project is a dif 1047297cult task The analytic hierarchy process (AHP) can
be applied as a complementary evaluation technique in order to
evaluate societal and environmental impact aspects of widespread
integration of ARE sources The methodology proposed and
developed in [9] gives the possibility of classifying the different
analysis criteria into a hierarchical model thus solving the
problem of selection between several solution alternatives The
use of a de1047297ned weighting system makes it easier to classify
alternatives an criteria entries by applying a pair-wise comparison
We can 1047297nd some background on AHP application for ARE
planning purposes Comprehensive reviews on AHP application
for ARE resource planning and project valuation can be found in
[10ndash12] In the following we discuss some of them An analytic
hierarchy process has been applied by [1314] for renewable
energy resources in Jordan The application of an AHP method
for Chinese ARE sources is found in [1516] In [1718] the use of the
AHP method is proposed for decision making and planning of ARE
generation projects in Canada and India respectively Similarly in
[1920] the AHP method is applied for renewable energy planningin Malaysia and Indonesia respectively An integrated VIKOR ndashAHP
methodology to the selection of the best energy policy and
production site is developed in [21] In the work presented in
[22] an integrated methodology is proposed combining linear
optimization and AHP The use of different advanced techniques
along with the AHP method has also been formulated in the
literature In [2324] fuzzy logic is used for the AHP approach for
an ARE application in Brazil and Korea respectively An analysis of
the energy return on energy invested impact of ARE sources to the
environment is presented in [25]
This paper presents an integral approach that considers a
technicalndasheconomical analysis for ARE sources technology selec-
tion but also societalndashenvironmental aspects are considered dur-
ing the analysis of the project considering the net energy or EROEIconcept [26] The use of a two level analytic hierarchy process
Fig 1 Proposed methodology
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 101
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 311
(AHP) is proposed for this purpose The goal is to de1047297ne a
parameter that quanti1047297es the energy ef 1047297ciency of energy sources
The results of the technicalndasheconomical optimization problem
obtained by linear levelized cost minimization are used as an
input to the decision making model based in AHP The methodol-
ogy has been coded using basic programming language and two
study cases are discussed
This paper is divided in three sections In a 1047297rst section the
proposed methodology is presented In the second section themethodology is presented and formulated for a study case Finally
in the last section several results are presented and discussed
2 Proposed methodology
The methodology proposed in this paper is divided in two
levels The 1047297rst part involves the technicalndasheconomic analysis of
the project The second part deals with the socialndashenvironmental
analysis The goal of the methodology is to 1047297nd the best adapted
ARE sources con1047297guration for a given project where both techni-
calndasheconomical and socialndashenvironmental aspect are integrally
considered
The experts of the system may initially de1047297ne the search matrix
for the optimization problem formulation as proposed in [27] Theproposed strategy is also based on the results obtained in a 1047297rst
preliminary analysis using the software in [8] and considering only
technicalndasheconomical aspects A second analysis includes the
socialndashenvironmental aspects using the AHP method As the AHP
is based on weighted classi1047297cation by pair-wise comparison it is
proposed that some of the weighting vectors should be directly
completed from the results obtained in the 1047297rst analysis using [8]
The criteria evaluation of the AHP method may be performed by
the same initial experts of the system The optimization process is
given by exhaustive by time-domain simulations with varying
system components size The different component sizes consid-
ered during simulation may be resumed in a multivariable search
matrix This methodology structure is resumed in Fig 1
21 Analytic hierarchy process (AHP)
The AHP is a method based on the formulation of a hierarchy
structure that is used to deal with complex decisions The method
classi1047297es the choices or solutions to a given problem by evaluation
according to different criteria and a de1047297ned weighting system
Weights are de1047297ned by a pair-wise comparison simplifying the
selection process
In the AHP method problems are modeled as hierarchies This
process is shown in Fig 2 In many cases each criteria may be
divided in several sub-criteria to further analyze the details of the
studied problem
Pair-wise comparisons are used to establish the importance
weight 1047297xed to each criteria These comparisons are based on thescale system proposed by [9] This scale is shown in Table 1
Comparisons using this scale in each level of analysis are used
to build the pair-wise comparison matrices Using the procedure
detailed in [9] the weights associated to each comparison can be
obtained For each comparison matrix two elements of interest are
computed the associated weight vector and the matrix incon-
sistency ratio This last value is used to estimate the consistency of
pair-wise comparisons Consider for example that a problem or a
criterion that needs to be evaluated for three candidate options
Consider for illustration purposes that option 1 is slightly more
important (factor 3 in the pair-wise comparison scale) than option
2 and that option 1 is strongly more important than option 3
(factor 5 in the pair-wise comparison scale) The problem now is tode1047297ne the pair-wise comparison between options 2 and 3 Not a
trivial solution since a system of equations needs to be solved A
graphical solution to this problem is presented in Fig 3 It may be
considered as an approximate solution that the comparison can
take a value of 2 in the proposed scale With these values the pair-
wise comparison matrix for this example is given by
Criterion Op1 Op2 Op3 Weight
Op1 1 3 5 06483
Op2 13 1 2 02297
Op3 15 12 1 01220
with an inconsistency ratio of 00036
In [9] it is considered that an inconsistency ratio lower than10 means that decisions expressed in the pair-wise matrix are
consistent Simple programming language is used to compute
Fig 2 The AHP method
Table 1
Pair-wise comparison table
Numerical rating Verbal judgment of preference between alternatives i and j
1 i is equally important to j
3 i is slightly more important than j
5 i is strongly more important than j
7 i is very strongly more important than j
9 i is extremely more important than j
2468 I ntermediate values
1 15 2 25 3 35
0
001
002
003
004
005
006
Fig 3 Illustrative example for computing inconsistency ratio
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110102
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 411
weights and inconsistency ratios The eigenvectors e of the pair-
wise comparison matrices are given by
e frac14 eigs M λmaxI n
eth1THORN
where M is the pair-wise decision matrix λmax is the maximum
eigenvalue and I n is the identity matrix of size n the size of M The
weight vector is obtained by computing the average values of each
row of the eigenvector
The inconsistency ratio is obtained by 1047297
rst computing theconsistency index
CI frac14 eth λmaxnTHORN
ethn1THORN eth2THORN
Then the inconsistency ratio is given by
CR frac14CI
RI eth3THORN
where RI is the random index given in Table 2
22 The AHP method and the ARE sources
The goal now is to de1047297ne a methodology based on the AHP
method for hybrid power generation projects using ARE sources A
1047297rst classi1047297cation of different criteria is proposed according toimportant variables considered as candidates to have signi1047297cant
weight on the decision making problem
Given the problem complexity and the quantity of variables to
be considered the classi1047297cation is simpli1047297ed to two sets of criteria
technicalndasheconomical aspects and socialndashenvironmental aspects A
similar classi1047297cation is proposed in [17] Each set of criteria is then
divided with the following structure
I ndash Technicalndasheconomical aspects (TecEconAsp)
A Resources availability (AvailRes) A1 Total energy produced (for network connected systems)
or capacity shortage (for stand-alone systems)
A2 Multiple equipment location possibilities
A3 Type of energy constant or intermittent B Technical feasibility (TecFeas)
B1 Technological maturity and local experience with the
technology
B2 Availability of the technology B3 Net energy associated
C Financial viability (FinVia)
C1 Total cost of the project in net present cost (NPC)
C2 Total levelized cost-of-energy (LCOE) in $kWh C3 Total operational and maintenance costs
II ndash Socialndashenvironmental aspects (SocEnvAsp)
D Ecological impact (EcolImp)
D1 Reduction or substitution of CO2 emissions
D2 Reduction of impact on surrounding ecosystem
degradation D3 Visual or sound impact on surrounding populated areas
E Educational potential (EduPot)
E1 The project is tangible and visible within the
community E2 Promotes awareness and public interest on energy
problems
E3 Community participation F Ecological and social impacts (EcoSocImp)
F1 The project promotes economical activity
F2 The project promotes social development
Using this classi1047297cation the hierarchy structure proposed is
given in Fig 4
23 Energy return on energy invested (EROEI)
One of the main contributions of the proposed methodology
consists in the incorporation of the use of the net energy concept
in the pair-wise comparison matrices at levels 2ndash4 of the
described methodology This concept and the energy return on
energy invested (EROEI) for each ARE source can be found in [5]
and [26] respectively
The goal is to include a parameter that quanti1047297es the energy
ef 1047297ciency of energy sources Instead of concentrating analysis
efforts on costs associated to energy production focus is given toldquohow much energy is needed to obtain energyrdquo
The EROEI is given by equation
EROEI frac14UAE =EE eth4THORN
where UAE is the usable acquired energy and EE is the energy
expended
In a 1047297rst level the EROEI variable is taken into account by
considering an important weight to variable B3 in the ldquotechnical
feasibilityrdquo pair-wise matrix On a second level the EROEI will play
an important role when comparing ARE sources technologies by
pair A classi1047297cation of energy sources technologies as a function of
their EROEI ratio is presented in [26] Accordingly wind turbines
have an EROEI of 181 while for solar photovoltaic this value is
around 681 This means that when evaluating the hybrid ARE
solution candidates from the HOMER optimization a higher scaled
weight will be granted to options with a greater nominal power
from wind turbines when contrasted to less favorable EROEI for
solar photovoltaic technologies For comparison purposes an
Table 2
Average random index matrix
Size of M 1 2 3 4 5
Random index 0 0 058 09 112
Size of M 6 7 8 9 10
Random index 124 132 141 145 149
Fig 4 Proposed hierarchy structure
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 103
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 511
example of a renewable energy with high EROEI is hydropower
with typical ratio higher than 1001 In the other hand biodiesel
and bitumen oil from tar sands (as in the Venezuelan Orinoco oil
belt) are among the energy sources with the lower EROEI ranging
from 13 to 41 The impacts of these assumptions are discussed
later in the study case section
The matrices for levels 0 1 and 2 are now presented they are
constructed under the assumptions described before They can
also contain the results of an expert meeting or a community pollevaluating the different pair-wise comparisons When selecting
the pair-wise comparison values in a three dimension matrix at
least one value is specially selected to minimize the inconsistency
ratio from Eq (3)
Goal TecEconAsp SocEnvAsp Weights
TecEconAsp 1 3 075
SocEnvAsp 13 1 025
TecEconAsp A AvailRes BTecFeas CFinVia Weights
A AvailRes 1 3 5 06370
BTecFeas 13 1 3 02583CFinVia 15 13 1 01047
SocEnvAsp DEcolImp EEduPot FEcoSocImp Weights
DEcolImp 1 3 13 02684
EEduPot 13 1 14 01172
FEcoSocImp 3 4 1 06144
A AvailRes A1 A2 A3 Weights
A1 1 7 1 04869
A2 17 1 15 00778
A3 1 5 1 04353
BTecFeas B1 B2 B3 Weights
B1 1 13 19 00704
B2 3 1 15 01782
B3 9 5 1 07514
CFinVia C1 C2 C3 Weights
C1 1 1 3 04286
C2 1 1 3 04286
C3 13 13 1 01429
DEcolImp D1 D2 D3 Weights
D1 1 5 7 07306D2 15 1 3 01884
D3 17 13 1 00810
EEduPot E1 E2 E3 Weights
E1 1 2 13 02493
E2 12 1 13 01571E3 3 3 1 05936
FEcoSocImp F1 F2 Weights
F1 1 15 01667
F2 5 1 08333
In the following sections the practical use of the AHP method is
given with some application examples For an evaluation of the
proposed methodology different study scenarios are presented
3 Study case
To illustrate the proposed planning methodology a study case
is presented in this section The principle is to design a hybrid
power generation system based on ARE sources in Margarita
Island in north-east Venezuela The application is based on an
isolated home without connection to the national electrical grid
For this study case solar wind and fuel resources are con-
sidered as available on the project site The diesel fuel is available
for portable electric power generation Data of wind resource in
Margarita Island are obtained from [28] Data for solar irradiance
was obtained from 10 min measurements in the central Venezue-
lan region (Caracas) These measures are available at httpcbm
usbveclima For illustration purposes of this example the statis-
tical data for solar irradiation in the central Venezuelan region is
used as the solar resource in the island Using an extrapolation to
the island geographical coordinates the HOMER is able to generate
stochastic time series hourly data for both wind and solar
resources Input data of equipment costs and electrical load pro1047297le
were obtained from typical HOMER data library The load pro1047297le
considered is assumed as a typical representation of small house
electrical energy consumption The availability curves of wind and
solar resources and the electrical load pro1047297le considered are
presented in Figs 5 and 6 respectively
Before de1047297ning the possible candidate options for a successful
decision making strategy comparisons associated with levels 0ndash2
(cf Fig 4) may be de1047297ned For this the assumptions considered for
the study are now de1047297ned These assumptions may change for
different project designs and as before in their de1047297nition may be
involved the planning team expert groups or even the organized
community
Some of the assumptions for the study and construction of thepair-wise comparison matrices are as follows
Fig 5 Solar and wind resources considered for Margarita Island
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110104
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 611
Within the technical and economic aspects the availability of
resources is considered as a key parameter with a higher
importance to the technical feasibility and economic viability In the social and environmental aspects it is given a slightly
higher importance to economic and social impact in the
community over the ecological impact and the educational
potential of the project A higher importance is considered for the following para-
meters the amount of energy produced the energy type
(continuous intermittent etc) the concept of net energy and
the importance in promoting social development in the com-
munity where the project is executed
Several scenarios for this study case are now presented for
different diesel prices with low intermediate and high price levels
31 Low diesel price scenario
In this 1047297rst case study the application example presented in
the previous section is considered under the assumption that
diesel fuel is available at current Venezuelan market prices ie
0029$L (price for July 2012)
The design of the hybrid system considers solar panels wind
turbines and diesel generators as energy sources The design
problem is reduced to 1047297nding the optimal size of each system
component For the speci1047297c hybrid generator system problem
search variables are the size (nominal power in W) of the powerconverter the battery stack the solar photovoltaic panels the
wind turbines and the diesel generator A backup battery system is
considered for increased system reliability The power converter
in this case an inverter is necessary to couple DC and AC buses
Additional converters may be considered for DCDC coupling
between ARE sources For example boost or buck converter may
be used to improve system controllability and operating perfor-
mance (improved ef 1047297ciency control to optimal operating point
MPPT maximum power point tracker algorithm etc) Restrictions
of 10 minimum capacity shortage and 25 minimum ARE sources
participation are considered With the given resource availability
data and with several proposed hybrid generation system con1047297g-
urations the HOMER is used for optimization and simulation of
results
With the system topology de1047297ned a search matrix is obtained
and it is possible to run the optimization algorithm using HOMER
to 1047297nd the optimal solution When the optimization process
completed the 1047297rst three solutions are presented in Table 3 They
are classi1047297ed in descending order by their total net presentcost (NPC)
The option 1 consists of 025 kW in solar panels 2 SW AIR X
400 W wind turbines a 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Option 2 is composed by
4 AIR X 400 W wind turbines 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Finally option 3 is given by
075 kW in solar panels 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter All batteries are model
S4KS25P 76 kWh (4 V 1900 Ah)
In Fig 7 the simulation results of the operating behavior of the
obtained system con1047297gurations are presented With these results
it is clear that the diesel generation participation is high This is
normal form of a NPC optimization since fuel prices are very low in
Venezuela These three options are 1047297xed as the candidate solutionsfor the second part of the methodology presented in the previous
section using the AHP method
The second part of the proposed methodology deals with the
evaluation of candidate options using the AHP method For this
each option is weighted by comparison with all the selected sub-
criteria (cf Fig 4) Previously described assumptions are used to
evaluate pair-wise comparisons between options and sub-criteria
Additionally as a reference an analysis of the negative impact of
ARE sources to the environment is presented in [25]
In this paper it is proposed that some evaluation weights
should be directly 1047297xed from results obtained with the optimiza-
tion performed using HOMER Then for example for sub-criterion
A1 ldquoElectrical energy capacity shortagerdquo the pair-wise matrix is
not evaluated and the vector weight is directly de1047297ned from
Fig 6 Electric load pro1047297le considered
Table 3
Results obtained with HOMER for the low diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 14 021 569 0301
Option 2 14 698 602 0315
Option 3 14 826 514 0318
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 105
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 711
optimization results with values 115 502 and 482 in kWhyear of
production capacity shortage for options 1 2 and 3 respectively
These values are normalized to the AHP method evaluation scale
(1ndash9) and the weight vector is respectively given by 023 038 and
039 It should be noted that in this case option 3 has the higher
weight because it represents the lower capacity shortage which is
a desired design requirement
The local weights matrix obtained is given in Table 4
The cells 1047297lled with light gray color denote the weights that
were 1047297xed directly from results of the optimization using HOMER
Pair-wise matrices used to obtain this weight matrix are presented
in this paper in Appendix A
The global weight vector is given in Table 5
With these results the global weights for each candidate
option are obtained as
Option 1 03413
Option 2 03247
Option 3 03340
Fig 7 Simulation results (electricity production) of hybrid sources for the three options considered
Table 4
Local weights matrix obtained for low diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 023 014 055 025 025 033 034 034 033 033 033 010 060 033 020 033 033
Op2 038 033 021 059 059 031 033 033 032 034 033 026 020 033 020 033 033
Op3 039 053 024 016 016 036 033 033 035 033 033 064 020 033 060 033 033
Table 5
Global weight vector obtained for low diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110106
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 811
With the results obtained after evaluation of each alternative
option 1 stands as still the best solution according to the evaluated
criteria An important aspect that may have an in1047298uence in this
result is the choice of option 1 as a hybrid energy source with a
high component of continuous energy capacity (the diesel gen-
erator represents the 74 of the energy produced) Using HOMER
a time response simulation is obtained for the selected solution
The simulation result is presented in Fig 8 and is representative of
the operation of the hybrid generator during a complete week The1047297gure shows the levels of energy production for each source the
electrical load demand and the batteries state-of-charge (SOC)
The results show the need of the diesel generator when the
batteries SOC falls beyond 50 during low wind energy
production
32 High diesel price scenario
The second case study scenario is given by the same example as
before but now considering international diesel fuel prices The
price taken as reference is at 161$L (average value in Europe
during July 2012) Using a similar model as in the previous case
the levelized cost optimization using HOMER is performed Results
are shown in Table 6 where the system characteristics arepresented The operating electricity production characteristic of
each con1047297guration is presented in Fig 9 In this case power
generation from solar panels dominates the 1047297rst and second
0 20 40 60 80 100 120 1400
05
1
15
0 20 40 60 80 100 120 14040
60
80
100
Fig 8 Time simulation results hybrid system operating during a week
Table 6
Results obtained with HOMER for the high diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 25 896 383 0590
Option 2 28 408 384 0629
Option 3 29 469 878 0632
Fig 9 Simulation results (electricity production) of hybrid sources for the three options considered in the second case study problem
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 107
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 911
options The third system option is composed by a mixed solar
winddiesel power generator In this second case study option 1 is
composed by 25 kW in solar panels 1 SW AIR X 400 W wind
turbine 1 string of 6 batteries and a 1 kW power inverter Option
2 is a solution with 3 kW in solar panels 1 string of 6 batteries and
a 1 kW power inverter Option 3 is given with 175 kW in solar
panels 3 SW AIR X 400 W wind turbines 500 W diesel generator
1 string of 6 batteries and a 1 kW power inverter All batteries are
model S4KS25P 76 kWh (4 V 1900 Ah)
With a similar analysis as in the previous case the local
weights matrix is given in Table 7 Here light gray cells denote
the weights that were 1047297xed directly from results of the
optimization using HOMER Similarly pair-wise matrices used to
obtain this weight matrix are presented in this paper in Appendix B
The global weight vector is then given in Table 8
Finally global weights of the three options are obtained as the
product of the local weights matrix and the global weight vector
The result is given by
Option 1 02834
Option 2 03056
Option 3 04110
In this case a re-arrangement of 1047297nal solutions is obtained after
completion of the AHP method The 1047297nal problem solution
selection is given by option 3 where as in the previous case a
higher participation of a continuous energy source has an in1047298u-
ence in the 1047297nal result The third option gives the best balance in
terms of net energy as the other two options are primarily
composed by solar panels A 1047297nal critical parameter was the
capacity shortage which turned out to be relatively high for the
1047297rst two options with unmet supplied energy of 304 kWhyearand 173 kWhyear respectively Time response simulation results
for a complete week of operation are presented in Fig 10 A main
characteristic is the high solar power production with the exceed-
ing generated energy used to keep desired levels of battery stack
SOC A good energy balance is obtained with this option with
virtually no capacity shortage during a year round operation
33 Intermediate diesel prices scenarios
Two additional scenarios were considered with intermediate
diesel price values of 05 and 12$L Similar considerations to the
previous cases were used for pair-wise matrix comparisons
For a diesel price of 05$L the results of the HOMER optimiza-
tion are very similar to the low diesel price case All the possiblesolutions in the group of most interesting optimization results
include a 500 W diesel generator This fact re1047298ects that the diesel
prices are still very low and do not have a big in1047298uence in the
project 1047297nal NPC values After applying the described methodol-
ogy the selected optimal system is given by a mix of solar (250 W)
wind (800 W) and diesel generation (500 W)
For a diesel price of 12$L the optimization yields three main
options to be considered
1 The 1047297rst option is given by the installation of PV (2500 W) and
wind (400 W)
2 The second option consists of a mix of PV (1500 W) wind
(1200 W) and diesel (500 W)
3 The third option is given by a purely PV(3000 W) system
Table 7
Local weights matrix obtained for the high diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 018 026 023 030 030 034 035 034 038 047 043 026 043 033 033 033 033
Op2 032 064 012 016 016 036 033 033 038 047 043 064 014 033 053 033 033
Op3 050 010 065 054 054 030 032 033 023 005 014 010 043 033 014 033 033
Table 8Global weight vector obtained for the high diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
0 20 40 60 80 100 120 1400
05
1
15
2
0 20 40 60 80 100 120 14060
70
80
90
100
Fig 10 Time simulation results hybrid system operating during a week for the
second case study problem
02 04 06 08 1 12 14 16200
400
600
800
1000
1200
1400
1600
1800
Fig 11 PV and wind systems sizes vs diesel prices for each scenario
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110108
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1011
The 1047297nal weight vector gives a 02835 weight for the 1047297rst
04093 for the second and 03072 for the third option The selected
solution is the second option the mix with high PV and wind
participation The storage device obtained for all optimal solutions
is a string of 6 batteries (456 kWh) with a 1 kW power inverter
Fig 11 shows the evolution of PV and wind installation sizes for
the selected solution as a function of the diesel prices considered
in the different scenarios presented in this paperIt can be observed that for the different considerations used in
the pair-wise matrix comparisons at all levels obtained results for
higher diesel prices tend to show an overall increased PV penetra-
tion ratio while for the low diesel price scenarios the selection of
wind power is privileged over PV More importantly the results
show the importance of considering social and environmental
aspects This is shown in the case of a high penetration of PV or
wind when the 1047297nal selected solution is not necessarily the result
of the technical and economical optimization
4 Conclusion
In this paper a hierarchical planning methodology for the
integral net energy design of small-scale hybrid renewable energysystems using ARE sources is proposed The approach considers
technicalndasheconomical as well as societalndashenvironmental aspects in
two different decision layers One of the main features of the
methodology is to complement the classical methodology that
considers the optimization and levelized cost analysis by using the
EROEI or net energy concept and the AHP method for decision
making problems to ensure the consideration of complex and
unconventional aspects such as social and environmental para-
meters in the design of the hybrid ARE systems The methodology
is evaluated using an application example in Margarita Island
under several simulation scenarios low intermediate and high
diesel fuel price cases The obtained results show that the aspects
considered in the proposed methodology have an in1047298uence on the
initial results obtained by using the classical approach of levelizedcost optimization
Acknowledgements
The authors would like to acknowledge the 1047297nancial and
technical support of the Venezuelan Planning School (Escuela
Venezolana de Plani1047297cacioacuten)
Appendix A Pair-wise matrices for low diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 13 01396
Op2 3 1 12 03325Op3 3 2 1 05278
A3 Op1 Op2 Op3 Weights
Op1 1 3 2 05499
Op2 13 1 1 02098
Op3 12 1 1 02402
B1 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
B2 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
D3 Op1 Op2 Op3 Weights
Op1 1 13 15 01047
Op2 3 1 13 02583Op3 5 3 1 06370
E1 Op1 Op2 Op3 Weights
Op1 1 3 3 06000
Op2 13 1 1 02000
Op3 13 1 1 02000
E3 Op1 Op2 Op3 Weights
Op1 1 1 13 02000
Op2 1 1 13 02000Op3 3 3 1 06000
Matrices D2 E2 F1 and F2 are 3 3 all-ones matrices
Appendix B Pair-wise matrices for high diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 3 02583Op2 3 1 5 06370
Op3 13 15 1 01047
A3 Op1 Op2 Op3 Weights
Op1 1 2 13 02297
Op2 12 1 15 01220
Op3 3 5 1 06483
B1 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634
Op3 2 3 1 05396
B2 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634Op3 2 3 1 05396
D1 Op1 Op2 Op3 Weights
Op1 1 1 9 04737
Op2 1 1 9 04737
Op3 19 19 1 00526
D2 Op1 Op2 Op3 Weights
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 109
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1111
Op1 1 1 3 04286Op2 1 1 3 04286
Op3 13 13 1 01429
D3 Op1 Op2 Op3 Weights
Op1 1 13 3 02583
Op2 3 1 5 06370
Op3 13 15 1 01047
E1 Op1 Op2 Op3 Weights
Op1 1 3 1 04286
Op2 13 1 13 01429Op3 1 3 1 04286
E3 Op1 Op2 Op3 Weights
Op1 1 12 3 03325
Op2 2 1 3 05278
Op3 13 13 1 01396
Matrices E2 F1 and F2 are 3 3 all-ones matrices
References
[1] International Energy Agency and Organisation for Economic Co-operation andDevelopment Key world energy statistics 2014
[2] Howes T The EUs New Renewable Energy Directive (200928EC) The newclimate policies of the European Union internal legislation and climatediplomacy vol 5 2010 p 117
[3] Mudacumura GM Mebratu D Haque MS Sustainable development policy andadministration Public administration and public policy Boca Raton CRCTaylor and Francis 2006
[4] Gonzaacutelez-Longatt F Gonzaacutelez JS Payoacuten MB Santos JMR Wind-resource atlasof Venezuela based on on-site anemometry observation Renew SustainEnergy Rev 201439898ndash911
[5] Martenson C The crash course the unsustainable future of our economyenergy and environment John Wiley amp Sons Inc Hoboken NJ 2011
[6] Connolly D Lund H Mathiesen BV Leahy M A review of computer tools foranalysing the integration of renewable energy into various energy systemsAppl Energy 201087(4)1059ndash82
[7] Leng GJ Renewable energy technologies project assessment tool RETScreenVarennes Quebec CANMET Energy Diversi1047297cation Research Laboratory 1998
[8] Lilienthal P HOMER micropower optimization model Report National Renew-able Energy Laboratory (NREL) Golden CO 2005
[9] Saaty TL Fundamentals of decision making and priority theory with theanalytic hierarchy processmdashvol VI RWS Publications Pittsburgh PA 1994
[10] Pohekar SD Ramachandran M Application of multi-criteria decision makingto sustainable energy planningmdasha review Renew Sustain Energy Rev 20048(4)365ndash81
[11] Banos R Manzano-Agugliaro F Montoya FG Gil C Alcayde A Goacutemez JOptimization methods applied to renewable and sustainable energy a reviewRenew Sustain Energy Rev 201115(4)1753ndash66
[12] Taha RA Daim T Multi-criteria applications in renewable energy analysis aliterature review London Springer 2013 p 17ndash30
[13] Elkarmi F Mustafa I Increasing the utilization of solar-energy technologies
(set) in Jordanmdashanalytic hierarchy process Energy Policy 199321(9)978ndash82[14] Akash BA Mamlook R Mohsen MS Multi-criteria selection of electric power
plants using analytical hierarchy process Electr Power Syst Res 199952(1)29ndash35
[15] Wu ZX Wei ZH Mitigation assessment results and priorities for Chinasenergy sector Appl Energy 199756(3ndash4)237ndash51
[16] Wang YM Chin KS Fuzzy analytic hierarchy process a logarithmic fuzzypreference programming methodology Int J Approx Reason 201152(4)541ndash53
[17] Nigim K Munier N Green J Pre-feasibility MCDM tools to aid communities inprioritizing local viable renewable energy sources Renew Energy 200429(11)1775ndash91
[18] Daniel J Vishal NV Albert B Selvarsan I Evaluation of the signi 1047297cant renew-able energy resources in India using analytical hierarchy process LondonSpringer 2010 p 13ndash26
[19] Ahmad S Tahar RM Selection of renewable energy sources for sustainabledevelopment of electricity generation system using analytic hierarchy processa case of Malaysia Renew Energy 201463458ndash66
[20] Tasri A Susilawati A Selection among renewable energy alternatives based ona fuzzy analytic hierarchy process in Indonesia Sustain Energy Technol Assess2014734ndash44
[21] Kaya T Kahraman C Multicriteria renewable energy planning using anintegrated fuzzy VIKOR amp AHP methodology the case of Istanbul Energy2010352517ndash27
[22] Zakerinia M Ghaderi F Piltan M Optimal portfolio selection between differentkinds of renewable energy sources 2010
[23] Barin A Canha LN Abaide A Magnago K Wottrich B Multicriteria analysis of the operation of renewable energy sources taking as basis the AHP methodand fuzzy logic concerning distributed generation systems Online J ElectronElectric Eng (OJEEE) 20091(1)52ndash7
[24] Heo E Kim J Boo KJ Analysis of the assessment factors for renewable energydissemination program evaluation using fuzzy AHP Renew Sustain EnergyRev 201014(8)2214ndash20
[25] Akhgari P Kamalan H Monavari M Utilizing AHP method in rating of highlighted environment parameters in most used renewable energies inelectricity production Eur J Sci Res 201160182ndash8
[26] Murphy DJ Hall CAS Year in reviewmdashEROI or energy return on (energy)
invested Ecol Econ Rev 20101185102ndash18[27] Urdaneta AJ Chankong V A multiobjective minimax approach to controller
settings for systems running under disturbances Control-Theory Adv Technol19895(4)391ndash411
[28] Gonzaacutelez-Longatt F Teraacuten R Meacutendez J Hernaacutendez A Guilleacuten F Evaluacioacuten delRecurso Eoacutelico en Venezuela Parte I Congreso Petrolero Energeacutetico ASME-UNEFA Puerto Cabello 2006 [in Spanish]
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110110
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 211
had an important decrease the alternative renewable energy
(ARE) sources are signaled to play an increasingly more important
role in this diversi1047297cation process Renewable energy sources have
grown from 01 of total of primary energy in 1973 (6106 MToe) to
11 of 13 371 MToe in 2013 [1] This implies an increase of 2400
in 40 years according to International Energy Agency [1]
Several reasons can explain the emergence of renewable
energy as demography conventional energy scarcity equilibrium
of planetary resource use Climate change is a fact Traditionallythe role of ARE sources has been referred to as part of the solution
to global warming [2] Also 80 of global energy resources are
used by 20 of the population living in developed countries [3]
ARE sources are an instrument to address the development of
world population instead of using conventional energy resources
Some advantages of ARE source are their participation on
energy diversi1047297cation their modularity and the capability of
decentralized operation They may contribute to equitable geo-
graphical energy distribution in rural countries and of course ARE
source are environmental friendly when compared to some con-
ventional sources as fossil fuel In a regional context ARE sources
are expected to play a more important leading role since wind
energy sources show a signi1047297cant development potential [4] A
very interesting comparison includes the concept of net energy
where the energy ef 1047297ciency of the given sources is computed
namely a focus is given to ldquohow much energy is needed to produce
energyrdquo [5] In this net energy analysis ARE sources are well
placed when compared to hydrocarbon sources
The analysis of the important role that ARE sources may play
includes the following factors
Their participation in the energy diversi1047297cation process Inevitable participation in the national energy approach as
conventional source are becoming scarce (as big scale hydro-
power) oil pro1047297tability should be maximized through exports
or due to immediate unavailability of some resources at big
scale as in the case of natural gas Identi1047297cation of the politicalndashterritorial importance of ARE
installations on remote and isolated sites with special interestin strengthening the nationals developing axes
Active role of ARE source in network stability when considered
as distributed energy sources with migration towards smart
grids networks In1047298uence on the electricity balance and reduction of power
transmission limits with installation of ARE sources near
energy consumption centers
The development of ARE resources are strongly dependent on
energy decision makers preferences These decision makers are
generally energy stakeholders that participate in the energy
market as investors and operators In general ARE projects are
strongly encouraged by national or regional regulatory boards
with strong incentives in order to achieve some economy of scalein alternative generating technologies Within this context when a
energy decision maker prospects the implementation and use of
ARE sources an ef 1047297cient methodology for a technicalndasheconomical
selection is required
There exists in literature several contributions about how to select
an appropriate electrical generating scheme based upon ARE sources
A comprehensive review can be found in [6] with a survey of
methodologies and computational tools to address this problem We
will highlight two of them First the RETscreen platform [7] provides
an option for an initial analysis of technical and 1047297nancial viability of potential renewable energy Another well developed tool is HOMER
[8] which is able to simulate and 1047297nd the best selected option for
energy source con1047297gurations in a given project In this paper it is
considered that the HOMER tool is known by the reader however a
detailed user manual is given in [8]
However despite ongoing ARE evaluation methodologies have
strong strengths on system optimization of technical and eco-
nomic variables It is observed that social and environmental
aspects are hard to treat and valuate in these proposals Quantify-
ing and evaluating the social and environmental viability of a
project is a dif 1047297cult task The analytic hierarchy process (AHP) can
be applied as a complementary evaluation technique in order to
evaluate societal and environmental impact aspects of widespread
integration of ARE sources The methodology proposed and
developed in [9] gives the possibility of classifying the different
analysis criteria into a hierarchical model thus solving the
problem of selection between several solution alternatives The
use of a de1047297ned weighting system makes it easier to classify
alternatives an criteria entries by applying a pair-wise comparison
We can 1047297nd some background on AHP application for ARE
planning purposes Comprehensive reviews on AHP application
for ARE resource planning and project valuation can be found in
[10ndash12] In the following we discuss some of them An analytic
hierarchy process has been applied by [1314] for renewable
energy resources in Jordan The application of an AHP method
for Chinese ARE sources is found in [1516] In [1718] the use of the
AHP method is proposed for decision making and planning of ARE
generation projects in Canada and India respectively Similarly in
[1920] the AHP method is applied for renewable energy planningin Malaysia and Indonesia respectively An integrated VIKOR ndashAHP
methodology to the selection of the best energy policy and
production site is developed in [21] In the work presented in
[22] an integrated methodology is proposed combining linear
optimization and AHP The use of different advanced techniques
along with the AHP method has also been formulated in the
literature In [2324] fuzzy logic is used for the AHP approach for
an ARE application in Brazil and Korea respectively An analysis of
the energy return on energy invested impact of ARE sources to the
environment is presented in [25]
This paper presents an integral approach that considers a
technicalndasheconomical analysis for ARE sources technology selec-
tion but also societalndashenvironmental aspects are considered dur-
ing the analysis of the project considering the net energy or EROEIconcept [26] The use of a two level analytic hierarchy process
Fig 1 Proposed methodology
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 101
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 311
(AHP) is proposed for this purpose The goal is to de1047297ne a
parameter that quanti1047297es the energy ef 1047297ciency of energy sources
The results of the technicalndasheconomical optimization problem
obtained by linear levelized cost minimization are used as an
input to the decision making model based in AHP The methodol-
ogy has been coded using basic programming language and two
study cases are discussed
This paper is divided in three sections In a 1047297rst section the
proposed methodology is presented In the second section themethodology is presented and formulated for a study case Finally
in the last section several results are presented and discussed
2 Proposed methodology
The methodology proposed in this paper is divided in two
levels The 1047297rst part involves the technicalndasheconomic analysis of
the project The second part deals with the socialndashenvironmental
analysis The goal of the methodology is to 1047297nd the best adapted
ARE sources con1047297guration for a given project where both techni-
calndasheconomical and socialndashenvironmental aspect are integrally
considered
The experts of the system may initially de1047297ne the search matrix
for the optimization problem formulation as proposed in [27] Theproposed strategy is also based on the results obtained in a 1047297rst
preliminary analysis using the software in [8] and considering only
technicalndasheconomical aspects A second analysis includes the
socialndashenvironmental aspects using the AHP method As the AHP
is based on weighted classi1047297cation by pair-wise comparison it is
proposed that some of the weighting vectors should be directly
completed from the results obtained in the 1047297rst analysis using [8]
The criteria evaluation of the AHP method may be performed by
the same initial experts of the system The optimization process is
given by exhaustive by time-domain simulations with varying
system components size The different component sizes consid-
ered during simulation may be resumed in a multivariable search
matrix This methodology structure is resumed in Fig 1
21 Analytic hierarchy process (AHP)
The AHP is a method based on the formulation of a hierarchy
structure that is used to deal with complex decisions The method
classi1047297es the choices or solutions to a given problem by evaluation
according to different criteria and a de1047297ned weighting system
Weights are de1047297ned by a pair-wise comparison simplifying the
selection process
In the AHP method problems are modeled as hierarchies This
process is shown in Fig 2 In many cases each criteria may be
divided in several sub-criteria to further analyze the details of the
studied problem
Pair-wise comparisons are used to establish the importance
weight 1047297xed to each criteria These comparisons are based on thescale system proposed by [9] This scale is shown in Table 1
Comparisons using this scale in each level of analysis are used
to build the pair-wise comparison matrices Using the procedure
detailed in [9] the weights associated to each comparison can be
obtained For each comparison matrix two elements of interest are
computed the associated weight vector and the matrix incon-
sistency ratio This last value is used to estimate the consistency of
pair-wise comparisons Consider for example that a problem or a
criterion that needs to be evaluated for three candidate options
Consider for illustration purposes that option 1 is slightly more
important (factor 3 in the pair-wise comparison scale) than option
2 and that option 1 is strongly more important than option 3
(factor 5 in the pair-wise comparison scale) The problem now is tode1047297ne the pair-wise comparison between options 2 and 3 Not a
trivial solution since a system of equations needs to be solved A
graphical solution to this problem is presented in Fig 3 It may be
considered as an approximate solution that the comparison can
take a value of 2 in the proposed scale With these values the pair-
wise comparison matrix for this example is given by
Criterion Op1 Op2 Op3 Weight
Op1 1 3 5 06483
Op2 13 1 2 02297
Op3 15 12 1 01220
with an inconsistency ratio of 00036
In [9] it is considered that an inconsistency ratio lower than10 means that decisions expressed in the pair-wise matrix are
consistent Simple programming language is used to compute
Fig 2 The AHP method
Table 1
Pair-wise comparison table
Numerical rating Verbal judgment of preference between alternatives i and j
1 i is equally important to j
3 i is slightly more important than j
5 i is strongly more important than j
7 i is very strongly more important than j
9 i is extremely more important than j
2468 I ntermediate values
1 15 2 25 3 35
0
001
002
003
004
005
006
Fig 3 Illustrative example for computing inconsistency ratio
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110102
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 411
weights and inconsistency ratios The eigenvectors e of the pair-
wise comparison matrices are given by
e frac14 eigs M λmaxI n
eth1THORN
where M is the pair-wise decision matrix λmax is the maximum
eigenvalue and I n is the identity matrix of size n the size of M The
weight vector is obtained by computing the average values of each
row of the eigenvector
The inconsistency ratio is obtained by 1047297
rst computing theconsistency index
CI frac14 eth λmaxnTHORN
ethn1THORN eth2THORN
Then the inconsistency ratio is given by
CR frac14CI
RI eth3THORN
where RI is the random index given in Table 2
22 The AHP method and the ARE sources
The goal now is to de1047297ne a methodology based on the AHP
method for hybrid power generation projects using ARE sources A
1047297rst classi1047297cation of different criteria is proposed according toimportant variables considered as candidates to have signi1047297cant
weight on the decision making problem
Given the problem complexity and the quantity of variables to
be considered the classi1047297cation is simpli1047297ed to two sets of criteria
technicalndasheconomical aspects and socialndashenvironmental aspects A
similar classi1047297cation is proposed in [17] Each set of criteria is then
divided with the following structure
I ndash Technicalndasheconomical aspects (TecEconAsp)
A Resources availability (AvailRes) A1 Total energy produced (for network connected systems)
or capacity shortage (for stand-alone systems)
A2 Multiple equipment location possibilities
A3 Type of energy constant or intermittent B Technical feasibility (TecFeas)
B1 Technological maturity and local experience with the
technology
B2 Availability of the technology B3 Net energy associated
C Financial viability (FinVia)
C1 Total cost of the project in net present cost (NPC)
C2 Total levelized cost-of-energy (LCOE) in $kWh C3 Total operational and maintenance costs
II ndash Socialndashenvironmental aspects (SocEnvAsp)
D Ecological impact (EcolImp)
D1 Reduction or substitution of CO2 emissions
D2 Reduction of impact on surrounding ecosystem
degradation D3 Visual or sound impact on surrounding populated areas
E Educational potential (EduPot)
E1 The project is tangible and visible within the
community E2 Promotes awareness and public interest on energy
problems
E3 Community participation F Ecological and social impacts (EcoSocImp)
F1 The project promotes economical activity
F2 The project promotes social development
Using this classi1047297cation the hierarchy structure proposed is
given in Fig 4
23 Energy return on energy invested (EROEI)
One of the main contributions of the proposed methodology
consists in the incorporation of the use of the net energy concept
in the pair-wise comparison matrices at levels 2ndash4 of the
described methodology This concept and the energy return on
energy invested (EROEI) for each ARE source can be found in [5]
and [26] respectively
The goal is to include a parameter that quanti1047297es the energy
ef 1047297ciency of energy sources Instead of concentrating analysis
efforts on costs associated to energy production focus is given toldquohow much energy is needed to obtain energyrdquo
The EROEI is given by equation
EROEI frac14UAE =EE eth4THORN
where UAE is the usable acquired energy and EE is the energy
expended
In a 1047297rst level the EROEI variable is taken into account by
considering an important weight to variable B3 in the ldquotechnical
feasibilityrdquo pair-wise matrix On a second level the EROEI will play
an important role when comparing ARE sources technologies by
pair A classi1047297cation of energy sources technologies as a function of
their EROEI ratio is presented in [26] Accordingly wind turbines
have an EROEI of 181 while for solar photovoltaic this value is
around 681 This means that when evaluating the hybrid ARE
solution candidates from the HOMER optimization a higher scaled
weight will be granted to options with a greater nominal power
from wind turbines when contrasted to less favorable EROEI for
solar photovoltaic technologies For comparison purposes an
Table 2
Average random index matrix
Size of M 1 2 3 4 5
Random index 0 0 058 09 112
Size of M 6 7 8 9 10
Random index 124 132 141 145 149
Fig 4 Proposed hierarchy structure
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 103
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 511
example of a renewable energy with high EROEI is hydropower
with typical ratio higher than 1001 In the other hand biodiesel
and bitumen oil from tar sands (as in the Venezuelan Orinoco oil
belt) are among the energy sources with the lower EROEI ranging
from 13 to 41 The impacts of these assumptions are discussed
later in the study case section
The matrices for levels 0 1 and 2 are now presented they are
constructed under the assumptions described before They can
also contain the results of an expert meeting or a community pollevaluating the different pair-wise comparisons When selecting
the pair-wise comparison values in a three dimension matrix at
least one value is specially selected to minimize the inconsistency
ratio from Eq (3)
Goal TecEconAsp SocEnvAsp Weights
TecEconAsp 1 3 075
SocEnvAsp 13 1 025
TecEconAsp A AvailRes BTecFeas CFinVia Weights
A AvailRes 1 3 5 06370
BTecFeas 13 1 3 02583CFinVia 15 13 1 01047
SocEnvAsp DEcolImp EEduPot FEcoSocImp Weights
DEcolImp 1 3 13 02684
EEduPot 13 1 14 01172
FEcoSocImp 3 4 1 06144
A AvailRes A1 A2 A3 Weights
A1 1 7 1 04869
A2 17 1 15 00778
A3 1 5 1 04353
BTecFeas B1 B2 B3 Weights
B1 1 13 19 00704
B2 3 1 15 01782
B3 9 5 1 07514
CFinVia C1 C2 C3 Weights
C1 1 1 3 04286
C2 1 1 3 04286
C3 13 13 1 01429
DEcolImp D1 D2 D3 Weights
D1 1 5 7 07306D2 15 1 3 01884
D3 17 13 1 00810
EEduPot E1 E2 E3 Weights
E1 1 2 13 02493
E2 12 1 13 01571E3 3 3 1 05936
FEcoSocImp F1 F2 Weights
F1 1 15 01667
F2 5 1 08333
In the following sections the practical use of the AHP method is
given with some application examples For an evaluation of the
proposed methodology different study scenarios are presented
3 Study case
To illustrate the proposed planning methodology a study case
is presented in this section The principle is to design a hybrid
power generation system based on ARE sources in Margarita
Island in north-east Venezuela The application is based on an
isolated home without connection to the national electrical grid
For this study case solar wind and fuel resources are con-
sidered as available on the project site The diesel fuel is available
for portable electric power generation Data of wind resource in
Margarita Island are obtained from [28] Data for solar irradiance
was obtained from 10 min measurements in the central Venezue-
lan region (Caracas) These measures are available at httpcbm
usbveclima For illustration purposes of this example the statis-
tical data for solar irradiation in the central Venezuelan region is
used as the solar resource in the island Using an extrapolation to
the island geographical coordinates the HOMER is able to generate
stochastic time series hourly data for both wind and solar
resources Input data of equipment costs and electrical load pro1047297le
were obtained from typical HOMER data library The load pro1047297le
considered is assumed as a typical representation of small house
electrical energy consumption The availability curves of wind and
solar resources and the electrical load pro1047297le considered are
presented in Figs 5 and 6 respectively
Before de1047297ning the possible candidate options for a successful
decision making strategy comparisons associated with levels 0ndash2
(cf Fig 4) may be de1047297ned For this the assumptions considered for
the study are now de1047297ned These assumptions may change for
different project designs and as before in their de1047297nition may be
involved the planning team expert groups or even the organized
community
Some of the assumptions for the study and construction of thepair-wise comparison matrices are as follows
Fig 5 Solar and wind resources considered for Margarita Island
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110104
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 611
Within the technical and economic aspects the availability of
resources is considered as a key parameter with a higher
importance to the technical feasibility and economic viability In the social and environmental aspects it is given a slightly
higher importance to economic and social impact in the
community over the ecological impact and the educational
potential of the project A higher importance is considered for the following para-
meters the amount of energy produced the energy type
(continuous intermittent etc) the concept of net energy and
the importance in promoting social development in the com-
munity where the project is executed
Several scenarios for this study case are now presented for
different diesel prices with low intermediate and high price levels
31 Low diesel price scenario
In this 1047297rst case study the application example presented in
the previous section is considered under the assumption that
diesel fuel is available at current Venezuelan market prices ie
0029$L (price for July 2012)
The design of the hybrid system considers solar panels wind
turbines and diesel generators as energy sources The design
problem is reduced to 1047297nding the optimal size of each system
component For the speci1047297c hybrid generator system problem
search variables are the size (nominal power in W) of the powerconverter the battery stack the solar photovoltaic panels the
wind turbines and the diesel generator A backup battery system is
considered for increased system reliability The power converter
in this case an inverter is necessary to couple DC and AC buses
Additional converters may be considered for DCDC coupling
between ARE sources For example boost or buck converter may
be used to improve system controllability and operating perfor-
mance (improved ef 1047297ciency control to optimal operating point
MPPT maximum power point tracker algorithm etc) Restrictions
of 10 minimum capacity shortage and 25 minimum ARE sources
participation are considered With the given resource availability
data and with several proposed hybrid generation system con1047297g-
urations the HOMER is used for optimization and simulation of
results
With the system topology de1047297ned a search matrix is obtained
and it is possible to run the optimization algorithm using HOMER
to 1047297nd the optimal solution When the optimization process
completed the 1047297rst three solutions are presented in Table 3 They
are classi1047297ed in descending order by their total net presentcost (NPC)
The option 1 consists of 025 kW in solar panels 2 SW AIR X
400 W wind turbines a 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Option 2 is composed by
4 AIR X 400 W wind turbines 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Finally option 3 is given by
075 kW in solar panels 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter All batteries are model
S4KS25P 76 kWh (4 V 1900 Ah)
In Fig 7 the simulation results of the operating behavior of the
obtained system con1047297gurations are presented With these results
it is clear that the diesel generation participation is high This is
normal form of a NPC optimization since fuel prices are very low in
Venezuela These three options are 1047297xed as the candidate solutionsfor the second part of the methodology presented in the previous
section using the AHP method
The second part of the proposed methodology deals with the
evaluation of candidate options using the AHP method For this
each option is weighted by comparison with all the selected sub-
criteria (cf Fig 4) Previously described assumptions are used to
evaluate pair-wise comparisons between options and sub-criteria
Additionally as a reference an analysis of the negative impact of
ARE sources to the environment is presented in [25]
In this paper it is proposed that some evaluation weights
should be directly 1047297xed from results obtained with the optimiza-
tion performed using HOMER Then for example for sub-criterion
A1 ldquoElectrical energy capacity shortagerdquo the pair-wise matrix is
not evaluated and the vector weight is directly de1047297ned from
Fig 6 Electric load pro1047297le considered
Table 3
Results obtained with HOMER for the low diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 14 021 569 0301
Option 2 14 698 602 0315
Option 3 14 826 514 0318
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 105
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 711
optimization results with values 115 502 and 482 in kWhyear of
production capacity shortage for options 1 2 and 3 respectively
These values are normalized to the AHP method evaluation scale
(1ndash9) and the weight vector is respectively given by 023 038 and
039 It should be noted that in this case option 3 has the higher
weight because it represents the lower capacity shortage which is
a desired design requirement
The local weights matrix obtained is given in Table 4
The cells 1047297lled with light gray color denote the weights that
were 1047297xed directly from results of the optimization using HOMER
Pair-wise matrices used to obtain this weight matrix are presented
in this paper in Appendix A
The global weight vector is given in Table 5
With these results the global weights for each candidate
option are obtained as
Option 1 03413
Option 2 03247
Option 3 03340
Fig 7 Simulation results (electricity production) of hybrid sources for the three options considered
Table 4
Local weights matrix obtained for low diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 023 014 055 025 025 033 034 034 033 033 033 010 060 033 020 033 033
Op2 038 033 021 059 059 031 033 033 032 034 033 026 020 033 020 033 033
Op3 039 053 024 016 016 036 033 033 035 033 033 064 020 033 060 033 033
Table 5
Global weight vector obtained for low diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110106
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 811
With the results obtained after evaluation of each alternative
option 1 stands as still the best solution according to the evaluated
criteria An important aspect that may have an in1047298uence in this
result is the choice of option 1 as a hybrid energy source with a
high component of continuous energy capacity (the diesel gen-
erator represents the 74 of the energy produced) Using HOMER
a time response simulation is obtained for the selected solution
The simulation result is presented in Fig 8 and is representative of
the operation of the hybrid generator during a complete week The1047297gure shows the levels of energy production for each source the
electrical load demand and the batteries state-of-charge (SOC)
The results show the need of the diesel generator when the
batteries SOC falls beyond 50 during low wind energy
production
32 High diesel price scenario
The second case study scenario is given by the same example as
before but now considering international diesel fuel prices The
price taken as reference is at 161$L (average value in Europe
during July 2012) Using a similar model as in the previous case
the levelized cost optimization using HOMER is performed Results
are shown in Table 6 where the system characteristics arepresented The operating electricity production characteristic of
each con1047297guration is presented in Fig 9 In this case power
generation from solar panels dominates the 1047297rst and second
0 20 40 60 80 100 120 1400
05
1
15
0 20 40 60 80 100 120 14040
60
80
100
Fig 8 Time simulation results hybrid system operating during a week
Table 6
Results obtained with HOMER for the high diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 25 896 383 0590
Option 2 28 408 384 0629
Option 3 29 469 878 0632
Fig 9 Simulation results (electricity production) of hybrid sources for the three options considered in the second case study problem
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 107
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 911
options The third system option is composed by a mixed solar
winddiesel power generator In this second case study option 1 is
composed by 25 kW in solar panels 1 SW AIR X 400 W wind
turbine 1 string of 6 batteries and a 1 kW power inverter Option
2 is a solution with 3 kW in solar panels 1 string of 6 batteries and
a 1 kW power inverter Option 3 is given with 175 kW in solar
panels 3 SW AIR X 400 W wind turbines 500 W diesel generator
1 string of 6 batteries and a 1 kW power inverter All batteries are
model S4KS25P 76 kWh (4 V 1900 Ah)
With a similar analysis as in the previous case the local
weights matrix is given in Table 7 Here light gray cells denote
the weights that were 1047297xed directly from results of the
optimization using HOMER Similarly pair-wise matrices used to
obtain this weight matrix are presented in this paper in Appendix B
The global weight vector is then given in Table 8
Finally global weights of the three options are obtained as the
product of the local weights matrix and the global weight vector
The result is given by
Option 1 02834
Option 2 03056
Option 3 04110
In this case a re-arrangement of 1047297nal solutions is obtained after
completion of the AHP method The 1047297nal problem solution
selection is given by option 3 where as in the previous case a
higher participation of a continuous energy source has an in1047298u-
ence in the 1047297nal result The third option gives the best balance in
terms of net energy as the other two options are primarily
composed by solar panels A 1047297nal critical parameter was the
capacity shortage which turned out to be relatively high for the
1047297rst two options with unmet supplied energy of 304 kWhyearand 173 kWhyear respectively Time response simulation results
for a complete week of operation are presented in Fig 10 A main
characteristic is the high solar power production with the exceed-
ing generated energy used to keep desired levels of battery stack
SOC A good energy balance is obtained with this option with
virtually no capacity shortage during a year round operation
33 Intermediate diesel prices scenarios
Two additional scenarios were considered with intermediate
diesel price values of 05 and 12$L Similar considerations to the
previous cases were used for pair-wise matrix comparisons
For a diesel price of 05$L the results of the HOMER optimiza-
tion are very similar to the low diesel price case All the possiblesolutions in the group of most interesting optimization results
include a 500 W diesel generator This fact re1047298ects that the diesel
prices are still very low and do not have a big in1047298uence in the
project 1047297nal NPC values After applying the described methodol-
ogy the selected optimal system is given by a mix of solar (250 W)
wind (800 W) and diesel generation (500 W)
For a diesel price of 12$L the optimization yields three main
options to be considered
1 The 1047297rst option is given by the installation of PV (2500 W) and
wind (400 W)
2 The second option consists of a mix of PV (1500 W) wind
(1200 W) and diesel (500 W)
3 The third option is given by a purely PV(3000 W) system
Table 7
Local weights matrix obtained for the high diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 018 026 023 030 030 034 035 034 038 047 043 026 043 033 033 033 033
Op2 032 064 012 016 016 036 033 033 038 047 043 064 014 033 053 033 033
Op3 050 010 065 054 054 030 032 033 023 005 014 010 043 033 014 033 033
Table 8Global weight vector obtained for the high diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
0 20 40 60 80 100 120 1400
05
1
15
2
0 20 40 60 80 100 120 14060
70
80
90
100
Fig 10 Time simulation results hybrid system operating during a week for the
second case study problem
02 04 06 08 1 12 14 16200
400
600
800
1000
1200
1400
1600
1800
Fig 11 PV and wind systems sizes vs diesel prices for each scenario
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110108
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1011
The 1047297nal weight vector gives a 02835 weight for the 1047297rst
04093 for the second and 03072 for the third option The selected
solution is the second option the mix with high PV and wind
participation The storage device obtained for all optimal solutions
is a string of 6 batteries (456 kWh) with a 1 kW power inverter
Fig 11 shows the evolution of PV and wind installation sizes for
the selected solution as a function of the diesel prices considered
in the different scenarios presented in this paperIt can be observed that for the different considerations used in
the pair-wise matrix comparisons at all levels obtained results for
higher diesel prices tend to show an overall increased PV penetra-
tion ratio while for the low diesel price scenarios the selection of
wind power is privileged over PV More importantly the results
show the importance of considering social and environmental
aspects This is shown in the case of a high penetration of PV or
wind when the 1047297nal selected solution is not necessarily the result
of the technical and economical optimization
4 Conclusion
In this paper a hierarchical planning methodology for the
integral net energy design of small-scale hybrid renewable energysystems using ARE sources is proposed The approach considers
technicalndasheconomical as well as societalndashenvironmental aspects in
two different decision layers One of the main features of the
methodology is to complement the classical methodology that
considers the optimization and levelized cost analysis by using the
EROEI or net energy concept and the AHP method for decision
making problems to ensure the consideration of complex and
unconventional aspects such as social and environmental para-
meters in the design of the hybrid ARE systems The methodology
is evaluated using an application example in Margarita Island
under several simulation scenarios low intermediate and high
diesel fuel price cases The obtained results show that the aspects
considered in the proposed methodology have an in1047298uence on the
initial results obtained by using the classical approach of levelizedcost optimization
Acknowledgements
The authors would like to acknowledge the 1047297nancial and
technical support of the Venezuelan Planning School (Escuela
Venezolana de Plani1047297cacioacuten)
Appendix A Pair-wise matrices for low diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 13 01396
Op2 3 1 12 03325Op3 3 2 1 05278
A3 Op1 Op2 Op3 Weights
Op1 1 3 2 05499
Op2 13 1 1 02098
Op3 12 1 1 02402
B1 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
B2 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
D3 Op1 Op2 Op3 Weights
Op1 1 13 15 01047
Op2 3 1 13 02583Op3 5 3 1 06370
E1 Op1 Op2 Op3 Weights
Op1 1 3 3 06000
Op2 13 1 1 02000
Op3 13 1 1 02000
E3 Op1 Op2 Op3 Weights
Op1 1 1 13 02000
Op2 1 1 13 02000Op3 3 3 1 06000
Matrices D2 E2 F1 and F2 are 3 3 all-ones matrices
Appendix B Pair-wise matrices for high diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 3 02583Op2 3 1 5 06370
Op3 13 15 1 01047
A3 Op1 Op2 Op3 Weights
Op1 1 2 13 02297
Op2 12 1 15 01220
Op3 3 5 1 06483
B1 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634
Op3 2 3 1 05396
B2 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634Op3 2 3 1 05396
D1 Op1 Op2 Op3 Weights
Op1 1 1 9 04737
Op2 1 1 9 04737
Op3 19 19 1 00526
D2 Op1 Op2 Op3 Weights
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 109
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1111
Op1 1 1 3 04286Op2 1 1 3 04286
Op3 13 13 1 01429
D3 Op1 Op2 Op3 Weights
Op1 1 13 3 02583
Op2 3 1 5 06370
Op3 13 15 1 01047
E1 Op1 Op2 Op3 Weights
Op1 1 3 1 04286
Op2 13 1 13 01429Op3 1 3 1 04286
E3 Op1 Op2 Op3 Weights
Op1 1 12 3 03325
Op2 2 1 3 05278
Op3 13 13 1 01396
Matrices E2 F1 and F2 are 3 3 all-ones matrices
References
[1] International Energy Agency and Organisation for Economic Co-operation andDevelopment Key world energy statistics 2014
[2] Howes T The EUs New Renewable Energy Directive (200928EC) The newclimate policies of the European Union internal legislation and climatediplomacy vol 5 2010 p 117
[3] Mudacumura GM Mebratu D Haque MS Sustainable development policy andadministration Public administration and public policy Boca Raton CRCTaylor and Francis 2006
[4] Gonzaacutelez-Longatt F Gonzaacutelez JS Payoacuten MB Santos JMR Wind-resource atlasof Venezuela based on on-site anemometry observation Renew SustainEnergy Rev 201439898ndash911
[5] Martenson C The crash course the unsustainable future of our economyenergy and environment John Wiley amp Sons Inc Hoboken NJ 2011
[6] Connolly D Lund H Mathiesen BV Leahy M A review of computer tools foranalysing the integration of renewable energy into various energy systemsAppl Energy 201087(4)1059ndash82
[7] Leng GJ Renewable energy technologies project assessment tool RETScreenVarennes Quebec CANMET Energy Diversi1047297cation Research Laboratory 1998
[8] Lilienthal P HOMER micropower optimization model Report National Renew-able Energy Laboratory (NREL) Golden CO 2005
[9] Saaty TL Fundamentals of decision making and priority theory with theanalytic hierarchy processmdashvol VI RWS Publications Pittsburgh PA 1994
[10] Pohekar SD Ramachandran M Application of multi-criteria decision makingto sustainable energy planningmdasha review Renew Sustain Energy Rev 20048(4)365ndash81
[11] Banos R Manzano-Agugliaro F Montoya FG Gil C Alcayde A Goacutemez JOptimization methods applied to renewable and sustainable energy a reviewRenew Sustain Energy Rev 201115(4)1753ndash66
[12] Taha RA Daim T Multi-criteria applications in renewable energy analysis aliterature review London Springer 2013 p 17ndash30
[13] Elkarmi F Mustafa I Increasing the utilization of solar-energy technologies
(set) in Jordanmdashanalytic hierarchy process Energy Policy 199321(9)978ndash82[14] Akash BA Mamlook R Mohsen MS Multi-criteria selection of electric power
plants using analytical hierarchy process Electr Power Syst Res 199952(1)29ndash35
[15] Wu ZX Wei ZH Mitigation assessment results and priorities for Chinasenergy sector Appl Energy 199756(3ndash4)237ndash51
[16] Wang YM Chin KS Fuzzy analytic hierarchy process a logarithmic fuzzypreference programming methodology Int J Approx Reason 201152(4)541ndash53
[17] Nigim K Munier N Green J Pre-feasibility MCDM tools to aid communities inprioritizing local viable renewable energy sources Renew Energy 200429(11)1775ndash91
[18] Daniel J Vishal NV Albert B Selvarsan I Evaluation of the signi 1047297cant renew-able energy resources in India using analytical hierarchy process LondonSpringer 2010 p 13ndash26
[19] Ahmad S Tahar RM Selection of renewable energy sources for sustainabledevelopment of electricity generation system using analytic hierarchy processa case of Malaysia Renew Energy 201463458ndash66
[20] Tasri A Susilawati A Selection among renewable energy alternatives based ona fuzzy analytic hierarchy process in Indonesia Sustain Energy Technol Assess2014734ndash44
[21] Kaya T Kahraman C Multicriteria renewable energy planning using anintegrated fuzzy VIKOR amp AHP methodology the case of Istanbul Energy2010352517ndash27
[22] Zakerinia M Ghaderi F Piltan M Optimal portfolio selection between differentkinds of renewable energy sources 2010
[23] Barin A Canha LN Abaide A Magnago K Wottrich B Multicriteria analysis of the operation of renewable energy sources taking as basis the AHP methodand fuzzy logic concerning distributed generation systems Online J ElectronElectric Eng (OJEEE) 20091(1)52ndash7
[24] Heo E Kim J Boo KJ Analysis of the assessment factors for renewable energydissemination program evaluation using fuzzy AHP Renew Sustain EnergyRev 201014(8)2214ndash20
[25] Akhgari P Kamalan H Monavari M Utilizing AHP method in rating of highlighted environment parameters in most used renewable energies inelectricity production Eur J Sci Res 201160182ndash8
[26] Murphy DJ Hall CAS Year in reviewmdashEROI or energy return on (energy)
invested Ecol Econ Rev 20101185102ndash18[27] Urdaneta AJ Chankong V A multiobjective minimax approach to controller
settings for systems running under disturbances Control-Theory Adv Technol19895(4)391ndash411
[28] Gonzaacutelez-Longatt F Teraacuten R Meacutendez J Hernaacutendez A Guilleacuten F Evaluacioacuten delRecurso Eoacutelico en Venezuela Parte I Congreso Petrolero Energeacutetico ASME-UNEFA Puerto Cabello 2006 [in Spanish]
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110110
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 311
(AHP) is proposed for this purpose The goal is to de1047297ne a
parameter that quanti1047297es the energy ef 1047297ciency of energy sources
The results of the technicalndasheconomical optimization problem
obtained by linear levelized cost minimization are used as an
input to the decision making model based in AHP The methodol-
ogy has been coded using basic programming language and two
study cases are discussed
This paper is divided in three sections In a 1047297rst section the
proposed methodology is presented In the second section themethodology is presented and formulated for a study case Finally
in the last section several results are presented and discussed
2 Proposed methodology
The methodology proposed in this paper is divided in two
levels The 1047297rst part involves the technicalndasheconomic analysis of
the project The second part deals with the socialndashenvironmental
analysis The goal of the methodology is to 1047297nd the best adapted
ARE sources con1047297guration for a given project where both techni-
calndasheconomical and socialndashenvironmental aspect are integrally
considered
The experts of the system may initially de1047297ne the search matrix
for the optimization problem formulation as proposed in [27] Theproposed strategy is also based on the results obtained in a 1047297rst
preliminary analysis using the software in [8] and considering only
technicalndasheconomical aspects A second analysis includes the
socialndashenvironmental aspects using the AHP method As the AHP
is based on weighted classi1047297cation by pair-wise comparison it is
proposed that some of the weighting vectors should be directly
completed from the results obtained in the 1047297rst analysis using [8]
The criteria evaluation of the AHP method may be performed by
the same initial experts of the system The optimization process is
given by exhaustive by time-domain simulations with varying
system components size The different component sizes consid-
ered during simulation may be resumed in a multivariable search
matrix This methodology structure is resumed in Fig 1
21 Analytic hierarchy process (AHP)
The AHP is a method based on the formulation of a hierarchy
structure that is used to deal with complex decisions The method
classi1047297es the choices or solutions to a given problem by evaluation
according to different criteria and a de1047297ned weighting system
Weights are de1047297ned by a pair-wise comparison simplifying the
selection process
In the AHP method problems are modeled as hierarchies This
process is shown in Fig 2 In many cases each criteria may be
divided in several sub-criteria to further analyze the details of the
studied problem
Pair-wise comparisons are used to establish the importance
weight 1047297xed to each criteria These comparisons are based on thescale system proposed by [9] This scale is shown in Table 1
Comparisons using this scale in each level of analysis are used
to build the pair-wise comparison matrices Using the procedure
detailed in [9] the weights associated to each comparison can be
obtained For each comparison matrix two elements of interest are
computed the associated weight vector and the matrix incon-
sistency ratio This last value is used to estimate the consistency of
pair-wise comparisons Consider for example that a problem or a
criterion that needs to be evaluated for three candidate options
Consider for illustration purposes that option 1 is slightly more
important (factor 3 in the pair-wise comparison scale) than option
2 and that option 1 is strongly more important than option 3
(factor 5 in the pair-wise comparison scale) The problem now is tode1047297ne the pair-wise comparison between options 2 and 3 Not a
trivial solution since a system of equations needs to be solved A
graphical solution to this problem is presented in Fig 3 It may be
considered as an approximate solution that the comparison can
take a value of 2 in the proposed scale With these values the pair-
wise comparison matrix for this example is given by
Criterion Op1 Op2 Op3 Weight
Op1 1 3 5 06483
Op2 13 1 2 02297
Op3 15 12 1 01220
with an inconsistency ratio of 00036
In [9] it is considered that an inconsistency ratio lower than10 means that decisions expressed in the pair-wise matrix are
consistent Simple programming language is used to compute
Fig 2 The AHP method
Table 1
Pair-wise comparison table
Numerical rating Verbal judgment of preference between alternatives i and j
1 i is equally important to j
3 i is slightly more important than j
5 i is strongly more important than j
7 i is very strongly more important than j
9 i is extremely more important than j
2468 I ntermediate values
1 15 2 25 3 35
0
001
002
003
004
005
006
Fig 3 Illustrative example for computing inconsistency ratio
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110102
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 411
weights and inconsistency ratios The eigenvectors e of the pair-
wise comparison matrices are given by
e frac14 eigs M λmaxI n
eth1THORN
where M is the pair-wise decision matrix λmax is the maximum
eigenvalue and I n is the identity matrix of size n the size of M The
weight vector is obtained by computing the average values of each
row of the eigenvector
The inconsistency ratio is obtained by 1047297
rst computing theconsistency index
CI frac14 eth λmaxnTHORN
ethn1THORN eth2THORN
Then the inconsistency ratio is given by
CR frac14CI
RI eth3THORN
where RI is the random index given in Table 2
22 The AHP method and the ARE sources
The goal now is to de1047297ne a methodology based on the AHP
method for hybrid power generation projects using ARE sources A
1047297rst classi1047297cation of different criteria is proposed according toimportant variables considered as candidates to have signi1047297cant
weight on the decision making problem
Given the problem complexity and the quantity of variables to
be considered the classi1047297cation is simpli1047297ed to two sets of criteria
technicalndasheconomical aspects and socialndashenvironmental aspects A
similar classi1047297cation is proposed in [17] Each set of criteria is then
divided with the following structure
I ndash Technicalndasheconomical aspects (TecEconAsp)
A Resources availability (AvailRes) A1 Total energy produced (for network connected systems)
or capacity shortage (for stand-alone systems)
A2 Multiple equipment location possibilities
A3 Type of energy constant or intermittent B Technical feasibility (TecFeas)
B1 Technological maturity and local experience with the
technology
B2 Availability of the technology B3 Net energy associated
C Financial viability (FinVia)
C1 Total cost of the project in net present cost (NPC)
C2 Total levelized cost-of-energy (LCOE) in $kWh C3 Total operational and maintenance costs
II ndash Socialndashenvironmental aspects (SocEnvAsp)
D Ecological impact (EcolImp)
D1 Reduction or substitution of CO2 emissions
D2 Reduction of impact on surrounding ecosystem
degradation D3 Visual or sound impact on surrounding populated areas
E Educational potential (EduPot)
E1 The project is tangible and visible within the
community E2 Promotes awareness and public interest on energy
problems
E3 Community participation F Ecological and social impacts (EcoSocImp)
F1 The project promotes economical activity
F2 The project promotes social development
Using this classi1047297cation the hierarchy structure proposed is
given in Fig 4
23 Energy return on energy invested (EROEI)
One of the main contributions of the proposed methodology
consists in the incorporation of the use of the net energy concept
in the pair-wise comparison matrices at levels 2ndash4 of the
described methodology This concept and the energy return on
energy invested (EROEI) for each ARE source can be found in [5]
and [26] respectively
The goal is to include a parameter that quanti1047297es the energy
ef 1047297ciency of energy sources Instead of concentrating analysis
efforts on costs associated to energy production focus is given toldquohow much energy is needed to obtain energyrdquo
The EROEI is given by equation
EROEI frac14UAE =EE eth4THORN
where UAE is the usable acquired energy and EE is the energy
expended
In a 1047297rst level the EROEI variable is taken into account by
considering an important weight to variable B3 in the ldquotechnical
feasibilityrdquo pair-wise matrix On a second level the EROEI will play
an important role when comparing ARE sources technologies by
pair A classi1047297cation of energy sources technologies as a function of
their EROEI ratio is presented in [26] Accordingly wind turbines
have an EROEI of 181 while for solar photovoltaic this value is
around 681 This means that when evaluating the hybrid ARE
solution candidates from the HOMER optimization a higher scaled
weight will be granted to options with a greater nominal power
from wind turbines when contrasted to less favorable EROEI for
solar photovoltaic technologies For comparison purposes an
Table 2
Average random index matrix
Size of M 1 2 3 4 5
Random index 0 0 058 09 112
Size of M 6 7 8 9 10
Random index 124 132 141 145 149
Fig 4 Proposed hierarchy structure
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 103
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 511
example of a renewable energy with high EROEI is hydropower
with typical ratio higher than 1001 In the other hand biodiesel
and bitumen oil from tar sands (as in the Venezuelan Orinoco oil
belt) are among the energy sources with the lower EROEI ranging
from 13 to 41 The impacts of these assumptions are discussed
later in the study case section
The matrices for levels 0 1 and 2 are now presented they are
constructed under the assumptions described before They can
also contain the results of an expert meeting or a community pollevaluating the different pair-wise comparisons When selecting
the pair-wise comparison values in a three dimension matrix at
least one value is specially selected to minimize the inconsistency
ratio from Eq (3)
Goal TecEconAsp SocEnvAsp Weights
TecEconAsp 1 3 075
SocEnvAsp 13 1 025
TecEconAsp A AvailRes BTecFeas CFinVia Weights
A AvailRes 1 3 5 06370
BTecFeas 13 1 3 02583CFinVia 15 13 1 01047
SocEnvAsp DEcolImp EEduPot FEcoSocImp Weights
DEcolImp 1 3 13 02684
EEduPot 13 1 14 01172
FEcoSocImp 3 4 1 06144
A AvailRes A1 A2 A3 Weights
A1 1 7 1 04869
A2 17 1 15 00778
A3 1 5 1 04353
BTecFeas B1 B2 B3 Weights
B1 1 13 19 00704
B2 3 1 15 01782
B3 9 5 1 07514
CFinVia C1 C2 C3 Weights
C1 1 1 3 04286
C2 1 1 3 04286
C3 13 13 1 01429
DEcolImp D1 D2 D3 Weights
D1 1 5 7 07306D2 15 1 3 01884
D3 17 13 1 00810
EEduPot E1 E2 E3 Weights
E1 1 2 13 02493
E2 12 1 13 01571E3 3 3 1 05936
FEcoSocImp F1 F2 Weights
F1 1 15 01667
F2 5 1 08333
In the following sections the practical use of the AHP method is
given with some application examples For an evaluation of the
proposed methodology different study scenarios are presented
3 Study case
To illustrate the proposed planning methodology a study case
is presented in this section The principle is to design a hybrid
power generation system based on ARE sources in Margarita
Island in north-east Venezuela The application is based on an
isolated home without connection to the national electrical grid
For this study case solar wind and fuel resources are con-
sidered as available on the project site The diesel fuel is available
for portable electric power generation Data of wind resource in
Margarita Island are obtained from [28] Data for solar irradiance
was obtained from 10 min measurements in the central Venezue-
lan region (Caracas) These measures are available at httpcbm
usbveclima For illustration purposes of this example the statis-
tical data for solar irradiation in the central Venezuelan region is
used as the solar resource in the island Using an extrapolation to
the island geographical coordinates the HOMER is able to generate
stochastic time series hourly data for both wind and solar
resources Input data of equipment costs and electrical load pro1047297le
were obtained from typical HOMER data library The load pro1047297le
considered is assumed as a typical representation of small house
electrical energy consumption The availability curves of wind and
solar resources and the electrical load pro1047297le considered are
presented in Figs 5 and 6 respectively
Before de1047297ning the possible candidate options for a successful
decision making strategy comparisons associated with levels 0ndash2
(cf Fig 4) may be de1047297ned For this the assumptions considered for
the study are now de1047297ned These assumptions may change for
different project designs and as before in their de1047297nition may be
involved the planning team expert groups or even the organized
community
Some of the assumptions for the study and construction of thepair-wise comparison matrices are as follows
Fig 5 Solar and wind resources considered for Margarita Island
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110104
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 611
Within the technical and economic aspects the availability of
resources is considered as a key parameter with a higher
importance to the technical feasibility and economic viability In the social and environmental aspects it is given a slightly
higher importance to economic and social impact in the
community over the ecological impact and the educational
potential of the project A higher importance is considered for the following para-
meters the amount of energy produced the energy type
(continuous intermittent etc) the concept of net energy and
the importance in promoting social development in the com-
munity where the project is executed
Several scenarios for this study case are now presented for
different diesel prices with low intermediate and high price levels
31 Low diesel price scenario
In this 1047297rst case study the application example presented in
the previous section is considered under the assumption that
diesel fuel is available at current Venezuelan market prices ie
0029$L (price for July 2012)
The design of the hybrid system considers solar panels wind
turbines and diesel generators as energy sources The design
problem is reduced to 1047297nding the optimal size of each system
component For the speci1047297c hybrid generator system problem
search variables are the size (nominal power in W) of the powerconverter the battery stack the solar photovoltaic panels the
wind turbines and the diesel generator A backup battery system is
considered for increased system reliability The power converter
in this case an inverter is necessary to couple DC and AC buses
Additional converters may be considered for DCDC coupling
between ARE sources For example boost or buck converter may
be used to improve system controllability and operating perfor-
mance (improved ef 1047297ciency control to optimal operating point
MPPT maximum power point tracker algorithm etc) Restrictions
of 10 minimum capacity shortage and 25 minimum ARE sources
participation are considered With the given resource availability
data and with several proposed hybrid generation system con1047297g-
urations the HOMER is used for optimization and simulation of
results
With the system topology de1047297ned a search matrix is obtained
and it is possible to run the optimization algorithm using HOMER
to 1047297nd the optimal solution When the optimization process
completed the 1047297rst three solutions are presented in Table 3 They
are classi1047297ed in descending order by their total net presentcost (NPC)
The option 1 consists of 025 kW in solar panels 2 SW AIR X
400 W wind turbines a 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Option 2 is composed by
4 AIR X 400 W wind turbines 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Finally option 3 is given by
075 kW in solar panels 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter All batteries are model
S4KS25P 76 kWh (4 V 1900 Ah)
In Fig 7 the simulation results of the operating behavior of the
obtained system con1047297gurations are presented With these results
it is clear that the diesel generation participation is high This is
normal form of a NPC optimization since fuel prices are very low in
Venezuela These three options are 1047297xed as the candidate solutionsfor the second part of the methodology presented in the previous
section using the AHP method
The second part of the proposed methodology deals with the
evaluation of candidate options using the AHP method For this
each option is weighted by comparison with all the selected sub-
criteria (cf Fig 4) Previously described assumptions are used to
evaluate pair-wise comparisons between options and sub-criteria
Additionally as a reference an analysis of the negative impact of
ARE sources to the environment is presented in [25]
In this paper it is proposed that some evaluation weights
should be directly 1047297xed from results obtained with the optimiza-
tion performed using HOMER Then for example for sub-criterion
A1 ldquoElectrical energy capacity shortagerdquo the pair-wise matrix is
not evaluated and the vector weight is directly de1047297ned from
Fig 6 Electric load pro1047297le considered
Table 3
Results obtained with HOMER for the low diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 14 021 569 0301
Option 2 14 698 602 0315
Option 3 14 826 514 0318
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 105
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 711
optimization results with values 115 502 and 482 in kWhyear of
production capacity shortage for options 1 2 and 3 respectively
These values are normalized to the AHP method evaluation scale
(1ndash9) and the weight vector is respectively given by 023 038 and
039 It should be noted that in this case option 3 has the higher
weight because it represents the lower capacity shortage which is
a desired design requirement
The local weights matrix obtained is given in Table 4
The cells 1047297lled with light gray color denote the weights that
were 1047297xed directly from results of the optimization using HOMER
Pair-wise matrices used to obtain this weight matrix are presented
in this paper in Appendix A
The global weight vector is given in Table 5
With these results the global weights for each candidate
option are obtained as
Option 1 03413
Option 2 03247
Option 3 03340
Fig 7 Simulation results (electricity production) of hybrid sources for the three options considered
Table 4
Local weights matrix obtained for low diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 023 014 055 025 025 033 034 034 033 033 033 010 060 033 020 033 033
Op2 038 033 021 059 059 031 033 033 032 034 033 026 020 033 020 033 033
Op3 039 053 024 016 016 036 033 033 035 033 033 064 020 033 060 033 033
Table 5
Global weight vector obtained for low diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110106
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 811
With the results obtained after evaluation of each alternative
option 1 stands as still the best solution according to the evaluated
criteria An important aspect that may have an in1047298uence in this
result is the choice of option 1 as a hybrid energy source with a
high component of continuous energy capacity (the diesel gen-
erator represents the 74 of the energy produced) Using HOMER
a time response simulation is obtained for the selected solution
The simulation result is presented in Fig 8 and is representative of
the operation of the hybrid generator during a complete week The1047297gure shows the levels of energy production for each source the
electrical load demand and the batteries state-of-charge (SOC)
The results show the need of the diesel generator when the
batteries SOC falls beyond 50 during low wind energy
production
32 High diesel price scenario
The second case study scenario is given by the same example as
before but now considering international diesel fuel prices The
price taken as reference is at 161$L (average value in Europe
during July 2012) Using a similar model as in the previous case
the levelized cost optimization using HOMER is performed Results
are shown in Table 6 where the system characteristics arepresented The operating electricity production characteristic of
each con1047297guration is presented in Fig 9 In this case power
generation from solar panels dominates the 1047297rst and second
0 20 40 60 80 100 120 1400
05
1
15
0 20 40 60 80 100 120 14040
60
80
100
Fig 8 Time simulation results hybrid system operating during a week
Table 6
Results obtained with HOMER for the high diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 25 896 383 0590
Option 2 28 408 384 0629
Option 3 29 469 878 0632
Fig 9 Simulation results (electricity production) of hybrid sources for the three options considered in the second case study problem
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 107
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 911
options The third system option is composed by a mixed solar
winddiesel power generator In this second case study option 1 is
composed by 25 kW in solar panels 1 SW AIR X 400 W wind
turbine 1 string of 6 batteries and a 1 kW power inverter Option
2 is a solution with 3 kW in solar panels 1 string of 6 batteries and
a 1 kW power inverter Option 3 is given with 175 kW in solar
panels 3 SW AIR X 400 W wind turbines 500 W diesel generator
1 string of 6 batteries and a 1 kW power inverter All batteries are
model S4KS25P 76 kWh (4 V 1900 Ah)
With a similar analysis as in the previous case the local
weights matrix is given in Table 7 Here light gray cells denote
the weights that were 1047297xed directly from results of the
optimization using HOMER Similarly pair-wise matrices used to
obtain this weight matrix are presented in this paper in Appendix B
The global weight vector is then given in Table 8
Finally global weights of the three options are obtained as the
product of the local weights matrix and the global weight vector
The result is given by
Option 1 02834
Option 2 03056
Option 3 04110
In this case a re-arrangement of 1047297nal solutions is obtained after
completion of the AHP method The 1047297nal problem solution
selection is given by option 3 where as in the previous case a
higher participation of a continuous energy source has an in1047298u-
ence in the 1047297nal result The third option gives the best balance in
terms of net energy as the other two options are primarily
composed by solar panels A 1047297nal critical parameter was the
capacity shortage which turned out to be relatively high for the
1047297rst two options with unmet supplied energy of 304 kWhyearand 173 kWhyear respectively Time response simulation results
for a complete week of operation are presented in Fig 10 A main
characteristic is the high solar power production with the exceed-
ing generated energy used to keep desired levels of battery stack
SOC A good energy balance is obtained with this option with
virtually no capacity shortage during a year round operation
33 Intermediate diesel prices scenarios
Two additional scenarios were considered with intermediate
diesel price values of 05 and 12$L Similar considerations to the
previous cases were used for pair-wise matrix comparisons
For a diesel price of 05$L the results of the HOMER optimiza-
tion are very similar to the low diesel price case All the possiblesolutions in the group of most interesting optimization results
include a 500 W diesel generator This fact re1047298ects that the diesel
prices are still very low and do not have a big in1047298uence in the
project 1047297nal NPC values After applying the described methodol-
ogy the selected optimal system is given by a mix of solar (250 W)
wind (800 W) and diesel generation (500 W)
For a diesel price of 12$L the optimization yields three main
options to be considered
1 The 1047297rst option is given by the installation of PV (2500 W) and
wind (400 W)
2 The second option consists of a mix of PV (1500 W) wind
(1200 W) and diesel (500 W)
3 The third option is given by a purely PV(3000 W) system
Table 7
Local weights matrix obtained for the high diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 018 026 023 030 030 034 035 034 038 047 043 026 043 033 033 033 033
Op2 032 064 012 016 016 036 033 033 038 047 043 064 014 033 053 033 033
Op3 050 010 065 054 054 030 032 033 023 005 014 010 043 033 014 033 033
Table 8Global weight vector obtained for the high diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
0 20 40 60 80 100 120 1400
05
1
15
2
0 20 40 60 80 100 120 14060
70
80
90
100
Fig 10 Time simulation results hybrid system operating during a week for the
second case study problem
02 04 06 08 1 12 14 16200
400
600
800
1000
1200
1400
1600
1800
Fig 11 PV and wind systems sizes vs diesel prices for each scenario
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110108
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1011
The 1047297nal weight vector gives a 02835 weight for the 1047297rst
04093 for the second and 03072 for the third option The selected
solution is the second option the mix with high PV and wind
participation The storage device obtained for all optimal solutions
is a string of 6 batteries (456 kWh) with a 1 kW power inverter
Fig 11 shows the evolution of PV and wind installation sizes for
the selected solution as a function of the diesel prices considered
in the different scenarios presented in this paperIt can be observed that for the different considerations used in
the pair-wise matrix comparisons at all levels obtained results for
higher diesel prices tend to show an overall increased PV penetra-
tion ratio while for the low diesel price scenarios the selection of
wind power is privileged over PV More importantly the results
show the importance of considering social and environmental
aspects This is shown in the case of a high penetration of PV or
wind when the 1047297nal selected solution is not necessarily the result
of the technical and economical optimization
4 Conclusion
In this paper a hierarchical planning methodology for the
integral net energy design of small-scale hybrid renewable energysystems using ARE sources is proposed The approach considers
technicalndasheconomical as well as societalndashenvironmental aspects in
two different decision layers One of the main features of the
methodology is to complement the classical methodology that
considers the optimization and levelized cost analysis by using the
EROEI or net energy concept and the AHP method for decision
making problems to ensure the consideration of complex and
unconventional aspects such as social and environmental para-
meters in the design of the hybrid ARE systems The methodology
is evaluated using an application example in Margarita Island
under several simulation scenarios low intermediate and high
diesel fuel price cases The obtained results show that the aspects
considered in the proposed methodology have an in1047298uence on the
initial results obtained by using the classical approach of levelizedcost optimization
Acknowledgements
The authors would like to acknowledge the 1047297nancial and
technical support of the Venezuelan Planning School (Escuela
Venezolana de Plani1047297cacioacuten)
Appendix A Pair-wise matrices for low diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 13 01396
Op2 3 1 12 03325Op3 3 2 1 05278
A3 Op1 Op2 Op3 Weights
Op1 1 3 2 05499
Op2 13 1 1 02098
Op3 12 1 1 02402
B1 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
B2 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
D3 Op1 Op2 Op3 Weights
Op1 1 13 15 01047
Op2 3 1 13 02583Op3 5 3 1 06370
E1 Op1 Op2 Op3 Weights
Op1 1 3 3 06000
Op2 13 1 1 02000
Op3 13 1 1 02000
E3 Op1 Op2 Op3 Weights
Op1 1 1 13 02000
Op2 1 1 13 02000Op3 3 3 1 06000
Matrices D2 E2 F1 and F2 are 3 3 all-ones matrices
Appendix B Pair-wise matrices for high diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 3 02583Op2 3 1 5 06370
Op3 13 15 1 01047
A3 Op1 Op2 Op3 Weights
Op1 1 2 13 02297
Op2 12 1 15 01220
Op3 3 5 1 06483
B1 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634
Op3 2 3 1 05396
B2 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634Op3 2 3 1 05396
D1 Op1 Op2 Op3 Weights
Op1 1 1 9 04737
Op2 1 1 9 04737
Op3 19 19 1 00526
D2 Op1 Op2 Op3 Weights
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 109
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1111
Op1 1 1 3 04286Op2 1 1 3 04286
Op3 13 13 1 01429
D3 Op1 Op2 Op3 Weights
Op1 1 13 3 02583
Op2 3 1 5 06370
Op3 13 15 1 01047
E1 Op1 Op2 Op3 Weights
Op1 1 3 1 04286
Op2 13 1 13 01429Op3 1 3 1 04286
E3 Op1 Op2 Op3 Weights
Op1 1 12 3 03325
Op2 2 1 3 05278
Op3 13 13 1 01396
Matrices E2 F1 and F2 are 3 3 all-ones matrices
References
[1] International Energy Agency and Organisation for Economic Co-operation andDevelopment Key world energy statistics 2014
[2] Howes T The EUs New Renewable Energy Directive (200928EC) The newclimate policies of the European Union internal legislation and climatediplomacy vol 5 2010 p 117
[3] Mudacumura GM Mebratu D Haque MS Sustainable development policy andadministration Public administration and public policy Boca Raton CRCTaylor and Francis 2006
[4] Gonzaacutelez-Longatt F Gonzaacutelez JS Payoacuten MB Santos JMR Wind-resource atlasof Venezuela based on on-site anemometry observation Renew SustainEnergy Rev 201439898ndash911
[5] Martenson C The crash course the unsustainable future of our economyenergy and environment John Wiley amp Sons Inc Hoboken NJ 2011
[6] Connolly D Lund H Mathiesen BV Leahy M A review of computer tools foranalysing the integration of renewable energy into various energy systemsAppl Energy 201087(4)1059ndash82
[7] Leng GJ Renewable energy technologies project assessment tool RETScreenVarennes Quebec CANMET Energy Diversi1047297cation Research Laboratory 1998
[8] Lilienthal P HOMER micropower optimization model Report National Renew-able Energy Laboratory (NREL) Golden CO 2005
[9] Saaty TL Fundamentals of decision making and priority theory with theanalytic hierarchy processmdashvol VI RWS Publications Pittsburgh PA 1994
[10] Pohekar SD Ramachandran M Application of multi-criteria decision makingto sustainable energy planningmdasha review Renew Sustain Energy Rev 20048(4)365ndash81
[11] Banos R Manzano-Agugliaro F Montoya FG Gil C Alcayde A Goacutemez JOptimization methods applied to renewable and sustainable energy a reviewRenew Sustain Energy Rev 201115(4)1753ndash66
[12] Taha RA Daim T Multi-criteria applications in renewable energy analysis aliterature review London Springer 2013 p 17ndash30
[13] Elkarmi F Mustafa I Increasing the utilization of solar-energy technologies
(set) in Jordanmdashanalytic hierarchy process Energy Policy 199321(9)978ndash82[14] Akash BA Mamlook R Mohsen MS Multi-criteria selection of electric power
plants using analytical hierarchy process Electr Power Syst Res 199952(1)29ndash35
[15] Wu ZX Wei ZH Mitigation assessment results and priorities for Chinasenergy sector Appl Energy 199756(3ndash4)237ndash51
[16] Wang YM Chin KS Fuzzy analytic hierarchy process a logarithmic fuzzypreference programming methodology Int J Approx Reason 201152(4)541ndash53
[17] Nigim K Munier N Green J Pre-feasibility MCDM tools to aid communities inprioritizing local viable renewable energy sources Renew Energy 200429(11)1775ndash91
[18] Daniel J Vishal NV Albert B Selvarsan I Evaluation of the signi 1047297cant renew-able energy resources in India using analytical hierarchy process LondonSpringer 2010 p 13ndash26
[19] Ahmad S Tahar RM Selection of renewable energy sources for sustainabledevelopment of electricity generation system using analytic hierarchy processa case of Malaysia Renew Energy 201463458ndash66
[20] Tasri A Susilawati A Selection among renewable energy alternatives based ona fuzzy analytic hierarchy process in Indonesia Sustain Energy Technol Assess2014734ndash44
[21] Kaya T Kahraman C Multicriteria renewable energy planning using anintegrated fuzzy VIKOR amp AHP methodology the case of Istanbul Energy2010352517ndash27
[22] Zakerinia M Ghaderi F Piltan M Optimal portfolio selection between differentkinds of renewable energy sources 2010
[23] Barin A Canha LN Abaide A Magnago K Wottrich B Multicriteria analysis of the operation of renewable energy sources taking as basis the AHP methodand fuzzy logic concerning distributed generation systems Online J ElectronElectric Eng (OJEEE) 20091(1)52ndash7
[24] Heo E Kim J Boo KJ Analysis of the assessment factors for renewable energydissemination program evaluation using fuzzy AHP Renew Sustain EnergyRev 201014(8)2214ndash20
[25] Akhgari P Kamalan H Monavari M Utilizing AHP method in rating of highlighted environment parameters in most used renewable energies inelectricity production Eur J Sci Res 201160182ndash8
[26] Murphy DJ Hall CAS Year in reviewmdashEROI or energy return on (energy)
invested Ecol Econ Rev 20101185102ndash18[27] Urdaneta AJ Chankong V A multiobjective minimax approach to controller
settings for systems running under disturbances Control-Theory Adv Technol19895(4)391ndash411
[28] Gonzaacutelez-Longatt F Teraacuten R Meacutendez J Hernaacutendez A Guilleacuten F Evaluacioacuten delRecurso Eoacutelico en Venezuela Parte I Congreso Petrolero Energeacutetico ASME-UNEFA Puerto Cabello 2006 [in Spanish]
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110110
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 411
weights and inconsistency ratios The eigenvectors e of the pair-
wise comparison matrices are given by
e frac14 eigs M λmaxI n
eth1THORN
where M is the pair-wise decision matrix λmax is the maximum
eigenvalue and I n is the identity matrix of size n the size of M The
weight vector is obtained by computing the average values of each
row of the eigenvector
The inconsistency ratio is obtained by 1047297
rst computing theconsistency index
CI frac14 eth λmaxnTHORN
ethn1THORN eth2THORN
Then the inconsistency ratio is given by
CR frac14CI
RI eth3THORN
where RI is the random index given in Table 2
22 The AHP method and the ARE sources
The goal now is to de1047297ne a methodology based on the AHP
method for hybrid power generation projects using ARE sources A
1047297rst classi1047297cation of different criteria is proposed according toimportant variables considered as candidates to have signi1047297cant
weight on the decision making problem
Given the problem complexity and the quantity of variables to
be considered the classi1047297cation is simpli1047297ed to two sets of criteria
technicalndasheconomical aspects and socialndashenvironmental aspects A
similar classi1047297cation is proposed in [17] Each set of criteria is then
divided with the following structure
I ndash Technicalndasheconomical aspects (TecEconAsp)
A Resources availability (AvailRes) A1 Total energy produced (for network connected systems)
or capacity shortage (for stand-alone systems)
A2 Multiple equipment location possibilities
A3 Type of energy constant or intermittent B Technical feasibility (TecFeas)
B1 Technological maturity and local experience with the
technology
B2 Availability of the technology B3 Net energy associated
C Financial viability (FinVia)
C1 Total cost of the project in net present cost (NPC)
C2 Total levelized cost-of-energy (LCOE) in $kWh C3 Total operational and maintenance costs
II ndash Socialndashenvironmental aspects (SocEnvAsp)
D Ecological impact (EcolImp)
D1 Reduction or substitution of CO2 emissions
D2 Reduction of impact on surrounding ecosystem
degradation D3 Visual or sound impact on surrounding populated areas
E Educational potential (EduPot)
E1 The project is tangible and visible within the
community E2 Promotes awareness and public interest on energy
problems
E3 Community participation F Ecological and social impacts (EcoSocImp)
F1 The project promotes economical activity
F2 The project promotes social development
Using this classi1047297cation the hierarchy structure proposed is
given in Fig 4
23 Energy return on energy invested (EROEI)
One of the main contributions of the proposed methodology
consists in the incorporation of the use of the net energy concept
in the pair-wise comparison matrices at levels 2ndash4 of the
described methodology This concept and the energy return on
energy invested (EROEI) for each ARE source can be found in [5]
and [26] respectively
The goal is to include a parameter that quanti1047297es the energy
ef 1047297ciency of energy sources Instead of concentrating analysis
efforts on costs associated to energy production focus is given toldquohow much energy is needed to obtain energyrdquo
The EROEI is given by equation
EROEI frac14UAE =EE eth4THORN
where UAE is the usable acquired energy and EE is the energy
expended
In a 1047297rst level the EROEI variable is taken into account by
considering an important weight to variable B3 in the ldquotechnical
feasibilityrdquo pair-wise matrix On a second level the EROEI will play
an important role when comparing ARE sources technologies by
pair A classi1047297cation of energy sources technologies as a function of
their EROEI ratio is presented in [26] Accordingly wind turbines
have an EROEI of 181 while for solar photovoltaic this value is
around 681 This means that when evaluating the hybrid ARE
solution candidates from the HOMER optimization a higher scaled
weight will be granted to options with a greater nominal power
from wind turbines when contrasted to less favorable EROEI for
solar photovoltaic technologies For comparison purposes an
Table 2
Average random index matrix
Size of M 1 2 3 4 5
Random index 0 0 058 09 112
Size of M 6 7 8 9 10
Random index 124 132 141 145 149
Fig 4 Proposed hierarchy structure
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 103
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 511
example of a renewable energy with high EROEI is hydropower
with typical ratio higher than 1001 In the other hand biodiesel
and bitumen oil from tar sands (as in the Venezuelan Orinoco oil
belt) are among the energy sources with the lower EROEI ranging
from 13 to 41 The impacts of these assumptions are discussed
later in the study case section
The matrices for levels 0 1 and 2 are now presented they are
constructed under the assumptions described before They can
also contain the results of an expert meeting or a community pollevaluating the different pair-wise comparisons When selecting
the pair-wise comparison values in a three dimension matrix at
least one value is specially selected to minimize the inconsistency
ratio from Eq (3)
Goal TecEconAsp SocEnvAsp Weights
TecEconAsp 1 3 075
SocEnvAsp 13 1 025
TecEconAsp A AvailRes BTecFeas CFinVia Weights
A AvailRes 1 3 5 06370
BTecFeas 13 1 3 02583CFinVia 15 13 1 01047
SocEnvAsp DEcolImp EEduPot FEcoSocImp Weights
DEcolImp 1 3 13 02684
EEduPot 13 1 14 01172
FEcoSocImp 3 4 1 06144
A AvailRes A1 A2 A3 Weights
A1 1 7 1 04869
A2 17 1 15 00778
A3 1 5 1 04353
BTecFeas B1 B2 B3 Weights
B1 1 13 19 00704
B2 3 1 15 01782
B3 9 5 1 07514
CFinVia C1 C2 C3 Weights
C1 1 1 3 04286
C2 1 1 3 04286
C3 13 13 1 01429
DEcolImp D1 D2 D3 Weights
D1 1 5 7 07306D2 15 1 3 01884
D3 17 13 1 00810
EEduPot E1 E2 E3 Weights
E1 1 2 13 02493
E2 12 1 13 01571E3 3 3 1 05936
FEcoSocImp F1 F2 Weights
F1 1 15 01667
F2 5 1 08333
In the following sections the practical use of the AHP method is
given with some application examples For an evaluation of the
proposed methodology different study scenarios are presented
3 Study case
To illustrate the proposed planning methodology a study case
is presented in this section The principle is to design a hybrid
power generation system based on ARE sources in Margarita
Island in north-east Venezuela The application is based on an
isolated home without connection to the national electrical grid
For this study case solar wind and fuel resources are con-
sidered as available on the project site The diesel fuel is available
for portable electric power generation Data of wind resource in
Margarita Island are obtained from [28] Data for solar irradiance
was obtained from 10 min measurements in the central Venezue-
lan region (Caracas) These measures are available at httpcbm
usbveclima For illustration purposes of this example the statis-
tical data for solar irradiation in the central Venezuelan region is
used as the solar resource in the island Using an extrapolation to
the island geographical coordinates the HOMER is able to generate
stochastic time series hourly data for both wind and solar
resources Input data of equipment costs and electrical load pro1047297le
were obtained from typical HOMER data library The load pro1047297le
considered is assumed as a typical representation of small house
electrical energy consumption The availability curves of wind and
solar resources and the electrical load pro1047297le considered are
presented in Figs 5 and 6 respectively
Before de1047297ning the possible candidate options for a successful
decision making strategy comparisons associated with levels 0ndash2
(cf Fig 4) may be de1047297ned For this the assumptions considered for
the study are now de1047297ned These assumptions may change for
different project designs and as before in their de1047297nition may be
involved the planning team expert groups or even the organized
community
Some of the assumptions for the study and construction of thepair-wise comparison matrices are as follows
Fig 5 Solar and wind resources considered for Margarita Island
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110104
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 611
Within the technical and economic aspects the availability of
resources is considered as a key parameter with a higher
importance to the technical feasibility and economic viability In the social and environmental aspects it is given a slightly
higher importance to economic and social impact in the
community over the ecological impact and the educational
potential of the project A higher importance is considered for the following para-
meters the amount of energy produced the energy type
(continuous intermittent etc) the concept of net energy and
the importance in promoting social development in the com-
munity where the project is executed
Several scenarios for this study case are now presented for
different diesel prices with low intermediate and high price levels
31 Low diesel price scenario
In this 1047297rst case study the application example presented in
the previous section is considered under the assumption that
diesel fuel is available at current Venezuelan market prices ie
0029$L (price for July 2012)
The design of the hybrid system considers solar panels wind
turbines and diesel generators as energy sources The design
problem is reduced to 1047297nding the optimal size of each system
component For the speci1047297c hybrid generator system problem
search variables are the size (nominal power in W) of the powerconverter the battery stack the solar photovoltaic panels the
wind turbines and the diesel generator A backup battery system is
considered for increased system reliability The power converter
in this case an inverter is necessary to couple DC and AC buses
Additional converters may be considered for DCDC coupling
between ARE sources For example boost or buck converter may
be used to improve system controllability and operating perfor-
mance (improved ef 1047297ciency control to optimal operating point
MPPT maximum power point tracker algorithm etc) Restrictions
of 10 minimum capacity shortage and 25 minimum ARE sources
participation are considered With the given resource availability
data and with several proposed hybrid generation system con1047297g-
urations the HOMER is used for optimization and simulation of
results
With the system topology de1047297ned a search matrix is obtained
and it is possible to run the optimization algorithm using HOMER
to 1047297nd the optimal solution When the optimization process
completed the 1047297rst three solutions are presented in Table 3 They
are classi1047297ed in descending order by their total net presentcost (NPC)
The option 1 consists of 025 kW in solar panels 2 SW AIR X
400 W wind turbines a 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Option 2 is composed by
4 AIR X 400 W wind turbines 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Finally option 3 is given by
075 kW in solar panels 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter All batteries are model
S4KS25P 76 kWh (4 V 1900 Ah)
In Fig 7 the simulation results of the operating behavior of the
obtained system con1047297gurations are presented With these results
it is clear that the diesel generation participation is high This is
normal form of a NPC optimization since fuel prices are very low in
Venezuela These three options are 1047297xed as the candidate solutionsfor the second part of the methodology presented in the previous
section using the AHP method
The second part of the proposed methodology deals with the
evaluation of candidate options using the AHP method For this
each option is weighted by comparison with all the selected sub-
criteria (cf Fig 4) Previously described assumptions are used to
evaluate pair-wise comparisons between options and sub-criteria
Additionally as a reference an analysis of the negative impact of
ARE sources to the environment is presented in [25]
In this paper it is proposed that some evaluation weights
should be directly 1047297xed from results obtained with the optimiza-
tion performed using HOMER Then for example for sub-criterion
A1 ldquoElectrical energy capacity shortagerdquo the pair-wise matrix is
not evaluated and the vector weight is directly de1047297ned from
Fig 6 Electric load pro1047297le considered
Table 3
Results obtained with HOMER for the low diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 14 021 569 0301
Option 2 14 698 602 0315
Option 3 14 826 514 0318
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 105
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 711
optimization results with values 115 502 and 482 in kWhyear of
production capacity shortage for options 1 2 and 3 respectively
These values are normalized to the AHP method evaluation scale
(1ndash9) and the weight vector is respectively given by 023 038 and
039 It should be noted that in this case option 3 has the higher
weight because it represents the lower capacity shortage which is
a desired design requirement
The local weights matrix obtained is given in Table 4
The cells 1047297lled with light gray color denote the weights that
were 1047297xed directly from results of the optimization using HOMER
Pair-wise matrices used to obtain this weight matrix are presented
in this paper in Appendix A
The global weight vector is given in Table 5
With these results the global weights for each candidate
option are obtained as
Option 1 03413
Option 2 03247
Option 3 03340
Fig 7 Simulation results (electricity production) of hybrid sources for the three options considered
Table 4
Local weights matrix obtained for low diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 023 014 055 025 025 033 034 034 033 033 033 010 060 033 020 033 033
Op2 038 033 021 059 059 031 033 033 032 034 033 026 020 033 020 033 033
Op3 039 053 024 016 016 036 033 033 035 033 033 064 020 033 060 033 033
Table 5
Global weight vector obtained for low diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110106
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 811
With the results obtained after evaluation of each alternative
option 1 stands as still the best solution according to the evaluated
criteria An important aspect that may have an in1047298uence in this
result is the choice of option 1 as a hybrid energy source with a
high component of continuous energy capacity (the diesel gen-
erator represents the 74 of the energy produced) Using HOMER
a time response simulation is obtained for the selected solution
The simulation result is presented in Fig 8 and is representative of
the operation of the hybrid generator during a complete week The1047297gure shows the levels of energy production for each source the
electrical load demand and the batteries state-of-charge (SOC)
The results show the need of the diesel generator when the
batteries SOC falls beyond 50 during low wind energy
production
32 High diesel price scenario
The second case study scenario is given by the same example as
before but now considering international diesel fuel prices The
price taken as reference is at 161$L (average value in Europe
during July 2012) Using a similar model as in the previous case
the levelized cost optimization using HOMER is performed Results
are shown in Table 6 where the system characteristics arepresented The operating electricity production characteristic of
each con1047297guration is presented in Fig 9 In this case power
generation from solar panels dominates the 1047297rst and second
0 20 40 60 80 100 120 1400
05
1
15
0 20 40 60 80 100 120 14040
60
80
100
Fig 8 Time simulation results hybrid system operating during a week
Table 6
Results obtained with HOMER for the high diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 25 896 383 0590
Option 2 28 408 384 0629
Option 3 29 469 878 0632
Fig 9 Simulation results (electricity production) of hybrid sources for the three options considered in the second case study problem
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 107
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 911
options The third system option is composed by a mixed solar
winddiesel power generator In this second case study option 1 is
composed by 25 kW in solar panels 1 SW AIR X 400 W wind
turbine 1 string of 6 batteries and a 1 kW power inverter Option
2 is a solution with 3 kW in solar panels 1 string of 6 batteries and
a 1 kW power inverter Option 3 is given with 175 kW in solar
panels 3 SW AIR X 400 W wind turbines 500 W diesel generator
1 string of 6 batteries and a 1 kW power inverter All batteries are
model S4KS25P 76 kWh (4 V 1900 Ah)
With a similar analysis as in the previous case the local
weights matrix is given in Table 7 Here light gray cells denote
the weights that were 1047297xed directly from results of the
optimization using HOMER Similarly pair-wise matrices used to
obtain this weight matrix are presented in this paper in Appendix B
The global weight vector is then given in Table 8
Finally global weights of the three options are obtained as the
product of the local weights matrix and the global weight vector
The result is given by
Option 1 02834
Option 2 03056
Option 3 04110
In this case a re-arrangement of 1047297nal solutions is obtained after
completion of the AHP method The 1047297nal problem solution
selection is given by option 3 where as in the previous case a
higher participation of a continuous energy source has an in1047298u-
ence in the 1047297nal result The third option gives the best balance in
terms of net energy as the other two options are primarily
composed by solar panels A 1047297nal critical parameter was the
capacity shortage which turned out to be relatively high for the
1047297rst two options with unmet supplied energy of 304 kWhyearand 173 kWhyear respectively Time response simulation results
for a complete week of operation are presented in Fig 10 A main
characteristic is the high solar power production with the exceed-
ing generated energy used to keep desired levels of battery stack
SOC A good energy balance is obtained with this option with
virtually no capacity shortage during a year round operation
33 Intermediate diesel prices scenarios
Two additional scenarios were considered with intermediate
diesel price values of 05 and 12$L Similar considerations to the
previous cases were used for pair-wise matrix comparisons
For a diesel price of 05$L the results of the HOMER optimiza-
tion are very similar to the low diesel price case All the possiblesolutions in the group of most interesting optimization results
include a 500 W diesel generator This fact re1047298ects that the diesel
prices are still very low and do not have a big in1047298uence in the
project 1047297nal NPC values After applying the described methodol-
ogy the selected optimal system is given by a mix of solar (250 W)
wind (800 W) and diesel generation (500 W)
For a diesel price of 12$L the optimization yields three main
options to be considered
1 The 1047297rst option is given by the installation of PV (2500 W) and
wind (400 W)
2 The second option consists of a mix of PV (1500 W) wind
(1200 W) and diesel (500 W)
3 The third option is given by a purely PV(3000 W) system
Table 7
Local weights matrix obtained for the high diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 018 026 023 030 030 034 035 034 038 047 043 026 043 033 033 033 033
Op2 032 064 012 016 016 036 033 033 038 047 043 064 014 033 053 033 033
Op3 050 010 065 054 054 030 032 033 023 005 014 010 043 033 014 033 033
Table 8Global weight vector obtained for the high diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
0 20 40 60 80 100 120 1400
05
1
15
2
0 20 40 60 80 100 120 14060
70
80
90
100
Fig 10 Time simulation results hybrid system operating during a week for the
second case study problem
02 04 06 08 1 12 14 16200
400
600
800
1000
1200
1400
1600
1800
Fig 11 PV and wind systems sizes vs diesel prices for each scenario
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110108
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1011
The 1047297nal weight vector gives a 02835 weight for the 1047297rst
04093 for the second and 03072 for the third option The selected
solution is the second option the mix with high PV and wind
participation The storage device obtained for all optimal solutions
is a string of 6 batteries (456 kWh) with a 1 kW power inverter
Fig 11 shows the evolution of PV and wind installation sizes for
the selected solution as a function of the diesel prices considered
in the different scenarios presented in this paperIt can be observed that for the different considerations used in
the pair-wise matrix comparisons at all levels obtained results for
higher diesel prices tend to show an overall increased PV penetra-
tion ratio while for the low diesel price scenarios the selection of
wind power is privileged over PV More importantly the results
show the importance of considering social and environmental
aspects This is shown in the case of a high penetration of PV or
wind when the 1047297nal selected solution is not necessarily the result
of the technical and economical optimization
4 Conclusion
In this paper a hierarchical planning methodology for the
integral net energy design of small-scale hybrid renewable energysystems using ARE sources is proposed The approach considers
technicalndasheconomical as well as societalndashenvironmental aspects in
two different decision layers One of the main features of the
methodology is to complement the classical methodology that
considers the optimization and levelized cost analysis by using the
EROEI or net energy concept and the AHP method for decision
making problems to ensure the consideration of complex and
unconventional aspects such as social and environmental para-
meters in the design of the hybrid ARE systems The methodology
is evaluated using an application example in Margarita Island
under several simulation scenarios low intermediate and high
diesel fuel price cases The obtained results show that the aspects
considered in the proposed methodology have an in1047298uence on the
initial results obtained by using the classical approach of levelizedcost optimization
Acknowledgements
The authors would like to acknowledge the 1047297nancial and
technical support of the Venezuelan Planning School (Escuela
Venezolana de Plani1047297cacioacuten)
Appendix A Pair-wise matrices for low diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 13 01396
Op2 3 1 12 03325Op3 3 2 1 05278
A3 Op1 Op2 Op3 Weights
Op1 1 3 2 05499
Op2 13 1 1 02098
Op3 12 1 1 02402
B1 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
B2 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
D3 Op1 Op2 Op3 Weights
Op1 1 13 15 01047
Op2 3 1 13 02583Op3 5 3 1 06370
E1 Op1 Op2 Op3 Weights
Op1 1 3 3 06000
Op2 13 1 1 02000
Op3 13 1 1 02000
E3 Op1 Op2 Op3 Weights
Op1 1 1 13 02000
Op2 1 1 13 02000Op3 3 3 1 06000
Matrices D2 E2 F1 and F2 are 3 3 all-ones matrices
Appendix B Pair-wise matrices for high diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 3 02583Op2 3 1 5 06370
Op3 13 15 1 01047
A3 Op1 Op2 Op3 Weights
Op1 1 2 13 02297
Op2 12 1 15 01220
Op3 3 5 1 06483
B1 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634
Op3 2 3 1 05396
B2 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634Op3 2 3 1 05396
D1 Op1 Op2 Op3 Weights
Op1 1 1 9 04737
Op2 1 1 9 04737
Op3 19 19 1 00526
D2 Op1 Op2 Op3 Weights
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 109
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1111
Op1 1 1 3 04286Op2 1 1 3 04286
Op3 13 13 1 01429
D3 Op1 Op2 Op3 Weights
Op1 1 13 3 02583
Op2 3 1 5 06370
Op3 13 15 1 01047
E1 Op1 Op2 Op3 Weights
Op1 1 3 1 04286
Op2 13 1 13 01429Op3 1 3 1 04286
E3 Op1 Op2 Op3 Weights
Op1 1 12 3 03325
Op2 2 1 3 05278
Op3 13 13 1 01396
Matrices E2 F1 and F2 are 3 3 all-ones matrices
References
[1] International Energy Agency and Organisation for Economic Co-operation andDevelopment Key world energy statistics 2014
[2] Howes T The EUs New Renewable Energy Directive (200928EC) The newclimate policies of the European Union internal legislation and climatediplomacy vol 5 2010 p 117
[3] Mudacumura GM Mebratu D Haque MS Sustainable development policy andadministration Public administration and public policy Boca Raton CRCTaylor and Francis 2006
[4] Gonzaacutelez-Longatt F Gonzaacutelez JS Payoacuten MB Santos JMR Wind-resource atlasof Venezuela based on on-site anemometry observation Renew SustainEnergy Rev 201439898ndash911
[5] Martenson C The crash course the unsustainable future of our economyenergy and environment John Wiley amp Sons Inc Hoboken NJ 2011
[6] Connolly D Lund H Mathiesen BV Leahy M A review of computer tools foranalysing the integration of renewable energy into various energy systemsAppl Energy 201087(4)1059ndash82
[7] Leng GJ Renewable energy technologies project assessment tool RETScreenVarennes Quebec CANMET Energy Diversi1047297cation Research Laboratory 1998
[8] Lilienthal P HOMER micropower optimization model Report National Renew-able Energy Laboratory (NREL) Golden CO 2005
[9] Saaty TL Fundamentals of decision making and priority theory with theanalytic hierarchy processmdashvol VI RWS Publications Pittsburgh PA 1994
[10] Pohekar SD Ramachandran M Application of multi-criteria decision makingto sustainable energy planningmdasha review Renew Sustain Energy Rev 20048(4)365ndash81
[11] Banos R Manzano-Agugliaro F Montoya FG Gil C Alcayde A Goacutemez JOptimization methods applied to renewable and sustainable energy a reviewRenew Sustain Energy Rev 201115(4)1753ndash66
[12] Taha RA Daim T Multi-criteria applications in renewable energy analysis aliterature review London Springer 2013 p 17ndash30
[13] Elkarmi F Mustafa I Increasing the utilization of solar-energy technologies
(set) in Jordanmdashanalytic hierarchy process Energy Policy 199321(9)978ndash82[14] Akash BA Mamlook R Mohsen MS Multi-criteria selection of electric power
plants using analytical hierarchy process Electr Power Syst Res 199952(1)29ndash35
[15] Wu ZX Wei ZH Mitigation assessment results and priorities for Chinasenergy sector Appl Energy 199756(3ndash4)237ndash51
[16] Wang YM Chin KS Fuzzy analytic hierarchy process a logarithmic fuzzypreference programming methodology Int J Approx Reason 201152(4)541ndash53
[17] Nigim K Munier N Green J Pre-feasibility MCDM tools to aid communities inprioritizing local viable renewable energy sources Renew Energy 200429(11)1775ndash91
[18] Daniel J Vishal NV Albert B Selvarsan I Evaluation of the signi 1047297cant renew-able energy resources in India using analytical hierarchy process LondonSpringer 2010 p 13ndash26
[19] Ahmad S Tahar RM Selection of renewable energy sources for sustainabledevelopment of electricity generation system using analytic hierarchy processa case of Malaysia Renew Energy 201463458ndash66
[20] Tasri A Susilawati A Selection among renewable energy alternatives based ona fuzzy analytic hierarchy process in Indonesia Sustain Energy Technol Assess2014734ndash44
[21] Kaya T Kahraman C Multicriteria renewable energy planning using anintegrated fuzzy VIKOR amp AHP methodology the case of Istanbul Energy2010352517ndash27
[22] Zakerinia M Ghaderi F Piltan M Optimal portfolio selection between differentkinds of renewable energy sources 2010
[23] Barin A Canha LN Abaide A Magnago K Wottrich B Multicriteria analysis of the operation of renewable energy sources taking as basis the AHP methodand fuzzy logic concerning distributed generation systems Online J ElectronElectric Eng (OJEEE) 20091(1)52ndash7
[24] Heo E Kim J Boo KJ Analysis of the assessment factors for renewable energydissemination program evaluation using fuzzy AHP Renew Sustain EnergyRev 201014(8)2214ndash20
[25] Akhgari P Kamalan H Monavari M Utilizing AHP method in rating of highlighted environment parameters in most used renewable energies inelectricity production Eur J Sci Res 201160182ndash8
[26] Murphy DJ Hall CAS Year in reviewmdashEROI or energy return on (energy)
invested Ecol Econ Rev 20101185102ndash18[27] Urdaneta AJ Chankong V A multiobjective minimax approach to controller
settings for systems running under disturbances Control-Theory Adv Technol19895(4)391ndash411
[28] Gonzaacutelez-Longatt F Teraacuten R Meacutendez J Hernaacutendez A Guilleacuten F Evaluacioacuten delRecurso Eoacutelico en Venezuela Parte I Congreso Petrolero Energeacutetico ASME-UNEFA Puerto Cabello 2006 [in Spanish]
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110110
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 511
example of a renewable energy with high EROEI is hydropower
with typical ratio higher than 1001 In the other hand biodiesel
and bitumen oil from tar sands (as in the Venezuelan Orinoco oil
belt) are among the energy sources with the lower EROEI ranging
from 13 to 41 The impacts of these assumptions are discussed
later in the study case section
The matrices for levels 0 1 and 2 are now presented they are
constructed under the assumptions described before They can
also contain the results of an expert meeting or a community pollevaluating the different pair-wise comparisons When selecting
the pair-wise comparison values in a three dimension matrix at
least one value is specially selected to minimize the inconsistency
ratio from Eq (3)
Goal TecEconAsp SocEnvAsp Weights
TecEconAsp 1 3 075
SocEnvAsp 13 1 025
TecEconAsp A AvailRes BTecFeas CFinVia Weights
A AvailRes 1 3 5 06370
BTecFeas 13 1 3 02583CFinVia 15 13 1 01047
SocEnvAsp DEcolImp EEduPot FEcoSocImp Weights
DEcolImp 1 3 13 02684
EEduPot 13 1 14 01172
FEcoSocImp 3 4 1 06144
A AvailRes A1 A2 A3 Weights
A1 1 7 1 04869
A2 17 1 15 00778
A3 1 5 1 04353
BTecFeas B1 B2 B3 Weights
B1 1 13 19 00704
B2 3 1 15 01782
B3 9 5 1 07514
CFinVia C1 C2 C3 Weights
C1 1 1 3 04286
C2 1 1 3 04286
C3 13 13 1 01429
DEcolImp D1 D2 D3 Weights
D1 1 5 7 07306D2 15 1 3 01884
D3 17 13 1 00810
EEduPot E1 E2 E3 Weights
E1 1 2 13 02493
E2 12 1 13 01571E3 3 3 1 05936
FEcoSocImp F1 F2 Weights
F1 1 15 01667
F2 5 1 08333
In the following sections the practical use of the AHP method is
given with some application examples For an evaluation of the
proposed methodology different study scenarios are presented
3 Study case
To illustrate the proposed planning methodology a study case
is presented in this section The principle is to design a hybrid
power generation system based on ARE sources in Margarita
Island in north-east Venezuela The application is based on an
isolated home without connection to the national electrical grid
For this study case solar wind and fuel resources are con-
sidered as available on the project site The diesel fuel is available
for portable electric power generation Data of wind resource in
Margarita Island are obtained from [28] Data for solar irradiance
was obtained from 10 min measurements in the central Venezue-
lan region (Caracas) These measures are available at httpcbm
usbveclima For illustration purposes of this example the statis-
tical data for solar irradiation in the central Venezuelan region is
used as the solar resource in the island Using an extrapolation to
the island geographical coordinates the HOMER is able to generate
stochastic time series hourly data for both wind and solar
resources Input data of equipment costs and electrical load pro1047297le
were obtained from typical HOMER data library The load pro1047297le
considered is assumed as a typical representation of small house
electrical energy consumption The availability curves of wind and
solar resources and the electrical load pro1047297le considered are
presented in Figs 5 and 6 respectively
Before de1047297ning the possible candidate options for a successful
decision making strategy comparisons associated with levels 0ndash2
(cf Fig 4) may be de1047297ned For this the assumptions considered for
the study are now de1047297ned These assumptions may change for
different project designs and as before in their de1047297nition may be
involved the planning team expert groups or even the organized
community
Some of the assumptions for the study and construction of thepair-wise comparison matrices are as follows
Fig 5 Solar and wind resources considered for Margarita Island
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110104
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 611
Within the technical and economic aspects the availability of
resources is considered as a key parameter with a higher
importance to the technical feasibility and economic viability In the social and environmental aspects it is given a slightly
higher importance to economic and social impact in the
community over the ecological impact and the educational
potential of the project A higher importance is considered for the following para-
meters the amount of energy produced the energy type
(continuous intermittent etc) the concept of net energy and
the importance in promoting social development in the com-
munity where the project is executed
Several scenarios for this study case are now presented for
different diesel prices with low intermediate and high price levels
31 Low diesel price scenario
In this 1047297rst case study the application example presented in
the previous section is considered under the assumption that
diesel fuel is available at current Venezuelan market prices ie
0029$L (price for July 2012)
The design of the hybrid system considers solar panels wind
turbines and diesel generators as energy sources The design
problem is reduced to 1047297nding the optimal size of each system
component For the speci1047297c hybrid generator system problem
search variables are the size (nominal power in W) of the powerconverter the battery stack the solar photovoltaic panels the
wind turbines and the diesel generator A backup battery system is
considered for increased system reliability The power converter
in this case an inverter is necessary to couple DC and AC buses
Additional converters may be considered for DCDC coupling
between ARE sources For example boost or buck converter may
be used to improve system controllability and operating perfor-
mance (improved ef 1047297ciency control to optimal operating point
MPPT maximum power point tracker algorithm etc) Restrictions
of 10 minimum capacity shortage and 25 minimum ARE sources
participation are considered With the given resource availability
data and with several proposed hybrid generation system con1047297g-
urations the HOMER is used for optimization and simulation of
results
With the system topology de1047297ned a search matrix is obtained
and it is possible to run the optimization algorithm using HOMER
to 1047297nd the optimal solution When the optimization process
completed the 1047297rst three solutions are presented in Table 3 They
are classi1047297ed in descending order by their total net presentcost (NPC)
The option 1 consists of 025 kW in solar panels 2 SW AIR X
400 W wind turbines a 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Option 2 is composed by
4 AIR X 400 W wind turbines 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Finally option 3 is given by
075 kW in solar panels 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter All batteries are model
S4KS25P 76 kWh (4 V 1900 Ah)
In Fig 7 the simulation results of the operating behavior of the
obtained system con1047297gurations are presented With these results
it is clear that the diesel generation participation is high This is
normal form of a NPC optimization since fuel prices are very low in
Venezuela These three options are 1047297xed as the candidate solutionsfor the second part of the methodology presented in the previous
section using the AHP method
The second part of the proposed methodology deals with the
evaluation of candidate options using the AHP method For this
each option is weighted by comparison with all the selected sub-
criteria (cf Fig 4) Previously described assumptions are used to
evaluate pair-wise comparisons between options and sub-criteria
Additionally as a reference an analysis of the negative impact of
ARE sources to the environment is presented in [25]
In this paper it is proposed that some evaluation weights
should be directly 1047297xed from results obtained with the optimiza-
tion performed using HOMER Then for example for sub-criterion
A1 ldquoElectrical energy capacity shortagerdquo the pair-wise matrix is
not evaluated and the vector weight is directly de1047297ned from
Fig 6 Electric load pro1047297le considered
Table 3
Results obtained with HOMER for the low diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 14 021 569 0301
Option 2 14 698 602 0315
Option 3 14 826 514 0318
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 105
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 711
optimization results with values 115 502 and 482 in kWhyear of
production capacity shortage for options 1 2 and 3 respectively
These values are normalized to the AHP method evaluation scale
(1ndash9) and the weight vector is respectively given by 023 038 and
039 It should be noted that in this case option 3 has the higher
weight because it represents the lower capacity shortage which is
a desired design requirement
The local weights matrix obtained is given in Table 4
The cells 1047297lled with light gray color denote the weights that
were 1047297xed directly from results of the optimization using HOMER
Pair-wise matrices used to obtain this weight matrix are presented
in this paper in Appendix A
The global weight vector is given in Table 5
With these results the global weights for each candidate
option are obtained as
Option 1 03413
Option 2 03247
Option 3 03340
Fig 7 Simulation results (electricity production) of hybrid sources for the three options considered
Table 4
Local weights matrix obtained for low diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 023 014 055 025 025 033 034 034 033 033 033 010 060 033 020 033 033
Op2 038 033 021 059 059 031 033 033 032 034 033 026 020 033 020 033 033
Op3 039 053 024 016 016 036 033 033 035 033 033 064 020 033 060 033 033
Table 5
Global weight vector obtained for low diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110106
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 811
With the results obtained after evaluation of each alternative
option 1 stands as still the best solution according to the evaluated
criteria An important aspect that may have an in1047298uence in this
result is the choice of option 1 as a hybrid energy source with a
high component of continuous energy capacity (the diesel gen-
erator represents the 74 of the energy produced) Using HOMER
a time response simulation is obtained for the selected solution
The simulation result is presented in Fig 8 and is representative of
the operation of the hybrid generator during a complete week The1047297gure shows the levels of energy production for each source the
electrical load demand and the batteries state-of-charge (SOC)
The results show the need of the diesel generator when the
batteries SOC falls beyond 50 during low wind energy
production
32 High diesel price scenario
The second case study scenario is given by the same example as
before but now considering international diesel fuel prices The
price taken as reference is at 161$L (average value in Europe
during July 2012) Using a similar model as in the previous case
the levelized cost optimization using HOMER is performed Results
are shown in Table 6 where the system characteristics arepresented The operating electricity production characteristic of
each con1047297guration is presented in Fig 9 In this case power
generation from solar panels dominates the 1047297rst and second
0 20 40 60 80 100 120 1400
05
1
15
0 20 40 60 80 100 120 14040
60
80
100
Fig 8 Time simulation results hybrid system operating during a week
Table 6
Results obtained with HOMER for the high diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 25 896 383 0590
Option 2 28 408 384 0629
Option 3 29 469 878 0632
Fig 9 Simulation results (electricity production) of hybrid sources for the three options considered in the second case study problem
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 107
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 911
options The third system option is composed by a mixed solar
winddiesel power generator In this second case study option 1 is
composed by 25 kW in solar panels 1 SW AIR X 400 W wind
turbine 1 string of 6 batteries and a 1 kW power inverter Option
2 is a solution with 3 kW in solar panels 1 string of 6 batteries and
a 1 kW power inverter Option 3 is given with 175 kW in solar
panels 3 SW AIR X 400 W wind turbines 500 W diesel generator
1 string of 6 batteries and a 1 kW power inverter All batteries are
model S4KS25P 76 kWh (4 V 1900 Ah)
With a similar analysis as in the previous case the local
weights matrix is given in Table 7 Here light gray cells denote
the weights that were 1047297xed directly from results of the
optimization using HOMER Similarly pair-wise matrices used to
obtain this weight matrix are presented in this paper in Appendix B
The global weight vector is then given in Table 8
Finally global weights of the three options are obtained as the
product of the local weights matrix and the global weight vector
The result is given by
Option 1 02834
Option 2 03056
Option 3 04110
In this case a re-arrangement of 1047297nal solutions is obtained after
completion of the AHP method The 1047297nal problem solution
selection is given by option 3 where as in the previous case a
higher participation of a continuous energy source has an in1047298u-
ence in the 1047297nal result The third option gives the best balance in
terms of net energy as the other two options are primarily
composed by solar panels A 1047297nal critical parameter was the
capacity shortage which turned out to be relatively high for the
1047297rst two options with unmet supplied energy of 304 kWhyearand 173 kWhyear respectively Time response simulation results
for a complete week of operation are presented in Fig 10 A main
characteristic is the high solar power production with the exceed-
ing generated energy used to keep desired levels of battery stack
SOC A good energy balance is obtained with this option with
virtually no capacity shortage during a year round operation
33 Intermediate diesel prices scenarios
Two additional scenarios were considered with intermediate
diesel price values of 05 and 12$L Similar considerations to the
previous cases were used for pair-wise matrix comparisons
For a diesel price of 05$L the results of the HOMER optimiza-
tion are very similar to the low diesel price case All the possiblesolutions in the group of most interesting optimization results
include a 500 W diesel generator This fact re1047298ects that the diesel
prices are still very low and do not have a big in1047298uence in the
project 1047297nal NPC values After applying the described methodol-
ogy the selected optimal system is given by a mix of solar (250 W)
wind (800 W) and diesel generation (500 W)
For a diesel price of 12$L the optimization yields three main
options to be considered
1 The 1047297rst option is given by the installation of PV (2500 W) and
wind (400 W)
2 The second option consists of a mix of PV (1500 W) wind
(1200 W) and diesel (500 W)
3 The third option is given by a purely PV(3000 W) system
Table 7
Local weights matrix obtained for the high diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 018 026 023 030 030 034 035 034 038 047 043 026 043 033 033 033 033
Op2 032 064 012 016 016 036 033 033 038 047 043 064 014 033 053 033 033
Op3 050 010 065 054 054 030 032 033 023 005 014 010 043 033 014 033 033
Table 8Global weight vector obtained for the high diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
0 20 40 60 80 100 120 1400
05
1
15
2
0 20 40 60 80 100 120 14060
70
80
90
100
Fig 10 Time simulation results hybrid system operating during a week for the
second case study problem
02 04 06 08 1 12 14 16200
400
600
800
1000
1200
1400
1600
1800
Fig 11 PV and wind systems sizes vs diesel prices for each scenario
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110108
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1011
The 1047297nal weight vector gives a 02835 weight for the 1047297rst
04093 for the second and 03072 for the third option The selected
solution is the second option the mix with high PV and wind
participation The storage device obtained for all optimal solutions
is a string of 6 batteries (456 kWh) with a 1 kW power inverter
Fig 11 shows the evolution of PV and wind installation sizes for
the selected solution as a function of the diesel prices considered
in the different scenarios presented in this paperIt can be observed that for the different considerations used in
the pair-wise matrix comparisons at all levels obtained results for
higher diesel prices tend to show an overall increased PV penetra-
tion ratio while for the low diesel price scenarios the selection of
wind power is privileged over PV More importantly the results
show the importance of considering social and environmental
aspects This is shown in the case of a high penetration of PV or
wind when the 1047297nal selected solution is not necessarily the result
of the technical and economical optimization
4 Conclusion
In this paper a hierarchical planning methodology for the
integral net energy design of small-scale hybrid renewable energysystems using ARE sources is proposed The approach considers
technicalndasheconomical as well as societalndashenvironmental aspects in
two different decision layers One of the main features of the
methodology is to complement the classical methodology that
considers the optimization and levelized cost analysis by using the
EROEI or net energy concept and the AHP method for decision
making problems to ensure the consideration of complex and
unconventional aspects such as social and environmental para-
meters in the design of the hybrid ARE systems The methodology
is evaluated using an application example in Margarita Island
under several simulation scenarios low intermediate and high
diesel fuel price cases The obtained results show that the aspects
considered in the proposed methodology have an in1047298uence on the
initial results obtained by using the classical approach of levelizedcost optimization
Acknowledgements
The authors would like to acknowledge the 1047297nancial and
technical support of the Venezuelan Planning School (Escuela
Venezolana de Plani1047297cacioacuten)
Appendix A Pair-wise matrices for low diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 13 01396
Op2 3 1 12 03325Op3 3 2 1 05278
A3 Op1 Op2 Op3 Weights
Op1 1 3 2 05499
Op2 13 1 1 02098
Op3 12 1 1 02402
B1 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
B2 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
D3 Op1 Op2 Op3 Weights
Op1 1 13 15 01047
Op2 3 1 13 02583Op3 5 3 1 06370
E1 Op1 Op2 Op3 Weights
Op1 1 3 3 06000
Op2 13 1 1 02000
Op3 13 1 1 02000
E3 Op1 Op2 Op3 Weights
Op1 1 1 13 02000
Op2 1 1 13 02000Op3 3 3 1 06000
Matrices D2 E2 F1 and F2 are 3 3 all-ones matrices
Appendix B Pair-wise matrices for high diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 3 02583Op2 3 1 5 06370
Op3 13 15 1 01047
A3 Op1 Op2 Op3 Weights
Op1 1 2 13 02297
Op2 12 1 15 01220
Op3 3 5 1 06483
B1 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634
Op3 2 3 1 05396
B2 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634Op3 2 3 1 05396
D1 Op1 Op2 Op3 Weights
Op1 1 1 9 04737
Op2 1 1 9 04737
Op3 19 19 1 00526
D2 Op1 Op2 Op3 Weights
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 109
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1111
Op1 1 1 3 04286Op2 1 1 3 04286
Op3 13 13 1 01429
D3 Op1 Op2 Op3 Weights
Op1 1 13 3 02583
Op2 3 1 5 06370
Op3 13 15 1 01047
E1 Op1 Op2 Op3 Weights
Op1 1 3 1 04286
Op2 13 1 13 01429Op3 1 3 1 04286
E3 Op1 Op2 Op3 Weights
Op1 1 12 3 03325
Op2 2 1 3 05278
Op3 13 13 1 01396
Matrices E2 F1 and F2 are 3 3 all-ones matrices
References
[1] International Energy Agency and Organisation for Economic Co-operation andDevelopment Key world energy statistics 2014
[2] Howes T The EUs New Renewable Energy Directive (200928EC) The newclimate policies of the European Union internal legislation and climatediplomacy vol 5 2010 p 117
[3] Mudacumura GM Mebratu D Haque MS Sustainable development policy andadministration Public administration and public policy Boca Raton CRCTaylor and Francis 2006
[4] Gonzaacutelez-Longatt F Gonzaacutelez JS Payoacuten MB Santos JMR Wind-resource atlasof Venezuela based on on-site anemometry observation Renew SustainEnergy Rev 201439898ndash911
[5] Martenson C The crash course the unsustainable future of our economyenergy and environment John Wiley amp Sons Inc Hoboken NJ 2011
[6] Connolly D Lund H Mathiesen BV Leahy M A review of computer tools foranalysing the integration of renewable energy into various energy systemsAppl Energy 201087(4)1059ndash82
[7] Leng GJ Renewable energy technologies project assessment tool RETScreenVarennes Quebec CANMET Energy Diversi1047297cation Research Laboratory 1998
[8] Lilienthal P HOMER micropower optimization model Report National Renew-able Energy Laboratory (NREL) Golden CO 2005
[9] Saaty TL Fundamentals of decision making and priority theory with theanalytic hierarchy processmdashvol VI RWS Publications Pittsburgh PA 1994
[10] Pohekar SD Ramachandran M Application of multi-criteria decision makingto sustainable energy planningmdasha review Renew Sustain Energy Rev 20048(4)365ndash81
[11] Banos R Manzano-Agugliaro F Montoya FG Gil C Alcayde A Goacutemez JOptimization methods applied to renewable and sustainable energy a reviewRenew Sustain Energy Rev 201115(4)1753ndash66
[12] Taha RA Daim T Multi-criteria applications in renewable energy analysis aliterature review London Springer 2013 p 17ndash30
[13] Elkarmi F Mustafa I Increasing the utilization of solar-energy technologies
(set) in Jordanmdashanalytic hierarchy process Energy Policy 199321(9)978ndash82[14] Akash BA Mamlook R Mohsen MS Multi-criteria selection of electric power
plants using analytical hierarchy process Electr Power Syst Res 199952(1)29ndash35
[15] Wu ZX Wei ZH Mitigation assessment results and priorities for Chinasenergy sector Appl Energy 199756(3ndash4)237ndash51
[16] Wang YM Chin KS Fuzzy analytic hierarchy process a logarithmic fuzzypreference programming methodology Int J Approx Reason 201152(4)541ndash53
[17] Nigim K Munier N Green J Pre-feasibility MCDM tools to aid communities inprioritizing local viable renewable energy sources Renew Energy 200429(11)1775ndash91
[18] Daniel J Vishal NV Albert B Selvarsan I Evaluation of the signi 1047297cant renew-able energy resources in India using analytical hierarchy process LondonSpringer 2010 p 13ndash26
[19] Ahmad S Tahar RM Selection of renewable energy sources for sustainabledevelopment of electricity generation system using analytic hierarchy processa case of Malaysia Renew Energy 201463458ndash66
[20] Tasri A Susilawati A Selection among renewable energy alternatives based ona fuzzy analytic hierarchy process in Indonesia Sustain Energy Technol Assess2014734ndash44
[21] Kaya T Kahraman C Multicriteria renewable energy planning using anintegrated fuzzy VIKOR amp AHP methodology the case of Istanbul Energy2010352517ndash27
[22] Zakerinia M Ghaderi F Piltan M Optimal portfolio selection between differentkinds of renewable energy sources 2010
[23] Barin A Canha LN Abaide A Magnago K Wottrich B Multicriteria analysis of the operation of renewable energy sources taking as basis the AHP methodand fuzzy logic concerning distributed generation systems Online J ElectronElectric Eng (OJEEE) 20091(1)52ndash7
[24] Heo E Kim J Boo KJ Analysis of the assessment factors for renewable energydissemination program evaluation using fuzzy AHP Renew Sustain EnergyRev 201014(8)2214ndash20
[25] Akhgari P Kamalan H Monavari M Utilizing AHP method in rating of highlighted environment parameters in most used renewable energies inelectricity production Eur J Sci Res 201160182ndash8
[26] Murphy DJ Hall CAS Year in reviewmdashEROI or energy return on (energy)
invested Ecol Econ Rev 20101185102ndash18[27] Urdaneta AJ Chankong V A multiobjective minimax approach to controller
settings for systems running under disturbances Control-Theory Adv Technol19895(4)391ndash411
[28] Gonzaacutelez-Longatt F Teraacuten R Meacutendez J Hernaacutendez A Guilleacuten F Evaluacioacuten delRecurso Eoacutelico en Venezuela Parte I Congreso Petrolero Energeacutetico ASME-UNEFA Puerto Cabello 2006 [in Spanish]
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110110
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 611
Within the technical and economic aspects the availability of
resources is considered as a key parameter with a higher
importance to the technical feasibility and economic viability In the social and environmental aspects it is given a slightly
higher importance to economic and social impact in the
community over the ecological impact and the educational
potential of the project A higher importance is considered for the following para-
meters the amount of energy produced the energy type
(continuous intermittent etc) the concept of net energy and
the importance in promoting social development in the com-
munity where the project is executed
Several scenarios for this study case are now presented for
different diesel prices with low intermediate and high price levels
31 Low diesel price scenario
In this 1047297rst case study the application example presented in
the previous section is considered under the assumption that
diesel fuel is available at current Venezuelan market prices ie
0029$L (price for July 2012)
The design of the hybrid system considers solar panels wind
turbines and diesel generators as energy sources The design
problem is reduced to 1047297nding the optimal size of each system
component For the speci1047297c hybrid generator system problem
search variables are the size (nominal power in W) of the powerconverter the battery stack the solar photovoltaic panels the
wind turbines and the diesel generator A backup battery system is
considered for increased system reliability The power converter
in this case an inverter is necessary to couple DC and AC buses
Additional converters may be considered for DCDC coupling
between ARE sources For example boost or buck converter may
be used to improve system controllability and operating perfor-
mance (improved ef 1047297ciency control to optimal operating point
MPPT maximum power point tracker algorithm etc) Restrictions
of 10 minimum capacity shortage and 25 minimum ARE sources
participation are considered With the given resource availability
data and with several proposed hybrid generation system con1047297g-
urations the HOMER is used for optimization and simulation of
results
With the system topology de1047297ned a search matrix is obtained
and it is possible to run the optimization algorithm using HOMER
to 1047297nd the optimal solution When the optimization process
completed the 1047297rst three solutions are presented in Table 3 They
are classi1047297ed in descending order by their total net presentcost (NPC)
The option 1 consists of 025 kW in solar panels 2 SW AIR X
400 W wind turbines a 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Option 2 is composed by
4 AIR X 400 W wind turbines 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter Finally option 3 is given by
075 kW in solar panels 500 W diesel generator 1 string of
6 batteries and a 1 kW power inverter All batteries are model
S4KS25P 76 kWh (4 V 1900 Ah)
In Fig 7 the simulation results of the operating behavior of the
obtained system con1047297gurations are presented With these results
it is clear that the diesel generation participation is high This is
normal form of a NPC optimization since fuel prices are very low in
Venezuela These three options are 1047297xed as the candidate solutionsfor the second part of the methodology presented in the previous
section using the AHP method
The second part of the proposed methodology deals with the
evaluation of candidate options using the AHP method For this
each option is weighted by comparison with all the selected sub-
criteria (cf Fig 4) Previously described assumptions are used to
evaluate pair-wise comparisons between options and sub-criteria
Additionally as a reference an analysis of the negative impact of
ARE sources to the environment is presented in [25]
In this paper it is proposed that some evaluation weights
should be directly 1047297xed from results obtained with the optimiza-
tion performed using HOMER Then for example for sub-criterion
A1 ldquoElectrical energy capacity shortagerdquo the pair-wise matrix is
not evaluated and the vector weight is directly de1047297ned from
Fig 6 Electric load pro1047297le considered
Table 3
Results obtained with HOMER for the low diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 14 021 569 0301
Option 2 14 698 602 0315
Option 3 14 826 514 0318
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 105
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 711
optimization results with values 115 502 and 482 in kWhyear of
production capacity shortage for options 1 2 and 3 respectively
These values are normalized to the AHP method evaluation scale
(1ndash9) and the weight vector is respectively given by 023 038 and
039 It should be noted that in this case option 3 has the higher
weight because it represents the lower capacity shortage which is
a desired design requirement
The local weights matrix obtained is given in Table 4
The cells 1047297lled with light gray color denote the weights that
were 1047297xed directly from results of the optimization using HOMER
Pair-wise matrices used to obtain this weight matrix are presented
in this paper in Appendix A
The global weight vector is given in Table 5
With these results the global weights for each candidate
option are obtained as
Option 1 03413
Option 2 03247
Option 3 03340
Fig 7 Simulation results (electricity production) of hybrid sources for the three options considered
Table 4
Local weights matrix obtained for low diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 023 014 055 025 025 033 034 034 033 033 033 010 060 033 020 033 033
Op2 038 033 021 059 059 031 033 033 032 034 033 026 020 033 020 033 033
Op3 039 053 024 016 016 036 033 033 035 033 033 064 020 033 060 033 033
Table 5
Global weight vector obtained for low diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110106
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 811
With the results obtained after evaluation of each alternative
option 1 stands as still the best solution according to the evaluated
criteria An important aspect that may have an in1047298uence in this
result is the choice of option 1 as a hybrid energy source with a
high component of continuous energy capacity (the diesel gen-
erator represents the 74 of the energy produced) Using HOMER
a time response simulation is obtained for the selected solution
The simulation result is presented in Fig 8 and is representative of
the operation of the hybrid generator during a complete week The1047297gure shows the levels of energy production for each source the
electrical load demand and the batteries state-of-charge (SOC)
The results show the need of the diesel generator when the
batteries SOC falls beyond 50 during low wind energy
production
32 High diesel price scenario
The second case study scenario is given by the same example as
before but now considering international diesel fuel prices The
price taken as reference is at 161$L (average value in Europe
during July 2012) Using a similar model as in the previous case
the levelized cost optimization using HOMER is performed Results
are shown in Table 6 where the system characteristics arepresented The operating electricity production characteristic of
each con1047297guration is presented in Fig 9 In this case power
generation from solar panels dominates the 1047297rst and second
0 20 40 60 80 100 120 1400
05
1
15
0 20 40 60 80 100 120 14040
60
80
100
Fig 8 Time simulation results hybrid system operating during a week
Table 6
Results obtained with HOMER for the high diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 25 896 383 0590
Option 2 28 408 384 0629
Option 3 29 469 878 0632
Fig 9 Simulation results (electricity production) of hybrid sources for the three options considered in the second case study problem
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 107
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 911
options The third system option is composed by a mixed solar
winddiesel power generator In this second case study option 1 is
composed by 25 kW in solar panels 1 SW AIR X 400 W wind
turbine 1 string of 6 batteries and a 1 kW power inverter Option
2 is a solution with 3 kW in solar panels 1 string of 6 batteries and
a 1 kW power inverter Option 3 is given with 175 kW in solar
panels 3 SW AIR X 400 W wind turbines 500 W diesel generator
1 string of 6 batteries and a 1 kW power inverter All batteries are
model S4KS25P 76 kWh (4 V 1900 Ah)
With a similar analysis as in the previous case the local
weights matrix is given in Table 7 Here light gray cells denote
the weights that were 1047297xed directly from results of the
optimization using HOMER Similarly pair-wise matrices used to
obtain this weight matrix are presented in this paper in Appendix B
The global weight vector is then given in Table 8
Finally global weights of the three options are obtained as the
product of the local weights matrix and the global weight vector
The result is given by
Option 1 02834
Option 2 03056
Option 3 04110
In this case a re-arrangement of 1047297nal solutions is obtained after
completion of the AHP method The 1047297nal problem solution
selection is given by option 3 where as in the previous case a
higher participation of a continuous energy source has an in1047298u-
ence in the 1047297nal result The third option gives the best balance in
terms of net energy as the other two options are primarily
composed by solar panels A 1047297nal critical parameter was the
capacity shortage which turned out to be relatively high for the
1047297rst two options with unmet supplied energy of 304 kWhyearand 173 kWhyear respectively Time response simulation results
for a complete week of operation are presented in Fig 10 A main
characteristic is the high solar power production with the exceed-
ing generated energy used to keep desired levels of battery stack
SOC A good energy balance is obtained with this option with
virtually no capacity shortage during a year round operation
33 Intermediate diesel prices scenarios
Two additional scenarios were considered with intermediate
diesel price values of 05 and 12$L Similar considerations to the
previous cases were used for pair-wise matrix comparisons
For a diesel price of 05$L the results of the HOMER optimiza-
tion are very similar to the low diesel price case All the possiblesolutions in the group of most interesting optimization results
include a 500 W diesel generator This fact re1047298ects that the diesel
prices are still very low and do not have a big in1047298uence in the
project 1047297nal NPC values After applying the described methodol-
ogy the selected optimal system is given by a mix of solar (250 W)
wind (800 W) and diesel generation (500 W)
For a diesel price of 12$L the optimization yields three main
options to be considered
1 The 1047297rst option is given by the installation of PV (2500 W) and
wind (400 W)
2 The second option consists of a mix of PV (1500 W) wind
(1200 W) and diesel (500 W)
3 The third option is given by a purely PV(3000 W) system
Table 7
Local weights matrix obtained for the high diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 018 026 023 030 030 034 035 034 038 047 043 026 043 033 033 033 033
Op2 032 064 012 016 016 036 033 033 038 047 043 064 014 033 053 033 033
Op3 050 010 065 054 054 030 032 033 023 005 014 010 043 033 014 033 033
Table 8Global weight vector obtained for the high diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
0 20 40 60 80 100 120 1400
05
1
15
2
0 20 40 60 80 100 120 14060
70
80
90
100
Fig 10 Time simulation results hybrid system operating during a week for the
second case study problem
02 04 06 08 1 12 14 16200
400
600
800
1000
1200
1400
1600
1800
Fig 11 PV and wind systems sizes vs diesel prices for each scenario
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110108
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1011
The 1047297nal weight vector gives a 02835 weight for the 1047297rst
04093 for the second and 03072 for the third option The selected
solution is the second option the mix with high PV and wind
participation The storage device obtained for all optimal solutions
is a string of 6 batteries (456 kWh) with a 1 kW power inverter
Fig 11 shows the evolution of PV and wind installation sizes for
the selected solution as a function of the diesel prices considered
in the different scenarios presented in this paperIt can be observed that for the different considerations used in
the pair-wise matrix comparisons at all levels obtained results for
higher diesel prices tend to show an overall increased PV penetra-
tion ratio while for the low diesel price scenarios the selection of
wind power is privileged over PV More importantly the results
show the importance of considering social and environmental
aspects This is shown in the case of a high penetration of PV or
wind when the 1047297nal selected solution is not necessarily the result
of the technical and economical optimization
4 Conclusion
In this paper a hierarchical planning methodology for the
integral net energy design of small-scale hybrid renewable energysystems using ARE sources is proposed The approach considers
technicalndasheconomical as well as societalndashenvironmental aspects in
two different decision layers One of the main features of the
methodology is to complement the classical methodology that
considers the optimization and levelized cost analysis by using the
EROEI or net energy concept and the AHP method for decision
making problems to ensure the consideration of complex and
unconventional aspects such as social and environmental para-
meters in the design of the hybrid ARE systems The methodology
is evaluated using an application example in Margarita Island
under several simulation scenarios low intermediate and high
diesel fuel price cases The obtained results show that the aspects
considered in the proposed methodology have an in1047298uence on the
initial results obtained by using the classical approach of levelizedcost optimization
Acknowledgements
The authors would like to acknowledge the 1047297nancial and
technical support of the Venezuelan Planning School (Escuela
Venezolana de Plani1047297cacioacuten)
Appendix A Pair-wise matrices for low diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 13 01396
Op2 3 1 12 03325Op3 3 2 1 05278
A3 Op1 Op2 Op3 Weights
Op1 1 3 2 05499
Op2 13 1 1 02098
Op3 12 1 1 02402
B1 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
B2 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
D3 Op1 Op2 Op3 Weights
Op1 1 13 15 01047
Op2 3 1 13 02583Op3 5 3 1 06370
E1 Op1 Op2 Op3 Weights
Op1 1 3 3 06000
Op2 13 1 1 02000
Op3 13 1 1 02000
E3 Op1 Op2 Op3 Weights
Op1 1 1 13 02000
Op2 1 1 13 02000Op3 3 3 1 06000
Matrices D2 E2 F1 and F2 are 3 3 all-ones matrices
Appendix B Pair-wise matrices for high diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 3 02583Op2 3 1 5 06370
Op3 13 15 1 01047
A3 Op1 Op2 Op3 Weights
Op1 1 2 13 02297
Op2 12 1 15 01220
Op3 3 5 1 06483
B1 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634
Op3 2 3 1 05396
B2 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634Op3 2 3 1 05396
D1 Op1 Op2 Op3 Weights
Op1 1 1 9 04737
Op2 1 1 9 04737
Op3 19 19 1 00526
D2 Op1 Op2 Op3 Weights
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 109
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1111
Op1 1 1 3 04286Op2 1 1 3 04286
Op3 13 13 1 01429
D3 Op1 Op2 Op3 Weights
Op1 1 13 3 02583
Op2 3 1 5 06370
Op3 13 15 1 01047
E1 Op1 Op2 Op3 Weights
Op1 1 3 1 04286
Op2 13 1 13 01429Op3 1 3 1 04286
E3 Op1 Op2 Op3 Weights
Op1 1 12 3 03325
Op2 2 1 3 05278
Op3 13 13 1 01396
Matrices E2 F1 and F2 are 3 3 all-ones matrices
References
[1] International Energy Agency and Organisation for Economic Co-operation andDevelopment Key world energy statistics 2014
[2] Howes T The EUs New Renewable Energy Directive (200928EC) The newclimate policies of the European Union internal legislation and climatediplomacy vol 5 2010 p 117
[3] Mudacumura GM Mebratu D Haque MS Sustainable development policy andadministration Public administration and public policy Boca Raton CRCTaylor and Francis 2006
[4] Gonzaacutelez-Longatt F Gonzaacutelez JS Payoacuten MB Santos JMR Wind-resource atlasof Venezuela based on on-site anemometry observation Renew SustainEnergy Rev 201439898ndash911
[5] Martenson C The crash course the unsustainable future of our economyenergy and environment John Wiley amp Sons Inc Hoboken NJ 2011
[6] Connolly D Lund H Mathiesen BV Leahy M A review of computer tools foranalysing the integration of renewable energy into various energy systemsAppl Energy 201087(4)1059ndash82
[7] Leng GJ Renewable energy technologies project assessment tool RETScreenVarennes Quebec CANMET Energy Diversi1047297cation Research Laboratory 1998
[8] Lilienthal P HOMER micropower optimization model Report National Renew-able Energy Laboratory (NREL) Golden CO 2005
[9] Saaty TL Fundamentals of decision making and priority theory with theanalytic hierarchy processmdashvol VI RWS Publications Pittsburgh PA 1994
[10] Pohekar SD Ramachandran M Application of multi-criteria decision makingto sustainable energy planningmdasha review Renew Sustain Energy Rev 20048(4)365ndash81
[11] Banos R Manzano-Agugliaro F Montoya FG Gil C Alcayde A Goacutemez JOptimization methods applied to renewable and sustainable energy a reviewRenew Sustain Energy Rev 201115(4)1753ndash66
[12] Taha RA Daim T Multi-criteria applications in renewable energy analysis aliterature review London Springer 2013 p 17ndash30
[13] Elkarmi F Mustafa I Increasing the utilization of solar-energy technologies
(set) in Jordanmdashanalytic hierarchy process Energy Policy 199321(9)978ndash82[14] Akash BA Mamlook R Mohsen MS Multi-criteria selection of electric power
plants using analytical hierarchy process Electr Power Syst Res 199952(1)29ndash35
[15] Wu ZX Wei ZH Mitigation assessment results and priorities for Chinasenergy sector Appl Energy 199756(3ndash4)237ndash51
[16] Wang YM Chin KS Fuzzy analytic hierarchy process a logarithmic fuzzypreference programming methodology Int J Approx Reason 201152(4)541ndash53
[17] Nigim K Munier N Green J Pre-feasibility MCDM tools to aid communities inprioritizing local viable renewable energy sources Renew Energy 200429(11)1775ndash91
[18] Daniel J Vishal NV Albert B Selvarsan I Evaluation of the signi 1047297cant renew-able energy resources in India using analytical hierarchy process LondonSpringer 2010 p 13ndash26
[19] Ahmad S Tahar RM Selection of renewable energy sources for sustainabledevelopment of electricity generation system using analytic hierarchy processa case of Malaysia Renew Energy 201463458ndash66
[20] Tasri A Susilawati A Selection among renewable energy alternatives based ona fuzzy analytic hierarchy process in Indonesia Sustain Energy Technol Assess2014734ndash44
[21] Kaya T Kahraman C Multicriteria renewable energy planning using anintegrated fuzzy VIKOR amp AHP methodology the case of Istanbul Energy2010352517ndash27
[22] Zakerinia M Ghaderi F Piltan M Optimal portfolio selection between differentkinds of renewable energy sources 2010
[23] Barin A Canha LN Abaide A Magnago K Wottrich B Multicriteria analysis of the operation of renewable energy sources taking as basis the AHP methodand fuzzy logic concerning distributed generation systems Online J ElectronElectric Eng (OJEEE) 20091(1)52ndash7
[24] Heo E Kim J Boo KJ Analysis of the assessment factors for renewable energydissemination program evaluation using fuzzy AHP Renew Sustain EnergyRev 201014(8)2214ndash20
[25] Akhgari P Kamalan H Monavari M Utilizing AHP method in rating of highlighted environment parameters in most used renewable energies inelectricity production Eur J Sci Res 201160182ndash8
[26] Murphy DJ Hall CAS Year in reviewmdashEROI or energy return on (energy)
invested Ecol Econ Rev 20101185102ndash18[27] Urdaneta AJ Chankong V A multiobjective minimax approach to controller
settings for systems running under disturbances Control-Theory Adv Technol19895(4)391ndash411
[28] Gonzaacutelez-Longatt F Teraacuten R Meacutendez J Hernaacutendez A Guilleacuten F Evaluacioacuten delRecurso Eoacutelico en Venezuela Parte I Congreso Petrolero Energeacutetico ASME-UNEFA Puerto Cabello 2006 [in Spanish]
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110110
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 711
optimization results with values 115 502 and 482 in kWhyear of
production capacity shortage for options 1 2 and 3 respectively
These values are normalized to the AHP method evaluation scale
(1ndash9) and the weight vector is respectively given by 023 038 and
039 It should be noted that in this case option 3 has the higher
weight because it represents the lower capacity shortage which is
a desired design requirement
The local weights matrix obtained is given in Table 4
The cells 1047297lled with light gray color denote the weights that
were 1047297xed directly from results of the optimization using HOMER
Pair-wise matrices used to obtain this weight matrix are presented
in this paper in Appendix A
The global weight vector is given in Table 5
With these results the global weights for each candidate
option are obtained as
Option 1 03413
Option 2 03247
Option 3 03340
Fig 7 Simulation results (electricity production) of hybrid sources for the three options considered
Table 4
Local weights matrix obtained for low diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 023 014 055 025 025 033 034 034 033 033 033 010 060 033 020 033 033
Op2 038 033 021 059 059 031 033 033 032 034 033 026 020 033 020 033 033
Op3 039 053 024 016 016 036 033 033 035 033 033 064 020 033 060 033 033
Table 5
Global weight vector obtained for low diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110106
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 811
With the results obtained after evaluation of each alternative
option 1 stands as still the best solution according to the evaluated
criteria An important aspect that may have an in1047298uence in this
result is the choice of option 1 as a hybrid energy source with a
high component of continuous energy capacity (the diesel gen-
erator represents the 74 of the energy produced) Using HOMER
a time response simulation is obtained for the selected solution
The simulation result is presented in Fig 8 and is representative of
the operation of the hybrid generator during a complete week The1047297gure shows the levels of energy production for each source the
electrical load demand and the batteries state-of-charge (SOC)
The results show the need of the diesel generator when the
batteries SOC falls beyond 50 during low wind energy
production
32 High diesel price scenario
The second case study scenario is given by the same example as
before but now considering international diesel fuel prices The
price taken as reference is at 161$L (average value in Europe
during July 2012) Using a similar model as in the previous case
the levelized cost optimization using HOMER is performed Results
are shown in Table 6 where the system characteristics arepresented The operating electricity production characteristic of
each con1047297guration is presented in Fig 9 In this case power
generation from solar panels dominates the 1047297rst and second
0 20 40 60 80 100 120 1400
05
1
15
0 20 40 60 80 100 120 14040
60
80
100
Fig 8 Time simulation results hybrid system operating during a week
Table 6
Results obtained with HOMER for the high diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 25 896 383 0590
Option 2 28 408 384 0629
Option 3 29 469 878 0632
Fig 9 Simulation results (electricity production) of hybrid sources for the three options considered in the second case study problem
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 107
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 911
options The third system option is composed by a mixed solar
winddiesel power generator In this second case study option 1 is
composed by 25 kW in solar panels 1 SW AIR X 400 W wind
turbine 1 string of 6 batteries and a 1 kW power inverter Option
2 is a solution with 3 kW in solar panels 1 string of 6 batteries and
a 1 kW power inverter Option 3 is given with 175 kW in solar
panels 3 SW AIR X 400 W wind turbines 500 W diesel generator
1 string of 6 batteries and a 1 kW power inverter All batteries are
model S4KS25P 76 kWh (4 V 1900 Ah)
With a similar analysis as in the previous case the local
weights matrix is given in Table 7 Here light gray cells denote
the weights that were 1047297xed directly from results of the
optimization using HOMER Similarly pair-wise matrices used to
obtain this weight matrix are presented in this paper in Appendix B
The global weight vector is then given in Table 8
Finally global weights of the three options are obtained as the
product of the local weights matrix and the global weight vector
The result is given by
Option 1 02834
Option 2 03056
Option 3 04110
In this case a re-arrangement of 1047297nal solutions is obtained after
completion of the AHP method The 1047297nal problem solution
selection is given by option 3 where as in the previous case a
higher participation of a continuous energy source has an in1047298u-
ence in the 1047297nal result The third option gives the best balance in
terms of net energy as the other two options are primarily
composed by solar panels A 1047297nal critical parameter was the
capacity shortage which turned out to be relatively high for the
1047297rst two options with unmet supplied energy of 304 kWhyearand 173 kWhyear respectively Time response simulation results
for a complete week of operation are presented in Fig 10 A main
characteristic is the high solar power production with the exceed-
ing generated energy used to keep desired levels of battery stack
SOC A good energy balance is obtained with this option with
virtually no capacity shortage during a year round operation
33 Intermediate diesel prices scenarios
Two additional scenarios were considered with intermediate
diesel price values of 05 and 12$L Similar considerations to the
previous cases were used for pair-wise matrix comparisons
For a diesel price of 05$L the results of the HOMER optimiza-
tion are very similar to the low diesel price case All the possiblesolutions in the group of most interesting optimization results
include a 500 W diesel generator This fact re1047298ects that the diesel
prices are still very low and do not have a big in1047298uence in the
project 1047297nal NPC values After applying the described methodol-
ogy the selected optimal system is given by a mix of solar (250 W)
wind (800 W) and diesel generation (500 W)
For a diesel price of 12$L the optimization yields three main
options to be considered
1 The 1047297rst option is given by the installation of PV (2500 W) and
wind (400 W)
2 The second option consists of a mix of PV (1500 W) wind
(1200 W) and diesel (500 W)
3 The third option is given by a purely PV(3000 W) system
Table 7
Local weights matrix obtained for the high diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 018 026 023 030 030 034 035 034 038 047 043 026 043 033 033 033 033
Op2 032 064 012 016 016 036 033 033 038 047 043 064 014 033 053 033 033
Op3 050 010 065 054 054 030 032 033 023 005 014 010 043 033 014 033 033
Table 8Global weight vector obtained for the high diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
0 20 40 60 80 100 120 1400
05
1
15
2
0 20 40 60 80 100 120 14060
70
80
90
100
Fig 10 Time simulation results hybrid system operating during a week for the
second case study problem
02 04 06 08 1 12 14 16200
400
600
800
1000
1200
1400
1600
1800
Fig 11 PV and wind systems sizes vs diesel prices for each scenario
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110108
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1011
The 1047297nal weight vector gives a 02835 weight for the 1047297rst
04093 for the second and 03072 for the third option The selected
solution is the second option the mix with high PV and wind
participation The storage device obtained for all optimal solutions
is a string of 6 batteries (456 kWh) with a 1 kW power inverter
Fig 11 shows the evolution of PV and wind installation sizes for
the selected solution as a function of the diesel prices considered
in the different scenarios presented in this paperIt can be observed that for the different considerations used in
the pair-wise matrix comparisons at all levels obtained results for
higher diesel prices tend to show an overall increased PV penetra-
tion ratio while for the low diesel price scenarios the selection of
wind power is privileged over PV More importantly the results
show the importance of considering social and environmental
aspects This is shown in the case of a high penetration of PV or
wind when the 1047297nal selected solution is not necessarily the result
of the technical and economical optimization
4 Conclusion
In this paper a hierarchical planning methodology for the
integral net energy design of small-scale hybrid renewable energysystems using ARE sources is proposed The approach considers
technicalndasheconomical as well as societalndashenvironmental aspects in
two different decision layers One of the main features of the
methodology is to complement the classical methodology that
considers the optimization and levelized cost analysis by using the
EROEI or net energy concept and the AHP method for decision
making problems to ensure the consideration of complex and
unconventional aspects such as social and environmental para-
meters in the design of the hybrid ARE systems The methodology
is evaluated using an application example in Margarita Island
under several simulation scenarios low intermediate and high
diesel fuel price cases The obtained results show that the aspects
considered in the proposed methodology have an in1047298uence on the
initial results obtained by using the classical approach of levelizedcost optimization
Acknowledgements
The authors would like to acknowledge the 1047297nancial and
technical support of the Venezuelan Planning School (Escuela
Venezolana de Plani1047297cacioacuten)
Appendix A Pair-wise matrices for low diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 13 01396
Op2 3 1 12 03325Op3 3 2 1 05278
A3 Op1 Op2 Op3 Weights
Op1 1 3 2 05499
Op2 13 1 1 02098
Op3 12 1 1 02402
B1 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
B2 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
D3 Op1 Op2 Op3 Weights
Op1 1 13 15 01047
Op2 3 1 13 02583Op3 5 3 1 06370
E1 Op1 Op2 Op3 Weights
Op1 1 3 3 06000
Op2 13 1 1 02000
Op3 13 1 1 02000
E3 Op1 Op2 Op3 Weights
Op1 1 1 13 02000
Op2 1 1 13 02000Op3 3 3 1 06000
Matrices D2 E2 F1 and F2 are 3 3 all-ones matrices
Appendix B Pair-wise matrices for high diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 3 02583Op2 3 1 5 06370
Op3 13 15 1 01047
A3 Op1 Op2 Op3 Weights
Op1 1 2 13 02297
Op2 12 1 15 01220
Op3 3 5 1 06483
B1 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634
Op3 2 3 1 05396
B2 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634Op3 2 3 1 05396
D1 Op1 Op2 Op3 Weights
Op1 1 1 9 04737
Op2 1 1 9 04737
Op3 19 19 1 00526
D2 Op1 Op2 Op3 Weights
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 109
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1111
Op1 1 1 3 04286Op2 1 1 3 04286
Op3 13 13 1 01429
D3 Op1 Op2 Op3 Weights
Op1 1 13 3 02583
Op2 3 1 5 06370
Op3 13 15 1 01047
E1 Op1 Op2 Op3 Weights
Op1 1 3 1 04286
Op2 13 1 13 01429Op3 1 3 1 04286
E3 Op1 Op2 Op3 Weights
Op1 1 12 3 03325
Op2 2 1 3 05278
Op3 13 13 1 01396
Matrices E2 F1 and F2 are 3 3 all-ones matrices
References
[1] International Energy Agency and Organisation for Economic Co-operation andDevelopment Key world energy statistics 2014
[2] Howes T The EUs New Renewable Energy Directive (200928EC) The newclimate policies of the European Union internal legislation and climatediplomacy vol 5 2010 p 117
[3] Mudacumura GM Mebratu D Haque MS Sustainable development policy andadministration Public administration and public policy Boca Raton CRCTaylor and Francis 2006
[4] Gonzaacutelez-Longatt F Gonzaacutelez JS Payoacuten MB Santos JMR Wind-resource atlasof Venezuela based on on-site anemometry observation Renew SustainEnergy Rev 201439898ndash911
[5] Martenson C The crash course the unsustainable future of our economyenergy and environment John Wiley amp Sons Inc Hoboken NJ 2011
[6] Connolly D Lund H Mathiesen BV Leahy M A review of computer tools foranalysing the integration of renewable energy into various energy systemsAppl Energy 201087(4)1059ndash82
[7] Leng GJ Renewable energy technologies project assessment tool RETScreenVarennes Quebec CANMET Energy Diversi1047297cation Research Laboratory 1998
[8] Lilienthal P HOMER micropower optimization model Report National Renew-able Energy Laboratory (NREL) Golden CO 2005
[9] Saaty TL Fundamentals of decision making and priority theory with theanalytic hierarchy processmdashvol VI RWS Publications Pittsburgh PA 1994
[10] Pohekar SD Ramachandran M Application of multi-criteria decision makingto sustainable energy planningmdasha review Renew Sustain Energy Rev 20048(4)365ndash81
[11] Banos R Manzano-Agugliaro F Montoya FG Gil C Alcayde A Goacutemez JOptimization methods applied to renewable and sustainable energy a reviewRenew Sustain Energy Rev 201115(4)1753ndash66
[12] Taha RA Daim T Multi-criteria applications in renewable energy analysis aliterature review London Springer 2013 p 17ndash30
[13] Elkarmi F Mustafa I Increasing the utilization of solar-energy technologies
(set) in Jordanmdashanalytic hierarchy process Energy Policy 199321(9)978ndash82[14] Akash BA Mamlook R Mohsen MS Multi-criteria selection of electric power
plants using analytical hierarchy process Electr Power Syst Res 199952(1)29ndash35
[15] Wu ZX Wei ZH Mitigation assessment results and priorities for Chinasenergy sector Appl Energy 199756(3ndash4)237ndash51
[16] Wang YM Chin KS Fuzzy analytic hierarchy process a logarithmic fuzzypreference programming methodology Int J Approx Reason 201152(4)541ndash53
[17] Nigim K Munier N Green J Pre-feasibility MCDM tools to aid communities inprioritizing local viable renewable energy sources Renew Energy 200429(11)1775ndash91
[18] Daniel J Vishal NV Albert B Selvarsan I Evaluation of the signi 1047297cant renew-able energy resources in India using analytical hierarchy process LondonSpringer 2010 p 13ndash26
[19] Ahmad S Tahar RM Selection of renewable energy sources for sustainabledevelopment of electricity generation system using analytic hierarchy processa case of Malaysia Renew Energy 201463458ndash66
[20] Tasri A Susilawati A Selection among renewable energy alternatives based ona fuzzy analytic hierarchy process in Indonesia Sustain Energy Technol Assess2014734ndash44
[21] Kaya T Kahraman C Multicriteria renewable energy planning using anintegrated fuzzy VIKOR amp AHP methodology the case of Istanbul Energy2010352517ndash27
[22] Zakerinia M Ghaderi F Piltan M Optimal portfolio selection between differentkinds of renewable energy sources 2010
[23] Barin A Canha LN Abaide A Magnago K Wottrich B Multicriteria analysis of the operation of renewable energy sources taking as basis the AHP methodand fuzzy logic concerning distributed generation systems Online J ElectronElectric Eng (OJEEE) 20091(1)52ndash7
[24] Heo E Kim J Boo KJ Analysis of the assessment factors for renewable energydissemination program evaluation using fuzzy AHP Renew Sustain EnergyRev 201014(8)2214ndash20
[25] Akhgari P Kamalan H Monavari M Utilizing AHP method in rating of highlighted environment parameters in most used renewable energies inelectricity production Eur J Sci Res 201160182ndash8
[26] Murphy DJ Hall CAS Year in reviewmdashEROI or energy return on (energy)
invested Ecol Econ Rev 20101185102ndash18[27] Urdaneta AJ Chankong V A multiobjective minimax approach to controller
settings for systems running under disturbances Control-Theory Adv Technol19895(4)391ndash411
[28] Gonzaacutelez-Longatt F Teraacuten R Meacutendez J Hernaacutendez A Guilleacuten F Evaluacioacuten delRecurso Eoacutelico en Venezuela Parte I Congreso Petrolero Energeacutetico ASME-UNEFA Puerto Cabello 2006 [in Spanish]
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110110
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 811
With the results obtained after evaluation of each alternative
option 1 stands as still the best solution according to the evaluated
criteria An important aspect that may have an in1047298uence in this
result is the choice of option 1 as a hybrid energy source with a
high component of continuous energy capacity (the diesel gen-
erator represents the 74 of the energy produced) Using HOMER
a time response simulation is obtained for the selected solution
The simulation result is presented in Fig 8 and is representative of
the operation of the hybrid generator during a complete week The1047297gure shows the levels of energy production for each source the
electrical load demand and the batteries state-of-charge (SOC)
The results show the need of the diesel generator when the
batteries SOC falls beyond 50 during low wind energy
production
32 High diesel price scenario
The second case study scenario is given by the same example as
before but now considering international diesel fuel prices The
price taken as reference is at 161$L (average value in Europe
during July 2012) Using a similar model as in the previous case
the levelized cost optimization using HOMER is performed Results
are shown in Table 6 where the system characteristics arepresented The operating electricity production characteristic of
each con1047297guration is presented in Fig 9 In this case power
generation from solar panels dominates the 1047297rst and second
0 20 40 60 80 100 120 1400
05
1
15
0 20 40 60 80 100 120 14040
60
80
100
Fig 8 Time simulation results hybrid system operating during a week
Table 6
Results obtained with HOMER for the high diesel price case study
Structure NPC ($) OampM costs ($year) LCOE ($kWh)
Option 1 25 896 383 0590
Option 2 28 408 384 0629
Option 3 29 469 878 0632
Fig 9 Simulation results (electricity production) of hybrid sources for the three options considered in the second case study problem
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 107
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 911
options The third system option is composed by a mixed solar
winddiesel power generator In this second case study option 1 is
composed by 25 kW in solar panels 1 SW AIR X 400 W wind
turbine 1 string of 6 batteries and a 1 kW power inverter Option
2 is a solution with 3 kW in solar panels 1 string of 6 batteries and
a 1 kW power inverter Option 3 is given with 175 kW in solar
panels 3 SW AIR X 400 W wind turbines 500 W diesel generator
1 string of 6 batteries and a 1 kW power inverter All batteries are
model S4KS25P 76 kWh (4 V 1900 Ah)
With a similar analysis as in the previous case the local
weights matrix is given in Table 7 Here light gray cells denote
the weights that were 1047297xed directly from results of the
optimization using HOMER Similarly pair-wise matrices used to
obtain this weight matrix are presented in this paper in Appendix B
The global weight vector is then given in Table 8
Finally global weights of the three options are obtained as the
product of the local weights matrix and the global weight vector
The result is given by
Option 1 02834
Option 2 03056
Option 3 04110
In this case a re-arrangement of 1047297nal solutions is obtained after
completion of the AHP method The 1047297nal problem solution
selection is given by option 3 where as in the previous case a
higher participation of a continuous energy source has an in1047298u-
ence in the 1047297nal result The third option gives the best balance in
terms of net energy as the other two options are primarily
composed by solar panels A 1047297nal critical parameter was the
capacity shortage which turned out to be relatively high for the
1047297rst two options with unmet supplied energy of 304 kWhyearand 173 kWhyear respectively Time response simulation results
for a complete week of operation are presented in Fig 10 A main
characteristic is the high solar power production with the exceed-
ing generated energy used to keep desired levels of battery stack
SOC A good energy balance is obtained with this option with
virtually no capacity shortage during a year round operation
33 Intermediate diesel prices scenarios
Two additional scenarios were considered with intermediate
diesel price values of 05 and 12$L Similar considerations to the
previous cases were used for pair-wise matrix comparisons
For a diesel price of 05$L the results of the HOMER optimiza-
tion are very similar to the low diesel price case All the possiblesolutions in the group of most interesting optimization results
include a 500 W diesel generator This fact re1047298ects that the diesel
prices are still very low and do not have a big in1047298uence in the
project 1047297nal NPC values After applying the described methodol-
ogy the selected optimal system is given by a mix of solar (250 W)
wind (800 W) and diesel generation (500 W)
For a diesel price of 12$L the optimization yields three main
options to be considered
1 The 1047297rst option is given by the installation of PV (2500 W) and
wind (400 W)
2 The second option consists of a mix of PV (1500 W) wind
(1200 W) and diesel (500 W)
3 The third option is given by a purely PV(3000 W) system
Table 7
Local weights matrix obtained for the high diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 018 026 023 030 030 034 035 034 038 047 043 026 043 033 033 033 033
Op2 032 064 012 016 016 036 033 033 038 047 043 064 014 033 053 033 033
Op3 050 010 065 054 054 030 032 033 023 005 014 010 043 033 014 033 033
Table 8Global weight vector obtained for the high diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
0 20 40 60 80 100 120 1400
05
1
15
2
0 20 40 60 80 100 120 14060
70
80
90
100
Fig 10 Time simulation results hybrid system operating during a week for the
second case study problem
02 04 06 08 1 12 14 16200
400
600
800
1000
1200
1400
1600
1800
Fig 11 PV and wind systems sizes vs diesel prices for each scenario
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110108
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1011
The 1047297nal weight vector gives a 02835 weight for the 1047297rst
04093 for the second and 03072 for the third option The selected
solution is the second option the mix with high PV and wind
participation The storage device obtained for all optimal solutions
is a string of 6 batteries (456 kWh) with a 1 kW power inverter
Fig 11 shows the evolution of PV and wind installation sizes for
the selected solution as a function of the diesel prices considered
in the different scenarios presented in this paperIt can be observed that for the different considerations used in
the pair-wise matrix comparisons at all levels obtained results for
higher diesel prices tend to show an overall increased PV penetra-
tion ratio while for the low diesel price scenarios the selection of
wind power is privileged over PV More importantly the results
show the importance of considering social and environmental
aspects This is shown in the case of a high penetration of PV or
wind when the 1047297nal selected solution is not necessarily the result
of the technical and economical optimization
4 Conclusion
In this paper a hierarchical planning methodology for the
integral net energy design of small-scale hybrid renewable energysystems using ARE sources is proposed The approach considers
technicalndasheconomical as well as societalndashenvironmental aspects in
two different decision layers One of the main features of the
methodology is to complement the classical methodology that
considers the optimization and levelized cost analysis by using the
EROEI or net energy concept and the AHP method for decision
making problems to ensure the consideration of complex and
unconventional aspects such as social and environmental para-
meters in the design of the hybrid ARE systems The methodology
is evaluated using an application example in Margarita Island
under several simulation scenarios low intermediate and high
diesel fuel price cases The obtained results show that the aspects
considered in the proposed methodology have an in1047298uence on the
initial results obtained by using the classical approach of levelizedcost optimization
Acknowledgements
The authors would like to acknowledge the 1047297nancial and
technical support of the Venezuelan Planning School (Escuela
Venezolana de Plani1047297cacioacuten)
Appendix A Pair-wise matrices for low diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 13 01396
Op2 3 1 12 03325Op3 3 2 1 05278
A3 Op1 Op2 Op3 Weights
Op1 1 3 2 05499
Op2 13 1 1 02098
Op3 12 1 1 02402
B1 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
B2 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
D3 Op1 Op2 Op3 Weights
Op1 1 13 15 01047
Op2 3 1 13 02583Op3 5 3 1 06370
E1 Op1 Op2 Op3 Weights
Op1 1 3 3 06000
Op2 13 1 1 02000
Op3 13 1 1 02000
E3 Op1 Op2 Op3 Weights
Op1 1 1 13 02000
Op2 1 1 13 02000Op3 3 3 1 06000
Matrices D2 E2 F1 and F2 are 3 3 all-ones matrices
Appendix B Pair-wise matrices for high diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 3 02583Op2 3 1 5 06370
Op3 13 15 1 01047
A3 Op1 Op2 Op3 Weights
Op1 1 2 13 02297
Op2 12 1 15 01220
Op3 3 5 1 06483
B1 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634
Op3 2 3 1 05396
B2 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634Op3 2 3 1 05396
D1 Op1 Op2 Op3 Weights
Op1 1 1 9 04737
Op2 1 1 9 04737
Op3 19 19 1 00526
D2 Op1 Op2 Op3 Weights
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 109
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1111
Op1 1 1 3 04286Op2 1 1 3 04286
Op3 13 13 1 01429
D3 Op1 Op2 Op3 Weights
Op1 1 13 3 02583
Op2 3 1 5 06370
Op3 13 15 1 01047
E1 Op1 Op2 Op3 Weights
Op1 1 3 1 04286
Op2 13 1 13 01429Op3 1 3 1 04286
E3 Op1 Op2 Op3 Weights
Op1 1 12 3 03325
Op2 2 1 3 05278
Op3 13 13 1 01396
Matrices E2 F1 and F2 are 3 3 all-ones matrices
References
[1] International Energy Agency and Organisation for Economic Co-operation andDevelopment Key world energy statistics 2014
[2] Howes T The EUs New Renewable Energy Directive (200928EC) The newclimate policies of the European Union internal legislation and climatediplomacy vol 5 2010 p 117
[3] Mudacumura GM Mebratu D Haque MS Sustainable development policy andadministration Public administration and public policy Boca Raton CRCTaylor and Francis 2006
[4] Gonzaacutelez-Longatt F Gonzaacutelez JS Payoacuten MB Santos JMR Wind-resource atlasof Venezuela based on on-site anemometry observation Renew SustainEnergy Rev 201439898ndash911
[5] Martenson C The crash course the unsustainable future of our economyenergy and environment John Wiley amp Sons Inc Hoboken NJ 2011
[6] Connolly D Lund H Mathiesen BV Leahy M A review of computer tools foranalysing the integration of renewable energy into various energy systemsAppl Energy 201087(4)1059ndash82
[7] Leng GJ Renewable energy technologies project assessment tool RETScreenVarennes Quebec CANMET Energy Diversi1047297cation Research Laboratory 1998
[8] Lilienthal P HOMER micropower optimization model Report National Renew-able Energy Laboratory (NREL) Golden CO 2005
[9] Saaty TL Fundamentals of decision making and priority theory with theanalytic hierarchy processmdashvol VI RWS Publications Pittsburgh PA 1994
[10] Pohekar SD Ramachandran M Application of multi-criteria decision makingto sustainable energy planningmdasha review Renew Sustain Energy Rev 20048(4)365ndash81
[11] Banos R Manzano-Agugliaro F Montoya FG Gil C Alcayde A Goacutemez JOptimization methods applied to renewable and sustainable energy a reviewRenew Sustain Energy Rev 201115(4)1753ndash66
[12] Taha RA Daim T Multi-criteria applications in renewable energy analysis aliterature review London Springer 2013 p 17ndash30
[13] Elkarmi F Mustafa I Increasing the utilization of solar-energy technologies
(set) in Jordanmdashanalytic hierarchy process Energy Policy 199321(9)978ndash82[14] Akash BA Mamlook R Mohsen MS Multi-criteria selection of electric power
plants using analytical hierarchy process Electr Power Syst Res 199952(1)29ndash35
[15] Wu ZX Wei ZH Mitigation assessment results and priorities for Chinasenergy sector Appl Energy 199756(3ndash4)237ndash51
[16] Wang YM Chin KS Fuzzy analytic hierarchy process a logarithmic fuzzypreference programming methodology Int J Approx Reason 201152(4)541ndash53
[17] Nigim K Munier N Green J Pre-feasibility MCDM tools to aid communities inprioritizing local viable renewable energy sources Renew Energy 200429(11)1775ndash91
[18] Daniel J Vishal NV Albert B Selvarsan I Evaluation of the signi 1047297cant renew-able energy resources in India using analytical hierarchy process LondonSpringer 2010 p 13ndash26
[19] Ahmad S Tahar RM Selection of renewable energy sources for sustainabledevelopment of electricity generation system using analytic hierarchy processa case of Malaysia Renew Energy 201463458ndash66
[20] Tasri A Susilawati A Selection among renewable energy alternatives based ona fuzzy analytic hierarchy process in Indonesia Sustain Energy Technol Assess2014734ndash44
[21] Kaya T Kahraman C Multicriteria renewable energy planning using anintegrated fuzzy VIKOR amp AHP methodology the case of Istanbul Energy2010352517ndash27
[22] Zakerinia M Ghaderi F Piltan M Optimal portfolio selection between differentkinds of renewable energy sources 2010
[23] Barin A Canha LN Abaide A Magnago K Wottrich B Multicriteria analysis of the operation of renewable energy sources taking as basis the AHP methodand fuzzy logic concerning distributed generation systems Online J ElectronElectric Eng (OJEEE) 20091(1)52ndash7
[24] Heo E Kim J Boo KJ Analysis of the assessment factors for renewable energydissemination program evaluation using fuzzy AHP Renew Sustain EnergyRev 201014(8)2214ndash20
[25] Akhgari P Kamalan H Monavari M Utilizing AHP method in rating of highlighted environment parameters in most used renewable energies inelectricity production Eur J Sci Res 201160182ndash8
[26] Murphy DJ Hall CAS Year in reviewmdashEROI or energy return on (energy)
invested Ecol Econ Rev 20101185102ndash18[27] Urdaneta AJ Chankong V A multiobjective minimax approach to controller
settings for systems running under disturbances Control-Theory Adv Technol19895(4)391ndash411
[28] Gonzaacutelez-Longatt F Teraacuten R Meacutendez J Hernaacutendez A Guilleacuten F Evaluacioacuten delRecurso Eoacutelico en Venezuela Parte I Congreso Petrolero Energeacutetico ASME-UNEFA Puerto Cabello 2006 [in Spanish]
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110110
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 911
options The third system option is composed by a mixed solar
winddiesel power generator In this second case study option 1 is
composed by 25 kW in solar panels 1 SW AIR X 400 W wind
turbine 1 string of 6 batteries and a 1 kW power inverter Option
2 is a solution with 3 kW in solar panels 1 string of 6 batteries and
a 1 kW power inverter Option 3 is given with 175 kW in solar
panels 3 SW AIR X 400 W wind turbines 500 W diesel generator
1 string of 6 batteries and a 1 kW power inverter All batteries are
model S4KS25P 76 kWh (4 V 1900 Ah)
With a similar analysis as in the previous case the local
weights matrix is given in Table 7 Here light gray cells denote
the weights that were 1047297xed directly from results of the
optimization using HOMER Similarly pair-wise matrices used to
obtain this weight matrix are presented in this paper in Appendix B
The global weight vector is then given in Table 8
Finally global weights of the three options are obtained as the
product of the local weights matrix and the global weight vector
The result is given by
Option 1 02834
Option 2 03056
Option 3 04110
In this case a re-arrangement of 1047297nal solutions is obtained after
completion of the AHP method The 1047297nal problem solution
selection is given by option 3 where as in the previous case a
higher participation of a continuous energy source has an in1047298u-
ence in the 1047297nal result The third option gives the best balance in
terms of net energy as the other two options are primarily
composed by solar panels A 1047297nal critical parameter was the
capacity shortage which turned out to be relatively high for the
1047297rst two options with unmet supplied energy of 304 kWhyearand 173 kWhyear respectively Time response simulation results
for a complete week of operation are presented in Fig 10 A main
characteristic is the high solar power production with the exceed-
ing generated energy used to keep desired levels of battery stack
SOC A good energy balance is obtained with this option with
virtually no capacity shortage during a year round operation
33 Intermediate diesel prices scenarios
Two additional scenarios were considered with intermediate
diesel price values of 05 and 12$L Similar considerations to the
previous cases were used for pair-wise matrix comparisons
For a diesel price of 05$L the results of the HOMER optimiza-
tion are very similar to the low diesel price case All the possiblesolutions in the group of most interesting optimization results
include a 500 W diesel generator This fact re1047298ects that the diesel
prices are still very low and do not have a big in1047298uence in the
project 1047297nal NPC values After applying the described methodol-
ogy the selected optimal system is given by a mix of solar (250 W)
wind (800 W) and diesel generation (500 W)
For a diesel price of 12$L the optimization yields three main
options to be considered
1 The 1047297rst option is given by the installation of PV (2500 W) and
wind (400 W)
2 The second option consists of a mix of PV (1500 W) wind
(1200 W) and diesel (500 W)
3 The third option is given by a purely PV(3000 W) system
Table 7
Local weights matrix obtained for the high diesel price case study
Option A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
Op1 018 026 023 030 030 034 035 034 038 047 043 026 043 033 033 033 033
Op2 032 064 012 016 016 036 033 033 038 047 043 064 014 033 053 033 033
Op3 050 010 065 054 054 030 032 033 023 005 014 010 043 033 014 033 033
Table 8Global weight vector obtained for the high diesel price case study
A1 A2 A3 B1 B2 B3 C1 C2 C3 D1 D2 D3 E1 E2 E3 F1 F2
023 004 021 001 003 015 003 003 001 005 001 001 001 000 002 003 013
0 20 40 60 80 100 120 1400
05
1
15
2
0 20 40 60 80 100 120 14060
70
80
90
100
Fig 10 Time simulation results hybrid system operating during a week for the
second case study problem
02 04 06 08 1 12 14 16200
400
600
800
1000
1200
1400
1600
1800
Fig 11 PV and wind systems sizes vs diesel prices for each scenario
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110108
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1011
The 1047297nal weight vector gives a 02835 weight for the 1047297rst
04093 for the second and 03072 for the third option The selected
solution is the second option the mix with high PV and wind
participation The storage device obtained for all optimal solutions
is a string of 6 batteries (456 kWh) with a 1 kW power inverter
Fig 11 shows the evolution of PV and wind installation sizes for
the selected solution as a function of the diesel prices considered
in the different scenarios presented in this paperIt can be observed that for the different considerations used in
the pair-wise matrix comparisons at all levels obtained results for
higher diesel prices tend to show an overall increased PV penetra-
tion ratio while for the low diesel price scenarios the selection of
wind power is privileged over PV More importantly the results
show the importance of considering social and environmental
aspects This is shown in the case of a high penetration of PV or
wind when the 1047297nal selected solution is not necessarily the result
of the technical and economical optimization
4 Conclusion
In this paper a hierarchical planning methodology for the
integral net energy design of small-scale hybrid renewable energysystems using ARE sources is proposed The approach considers
technicalndasheconomical as well as societalndashenvironmental aspects in
two different decision layers One of the main features of the
methodology is to complement the classical methodology that
considers the optimization and levelized cost analysis by using the
EROEI or net energy concept and the AHP method for decision
making problems to ensure the consideration of complex and
unconventional aspects such as social and environmental para-
meters in the design of the hybrid ARE systems The methodology
is evaluated using an application example in Margarita Island
under several simulation scenarios low intermediate and high
diesel fuel price cases The obtained results show that the aspects
considered in the proposed methodology have an in1047298uence on the
initial results obtained by using the classical approach of levelizedcost optimization
Acknowledgements
The authors would like to acknowledge the 1047297nancial and
technical support of the Venezuelan Planning School (Escuela
Venezolana de Plani1047297cacioacuten)
Appendix A Pair-wise matrices for low diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 13 01396
Op2 3 1 12 03325Op3 3 2 1 05278
A3 Op1 Op2 Op3 Weights
Op1 1 3 2 05499
Op2 13 1 1 02098
Op3 12 1 1 02402
B1 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
B2 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
D3 Op1 Op2 Op3 Weights
Op1 1 13 15 01047
Op2 3 1 13 02583Op3 5 3 1 06370
E1 Op1 Op2 Op3 Weights
Op1 1 3 3 06000
Op2 13 1 1 02000
Op3 13 1 1 02000
E3 Op1 Op2 Op3 Weights
Op1 1 1 13 02000
Op2 1 1 13 02000Op3 3 3 1 06000
Matrices D2 E2 F1 and F2 are 3 3 all-ones matrices
Appendix B Pair-wise matrices for high diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 3 02583Op2 3 1 5 06370
Op3 13 15 1 01047
A3 Op1 Op2 Op3 Weights
Op1 1 2 13 02297
Op2 12 1 15 01220
Op3 3 5 1 06483
B1 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634
Op3 2 3 1 05396
B2 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634Op3 2 3 1 05396
D1 Op1 Op2 Op3 Weights
Op1 1 1 9 04737
Op2 1 1 9 04737
Op3 19 19 1 00526
D2 Op1 Op2 Op3 Weights
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 109
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1111
Op1 1 1 3 04286Op2 1 1 3 04286
Op3 13 13 1 01429
D3 Op1 Op2 Op3 Weights
Op1 1 13 3 02583
Op2 3 1 5 06370
Op3 13 15 1 01047
E1 Op1 Op2 Op3 Weights
Op1 1 3 1 04286
Op2 13 1 13 01429Op3 1 3 1 04286
E3 Op1 Op2 Op3 Weights
Op1 1 12 3 03325
Op2 2 1 3 05278
Op3 13 13 1 01396
Matrices E2 F1 and F2 are 3 3 all-ones matrices
References
[1] International Energy Agency and Organisation for Economic Co-operation andDevelopment Key world energy statistics 2014
[2] Howes T The EUs New Renewable Energy Directive (200928EC) The newclimate policies of the European Union internal legislation and climatediplomacy vol 5 2010 p 117
[3] Mudacumura GM Mebratu D Haque MS Sustainable development policy andadministration Public administration and public policy Boca Raton CRCTaylor and Francis 2006
[4] Gonzaacutelez-Longatt F Gonzaacutelez JS Payoacuten MB Santos JMR Wind-resource atlasof Venezuela based on on-site anemometry observation Renew SustainEnergy Rev 201439898ndash911
[5] Martenson C The crash course the unsustainable future of our economyenergy and environment John Wiley amp Sons Inc Hoboken NJ 2011
[6] Connolly D Lund H Mathiesen BV Leahy M A review of computer tools foranalysing the integration of renewable energy into various energy systemsAppl Energy 201087(4)1059ndash82
[7] Leng GJ Renewable energy technologies project assessment tool RETScreenVarennes Quebec CANMET Energy Diversi1047297cation Research Laboratory 1998
[8] Lilienthal P HOMER micropower optimization model Report National Renew-able Energy Laboratory (NREL) Golden CO 2005
[9] Saaty TL Fundamentals of decision making and priority theory with theanalytic hierarchy processmdashvol VI RWS Publications Pittsburgh PA 1994
[10] Pohekar SD Ramachandran M Application of multi-criteria decision makingto sustainable energy planningmdasha review Renew Sustain Energy Rev 20048(4)365ndash81
[11] Banos R Manzano-Agugliaro F Montoya FG Gil C Alcayde A Goacutemez JOptimization methods applied to renewable and sustainable energy a reviewRenew Sustain Energy Rev 201115(4)1753ndash66
[12] Taha RA Daim T Multi-criteria applications in renewable energy analysis aliterature review London Springer 2013 p 17ndash30
[13] Elkarmi F Mustafa I Increasing the utilization of solar-energy technologies
(set) in Jordanmdashanalytic hierarchy process Energy Policy 199321(9)978ndash82[14] Akash BA Mamlook R Mohsen MS Multi-criteria selection of electric power
plants using analytical hierarchy process Electr Power Syst Res 199952(1)29ndash35
[15] Wu ZX Wei ZH Mitigation assessment results and priorities for Chinasenergy sector Appl Energy 199756(3ndash4)237ndash51
[16] Wang YM Chin KS Fuzzy analytic hierarchy process a logarithmic fuzzypreference programming methodology Int J Approx Reason 201152(4)541ndash53
[17] Nigim K Munier N Green J Pre-feasibility MCDM tools to aid communities inprioritizing local viable renewable energy sources Renew Energy 200429(11)1775ndash91
[18] Daniel J Vishal NV Albert B Selvarsan I Evaluation of the signi 1047297cant renew-able energy resources in India using analytical hierarchy process LondonSpringer 2010 p 13ndash26
[19] Ahmad S Tahar RM Selection of renewable energy sources for sustainabledevelopment of electricity generation system using analytic hierarchy processa case of Malaysia Renew Energy 201463458ndash66
[20] Tasri A Susilawati A Selection among renewable energy alternatives based ona fuzzy analytic hierarchy process in Indonesia Sustain Energy Technol Assess2014734ndash44
[21] Kaya T Kahraman C Multicriteria renewable energy planning using anintegrated fuzzy VIKOR amp AHP methodology the case of Istanbul Energy2010352517ndash27
[22] Zakerinia M Ghaderi F Piltan M Optimal portfolio selection between differentkinds of renewable energy sources 2010
[23] Barin A Canha LN Abaide A Magnago K Wottrich B Multicriteria analysis of the operation of renewable energy sources taking as basis the AHP methodand fuzzy logic concerning distributed generation systems Online J ElectronElectric Eng (OJEEE) 20091(1)52ndash7
[24] Heo E Kim J Boo KJ Analysis of the assessment factors for renewable energydissemination program evaluation using fuzzy AHP Renew Sustain EnergyRev 201014(8)2214ndash20
[25] Akhgari P Kamalan H Monavari M Utilizing AHP method in rating of highlighted environment parameters in most used renewable energies inelectricity production Eur J Sci Res 201160182ndash8
[26] Murphy DJ Hall CAS Year in reviewmdashEROI or energy return on (energy)
invested Ecol Econ Rev 20101185102ndash18[27] Urdaneta AJ Chankong V A multiobjective minimax approach to controller
settings for systems running under disturbances Control-Theory Adv Technol19895(4)391ndash411
[28] Gonzaacutelez-Longatt F Teraacuten R Meacutendez J Hernaacutendez A Guilleacuten F Evaluacioacuten delRecurso Eoacutelico en Venezuela Parte I Congreso Petrolero Energeacutetico ASME-UNEFA Puerto Cabello 2006 [in Spanish]
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110110
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1011
The 1047297nal weight vector gives a 02835 weight for the 1047297rst
04093 for the second and 03072 for the third option The selected
solution is the second option the mix with high PV and wind
participation The storage device obtained for all optimal solutions
is a string of 6 batteries (456 kWh) with a 1 kW power inverter
Fig 11 shows the evolution of PV and wind installation sizes for
the selected solution as a function of the diesel prices considered
in the different scenarios presented in this paperIt can be observed that for the different considerations used in
the pair-wise matrix comparisons at all levels obtained results for
higher diesel prices tend to show an overall increased PV penetra-
tion ratio while for the low diesel price scenarios the selection of
wind power is privileged over PV More importantly the results
show the importance of considering social and environmental
aspects This is shown in the case of a high penetration of PV or
wind when the 1047297nal selected solution is not necessarily the result
of the technical and economical optimization
4 Conclusion
In this paper a hierarchical planning methodology for the
integral net energy design of small-scale hybrid renewable energysystems using ARE sources is proposed The approach considers
technicalndasheconomical as well as societalndashenvironmental aspects in
two different decision layers One of the main features of the
methodology is to complement the classical methodology that
considers the optimization and levelized cost analysis by using the
EROEI or net energy concept and the AHP method for decision
making problems to ensure the consideration of complex and
unconventional aspects such as social and environmental para-
meters in the design of the hybrid ARE systems The methodology
is evaluated using an application example in Margarita Island
under several simulation scenarios low intermediate and high
diesel fuel price cases The obtained results show that the aspects
considered in the proposed methodology have an in1047298uence on the
initial results obtained by using the classical approach of levelizedcost optimization
Acknowledgements
The authors would like to acknowledge the 1047297nancial and
technical support of the Venezuelan Planning School (Escuela
Venezolana de Plani1047297cacioacuten)
Appendix A Pair-wise matrices for low diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 13 01396
Op2 3 1 12 03325Op3 3 2 1 05278
A3 Op1 Op2 Op3 Weights
Op1 1 3 2 05499
Op2 13 1 1 02098
Op3 12 1 1 02402
B1 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
B2 Op1 Op2 Op3 Weights
Op1 1 13 2 02493
Op2 3 1 3 05936
Op3 12 13 1 01571
D3 Op1 Op2 Op3 Weights
Op1 1 13 15 01047
Op2 3 1 13 02583Op3 5 3 1 06370
E1 Op1 Op2 Op3 Weights
Op1 1 3 3 06000
Op2 13 1 1 02000
Op3 13 1 1 02000
E3 Op1 Op2 Op3 Weights
Op1 1 1 13 02000
Op2 1 1 13 02000Op3 3 3 1 06000
Matrices D2 E2 F1 and F2 are 3 3 all-ones matrices
Appendix B Pair-wise matrices for high diesel price case study
scenario
A2 Op1 Op2 Op3 Weights
Op1 1 13 3 02583Op2 3 1 5 06370
Op3 13 15 1 01047
A3 Op1 Op2 Op3 Weights
Op1 1 2 13 02297
Op2 12 1 15 01220
Op3 3 5 1 06483
B1 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634
Op3 2 3 1 05396
B2 Op1 Op2 Op3 Weights
Op1 1 2 12 02970
Op2 12 1 13 01634Op3 2 3 1 05396
D1 Op1 Op2 Op3 Weights
Op1 1 1 9 04737
Op2 1 1 9 04737
Op3 19 19 1 00526
D2 Op1 Op2 Op3 Weights
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110 109
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1111
Op1 1 1 3 04286Op2 1 1 3 04286
Op3 13 13 1 01429
D3 Op1 Op2 Op3 Weights
Op1 1 13 3 02583
Op2 3 1 5 06370
Op3 13 15 1 01047
E1 Op1 Op2 Op3 Weights
Op1 1 3 1 04286
Op2 13 1 13 01429Op3 1 3 1 04286
E3 Op1 Op2 Op3 Weights
Op1 1 12 3 03325
Op2 2 1 3 05278
Op3 13 13 1 01396
Matrices E2 F1 and F2 are 3 3 all-ones matrices
References
[1] International Energy Agency and Organisation for Economic Co-operation andDevelopment Key world energy statistics 2014
[2] Howes T The EUs New Renewable Energy Directive (200928EC) The newclimate policies of the European Union internal legislation and climatediplomacy vol 5 2010 p 117
[3] Mudacumura GM Mebratu D Haque MS Sustainable development policy andadministration Public administration and public policy Boca Raton CRCTaylor and Francis 2006
[4] Gonzaacutelez-Longatt F Gonzaacutelez JS Payoacuten MB Santos JMR Wind-resource atlasof Venezuela based on on-site anemometry observation Renew SustainEnergy Rev 201439898ndash911
[5] Martenson C The crash course the unsustainable future of our economyenergy and environment John Wiley amp Sons Inc Hoboken NJ 2011
[6] Connolly D Lund H Mathiesen BV Leahy M A review of computer tools foranalysing the integration of renewable energy into various energy systemsAppl Energy 201087(4)1059ndash82
[7] Leng GJ Renewable energy technologies project assessment tool RETScreenVarennes Quebec CANMET Energy Diversi1047297cation Research Laboratory 1998
[8] Lilienthal P HOMER micropower optimization model Report National Renew-able Energy Laboratory (NREL) Golden CO 2005
[9] Saaty TL Fundamentals of decision making and priority theory with theanalytic hierarchy processmdashvol VI RWS Publications Pittsburgh PA 1994
[10] Pohekar SD Ramachandran M Application of multi-criteria decision makingto sustainable energy planningmdasha review Renew Sustain Energy Rev 20048(4)365ndash81
[11] Banos R Manzano-Agugliaro F Montoya FG Gil C Alcayde A Goacutemez JOptimization methods applied to renewable and sustainable energy a reviewRenew Sustain Energy Rev 201115(4)1753ndash66
[12] Taha RA Daim T Multi-criteria applications in renewable energy analysis aliterature review London Springer 2013 p 17ndash30
[13] Elkarmi F Mustafa I Increasing the utilization of solar-energy technologies
(set) in Jordanmdashanalytic hierarchy process Energy Policy 199321(9)978ndash82[14] Akash BA Mamlook R Mohsen MS Multi-criteria selection of electric power
plants using analytical hierarchy process Electr Power Syst Res 199952(1)29ndash35
[15] Wu ZX Wei ZH Mitigation assessment results and priorities for Chinasenergy sector Appl Energy 199756(3ndash4)237ndash51
[16] Wang YM Chin KS Fuzzy analytic hierarchy process a logarithmic fuzzypreference programming methodology Int J Approx Reason 201152(4)541ndash53
[17] Nigim K Munier N Green J Pre-feasibility MCDM tools to aid communities inprioritizing local viable renewable energy sources Renew Energy 200429(11)1775ndash91
[18] Daniel J Vishal NV Albert B Selvarsan I Evaluation of the signi 1047297cant renew-able energy resources in India using analytical hierarchy process LondonSpringer 2010 p 13ndash26
[19] Ahmad S Tahar RM Selection of renewable energy sources for sustainabledevelopment of electricity generation system using analytic hierarchy processa case of Malaysia Renew Energy 201463458ndash66
[20] Tasri A Susilawati A Selection among renewable energy alternatives based ona fuzzy analytic hierarchy process in Indonesia Sustain Energy Technol Assess2014734ndash44
[21] Kaya T Kahraman C Multicriteria renewable energy planning using anintegrated fuzzy VIKOR amp AHP methodology the case of Istanbul Energy2010352517ndash27
[22] Zakerinia M Ghaderi F Piltan M Optimal portfolio selection between differentkinds of renewable energy sources 2010
[23] Barin A Canha LN Abaide A Magnago K Wottrich B Multicriteria analysis of the operation of renewable energy sources taking as basis the AHP methodand fuzzy logic concerning distributed generation systems Online J ElectronElectric Eng (OJEEE) 20091(1)52ndash7
[24] Heo E Kim J Boo KJ Analysis of the assessment factors for renewable energydissemination program evaluation using fuzzy AHP Renew Sustain EnergyRev 201014(8)2214ndash20
[25] Akhgari P Kamalan H Monavari M Utilizing AHP method in rating of highlighted environment parameters in most used renewable energies inelectricity production Eur J Sci Res 201160182ndash8
[26] Murphy DJ Hall CAS Year in reviewmdashEROI or energy return on (energy)
invested Ecol Econ Rev 20101185102ndash18[27] Urdaneta AJ Chankong V A multiobjective minimax approach to controller
settings for systems running under disturbances Control-Theory Adv Technol19895(4)391ndash411
[28] Gonzaacutelez-Longatt F Teraacuten R Meacutendez J Hernaacutendez A Guilleacuten F Evaluacioacuten delRecurso Eoacutelico en Venezuela Parte I Congreso Petrolero Energeacutetico ASME-UNEFA Puerto Cabello 2006 [in Spanish]
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110110
8152019 2015Renewable and Sustainable Energy Reviews - A Hierarchical Methodologyfor the Integral Net Energy Design(1)
httpslidepdfcomreaderfull2015renewable-and-sustainable-energy-reviews-a-hierarchical-methodologyfor 1111
Op1 1 1 3 04286Op2 1 1 3 04286
Op3 13 13 1 01429
D3 Op1 Op2 Op3 Weights
Op1 1 13 3 02583
Op2 3 1 5 06370
Op3 13 15 1 01047
E1 Op1 Op2 Op3 Weights
Op1 1 3 1 04286
Op2 13 1 13 01429Op3 1 3 1 04286
E3 Op1 Op2 Op3 Weights
Op1 1 12 3 03325
Op2 2 1 3 05278
Op3 13 13 1 01396
Matrices E2 F1 and F2 are 3 3 all-ones matrices
References
[1] International Energy Agency and Organisation for Economic Co-operation andDevelopment Key world energy statistics 2014
[2] Howes T The EUs New Renewable Energy Directive (200928EC) The newclimate policies of the European Union internal legislation and climatediplomacy vol 5 2010 p 117
[3] Mudacumura GM Mebratu D Haque MS Sustainable development policy andadministration Public administration and public policy Boca Raton CRCTaylor and Francis 2006
[4] Gonzaacutelez-Longatt F Gonzaacutelez JS Payoacuten MB Santos JMR Wind-resource atlasof Venezuela based on on-site anemometry observation Renew SustainEnergy Rev 201439898ndash911
[5] Martenson C The crash course the unsustainable future of our economyenergy and environment John Wiley amp Sons Inc Hoboken NJ 2011
[6] Connolly D Lund H Mathiesen BV Leahy M A review of computer tools foranalysing the integration of renewable energy into various energy systemsAppl Energy 201087(4)1059ndash82
[7] Leng GJ Renewable energy technologies project assessment tool RETScreenVarennes Quebec CANMET Energy Diversi1047297cation Research Laboratory 1998
[8] Lilienthal P HOMER micropower optimization model Report National Renew-able Energy Laboratory (NREL) Golden CO 2005
[9] Saaty TL Fundamentals of decision making and priority theory with theanalytic hierarchy processmdashvol VI RWS Publications Pittsburgh PA 1994
[10] Pohekar SD Ramachandran M Application of multi-criteria decision makingto sustainable energy planningmdasha review Renew Sustain Energy Rev 20048(4)365ndash81
[11] Banos R Manzano-Agugliaro F Montoya FG Gil C Alcayde A Goacutemez JOptimization methods applied to renewable and sustainable energy a reviewRenew Sustain Energy Rev 201115(4)1753ndash66
[12] Taha RA Daim T Multi-criteria applications in renewable energy analysis aliterature review London Springer 2013 p 17ndash30
[13] Elkarmi F Mustafa I Increasing the utilization of solar-energy technologies
(set) in Jordanmdashanalytic hierarchy process Energy Policy 199321(9)978ndash82[14] Akash BA Mamlook R Mohsen MS Multi-criteria selection of electric power
plants using analytical hierarchy process Electr Power Syst Res 199952(1)29ndash35
[15] Wu ZX Wei ZH Mitigation assessment results and priorities for Chinasenergy sector Appl Energy 199756(3ndash4)237ndash51
[16] Wang YM Chin KS Fuzzy analytic hierarchy process a logarithmic fuzzypreference programming methodology Int J Approx Reason 201152(4)541ndash53
[17] Nigim K Munier N Green J Pre-feasibility MCDM tools to aid communities inprioritizing local viable renewable energy sources Renew Energy 200429(11)1775ndash91
[18] Daniel J Vishal NV Albert B Selvarsan I Evaluation of the signi 1047297cant renew-able energy resources in India using analytical hierarchy process LondonSpringer 2010 p 13ndash26
[19] Ahmad S Tahar RM Selection of renewable energy sources for sustainabledevelopment of electricity generation system using analytic hierarchy processa case of Malaysia Renew Energy 201463458ndash66
[20] Tasri A Susilawati A Selection among renewable energy alternatives based ona fuzzy analytic hierarchy process in Indonesia Sustain Energy Technol Assess2014734ndash44
[21] Kaya T Kahraman C Multicriteria renewable energy planning using anintegrated fuzzy VIKOR amp AHP methodology the case of Istanbul Energy2010352517ndash27
[22] Zakerinia M Ghaderi F Piltan M Optimal portfolio selection between differentkinds of renewable energy sources 2010
[23] Barin A Canha LN Abaide A Magnago K Wottrich B Multicriteria analysis of the operation of renewable energy sources taking as basis the AHP methodand fuzzy logic concerning distributed generation systems Online J ElectronElectric Eng (OJEEE) 20091(1)52ndash7
[24] Heo E Kim J Boo KJ Analysis of the assessment factors for renewable energydissemination program evaluation using fuzzy AHP Renew Sustain EnergyRev 201014(8)2214ndash20
[25] Akhgari P Kamalan H Monavari M Utilizing AHP method in rating of highlighted environment parameters in most used renewable energies inelectricity production Eur J Sci Res 201160182ndash8
[26] Murphy DJ Hall CAS Year in reviewmdashEROI or energy return on (energy)
invested Ecol Econ Rev 20101185102ndash18[27] Urdaneta AJ Chankong V A multiobjective minimax approach to controller
settings for systems running under disturbances Control-Theory Adv Technol19895(4)391ndash411
[28] Gonzaacutelez-Longatt F Teraacuten R Meacutendez J Hernaacutendez A Guilleacuten F Evaluacioacuten delRecurso Eoacutelico en Venezuela Parte I Congreso Petrolero Energeacutetico ASME-UNEFA Puerto Cabello 2006 [in Spanish]
D Hernaacutendez-Torres et al Renewable and Sustainable Energy Reviews 52 (2015) 100ndash110110