Emissions trading in Hong Kong and the Pearl River Delta region ...
Transcript of Emissions trading in Hong Kong and the Pearl River Delta region ...
Emissions trading in Hong Kong and the Pearl RiverDelta region—A modeling approach to tradedecisions in Hong Kong’s electricity industry
Richard C.M. Yam, W.H. Leung *
Department of Systems Engineering and Engineering Management, City University of Hong Kong, Hong Kong
2
e n v i r o n m e n t a l s c i e n c e & p o l i c y 3 1 ( 2 0 1 3 ) 1 – 1 2
a r t i c l e i n f o
Article history:
Received 18 October 2011
Received in revised form
5 March 2013
Accepted 19 March 2013
Published on line 23 April 2013
Keywords:
Air pollution
Regional emissions trading
Emission control
Modeling
a b s t r a c t
In 2002, the Hong Kong government and the Guangdong provincial government agreed to
reduce emissions of sulfur dioxide, nitrogen oxides, respirable suspended particulates, and
volatile organic compounds by 40%, 20%, 55%, and 55%, respectively. There was strong
public demand for the power stations in Hong Kong to reduce emissions. Emission caps
were introduced, with allowances for the trading of emission credits. However, local power
stations were using equipment built in the 1980s and 1990s, making it difficult for them to
meet the new emissions requirements. The situation presented a new challenge, which
involved a choice of either improving the existing equipment, or using emissions trading to
meet the emission caps. This study reviews the background on emissions in Hong Kong and
the surrounding regions, the ‘‘cap and trade’’ system, and the technologies used for power
generation and emission reduction. A modeling approach is adopted to simulate the
equipment, the electricity dispatching requirements, and the costs of either reducing
emissions or trading emission credits. Data from a power station in Hong Kong was chosen
for the simulation. Different options were simulated in the model to identify the optimal
strategy. The results were then compared with the plan for emission reduction. This study
demonstrates that a modeling approach using linear programming can analyze the com-
plicated options involving emission reduction and investments to achieve an optimized
business solution.
# 2013 Elsevier Ltd. All rights reserved.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/envsci
1. Introduction
Air pollution in Hong Kong (HK) has worsened since the early
2000s. The number of hours of reduced visibility increased
from 480 h in 1997 to 1490 h in 2003 (HKO, 2007). Hong Kong
faces two major air pollution problems: street-level and
region-wide air pollution. In 2002, the Hong Kong government
commissioned a study of the air quality in the surrounding
Pearl River Delta region (PRD). The PRD comprises an area of
* Corresponding author. Tel.: +852 69057108.E-mail addresses: [email protected], [email protected]
1462-9011/$ – see front matter # 2013 Elsevier Ltd. All rights reservedhttp://dx.doi.org/10.1016/j.envsci.2013.03.010
42,794 km with a population of 38.7 million around the
estuary of the Pearl River, whereas Hong Kong comprises only
1000 km2 and a population of 7 million. The report, entitled
‘‘Final Report – Study of Air Quality in the Pearl River Delta
Region’’ (CH2M HILL, 2002), revealed that the main sources of
pollution in the region came from the energy, industry, and
transportation sectors. Electricity generation contributed
most of the pollution in the energy sector.
The report considered four major air pollutants to be
important in the region: sulphur dioxide (SO2), nitrogen oxides
m (W.H. Leung).
.
e n v i r o n m e n t a l s c i e n c e & p o l i c y 3 1 ( 2 0 1 3 ) 1 – 1 22
(NOx), respirable suspended solids (RSP), and volatile organic
compounds (VOC) from hydrocarbons. In 1997, the total
emissions of SO2, NOx, RSP, and VOC in the region were
585,900, 643,000, 256,400, and 480,600 tons respectively, of
which Hong Kong produced 65,960, 123,000, 11,400, and
68,900 tons respectively. The energy sector in Hong Kong
contributed approximately 64,300, 59,700, and 6500 tons of
SO2, NOx, and RSP respectively, from generating 23,574 GWh of
electricity. The region has been one of the fastest growing
economic development areas in China, with an annual GDP
growth rate of more than 4.5% in Hong Kong (Census and
Statistics Department, 2007) and 11–12% in the PRD. Thus, the
report predicted increasing energy consumption and worsen-
ing emissions problems in the region.
In 2002, the Hong Kong government and the Guangdong
provincial government agreed to reduce, on a best-endeavor
basis, the regional emissions of sulfur dioxide, nitrogen
oxides, respirable suspended particulates, and volatile organic
compounds by 40%, 20%, 55%, and 55% respectively, by 2010,
using 1997 as the base year (HK Government news, 2002).
There was a strong demand for power companies to reduce
their emissions. Based on the agreement, the required
emission targets for Hong Kong’s electricity generation in
2010 were estimated at 25,000, 42,600, and 1260 tons of SO2,
NOx, and RSP respectively. These reduction targets were
substantial. Emissions trading was proposed as a solution for
the electricity generation industry to achieve the targets in
2002 (Liao, 2002). Emissions trading was completely new and
quite different from the emission controls that were in use at
that time. This option presented a new challenge for power
station managers who had to choose between improving their
existing equipment and/or using emissions trading to meet
the emission targets. The aim of this study was to explore a
modeling approach that would facilitate power station
managers in making such decisions.
Many different emissions trading systems have been
reported around the world (Leung et al., 2009). However, only
two major popular emissions trading programs have been
actively implemented: the Acid Rain Program based on Title IV
of the Clean Air Act adopted in the US (US EPA, 2006), and the
EU emissions trading program based on Directive 2003/87/EC
and Council Directive 96/61/EC (EC Directive, 2003) in Europe.
Both programs were developed from the ‘‘cap and trade’’
system. In HK, a pilot scheme for emissions trading among the
power stations based in HK and the PRD was also based on a
‘‘cap and trade’’ system (EPD HK, 2007a).
2. ‘‘Cap and trade’’ system
The ‘‘cap and trade’’ system is a means of preventing
emissions from exceeding government-set limits through
the trading of emission credits among the major emission
sources. An emissions monitoring station, usually the govern-
ment’s environmental protection agency, sets the air quality
objectives based on public expectations and assigns emission
caps to emissions permit holders, i.e. the power stations, who
either own or operate the emission sources. The emissions permit
holders must meet emission caps, either by improving their
equipment or by trading emission credits in the emissions
trading market. The emissions monitoring station monitors the
emissions from the emission sources, oversees the operation of
the emissions trading program, and enforces public policy by
imposing heavy penalties for any exceedance of emission
standards.
An effective ‘‘cap and trade’’ program requires legislation
so that it is enforceable by the authorities. The emission caps
must meet the air quality objectives and be respected by all of
the parties concerned. There should also be a global allocation
plan for emission caps, supported by agreement between the
emissions monitoring stations and the emissions permit holders.
The important rules in an allocation plan must be fair and
equitable and should allow for the new entry of permit
holders. The system requires a comprehensive emissions
measuring, monitoring, and reporting system. Continuous
emissions monitoring supplemented by an agreed calculation
system is commonly adopted, with heavy penalties for
exceeding limits to deter noncompliance. The pricing of
emission credits depends on the market and the cost of the
emission reduction work. Most credit trading is done through
agreed contracts that must be scrutinized by the emissions
monitoring stations.
To meet the emission caps, the emissions permit holders can
use cleaner fuels, or remove pollutants in the flue gas before
emitting it into the atmosphere. Alternatively, the permit
holders can buy emission credits to meet their emission caps,
or sell their emission credits to others when they have a
surplus. The emissions permit holders must strike a balance
between investing in emission-reduction equipment and
paying the price of emission credits. In the electricity industry,
the main concerns of the emissions permit holders are the
operating and maintenance costs, the capital investments and
the return on investment.
3. Technologies for electricity generation andpollution reduction
The emissions levels from power plants depend heavily on
which technologies the plants use for electricity generation.
The main fuels used for conventional power plants are coal
and oil. Natural gas is used in combined cycle generating
power plants.
In coal-fired or oil-fired power plants, fuels are burned in
the boiler to produce heat energy that is then converted into
mechanical energy for electricity generation. The energy
conversion efficiency for a power plant is the ratio between
the useful output of the power plant and the input in energy
terms. The theoretical energy conversion efficiency of this
process is between 35 and 42%, depending on the operating
conditions of the plant (Eastop and McConkey, 1986). With the
latest developments in technology and materials, it is possible
to achieve actual conversion efficiency of 42% at full load
(Alstom Power, 2011b).
In combined cycle plants, air is compressed in the air
compressor and gas fuel is injected into the combustion
chamber to burn with the compressed air. The combustion
process produces heat. Residual heat from the gas turbine is
used to heat water into steam in the heat-recovery steam
generator. The heat energy from the compressed air and
e n v i r o n m e n t a l s c i e n c e & p o l i c y 3 1 ( 2 0 1 3 ) 1 – 1 2 3
steam is then converted into mechanical energy for electricity
generation. The theoretical conversion efficiency is between
55 and 59% for this process (Eastop and McConkey, 1986). With
improvements in technologies and materials, it is possible to
achieve actual conversion efficiency of 59% in a combined
cycle power plant at full load (Alstom Power, 2011a). Both
processes, however, produce and emit pollutants into the
atmosphere.
3.1. Emission-reduction equipment
According to a World Bank study on clean coal technologies
(Tavoulareas and Charpentier, 1995), emissions can be
controlled by using cleaner fuel, by controlling the combustion
process, and/or by cleaning the flue gas before discharging it
into the atmosphere. As SO2 is produced when the fuel
contains sulfur, using fuel with less sulfur content produces
less SO2. SO2 emissions can also be reduced by cleaning the
flue gas after combustion with weak alkaline substances such
as limestone or lime slurry. This process can remove most of
the acidic SO2 in a flue gas desulfurization (FGD) plant.
NOx is produced when fuel burns in a combustion chamber.
Nitrogen combines with oxygen to form NOx. The amount of
NOx produced depends on the flame temperature. Low NOx
burners can reduce NOx by ‘‘staging’’ the combustion with
overfired air and reburning. Alternately, the NOx can be
cleaned using selective catalytic reduction (SCR) technology.
Ammonia is injected into the flue gas to react with NOx in the
catalytic reactor, producing molecular nitrogen and water
vapor, both of which are harmless.
Burning fuel with ash produces suspended particulates.
These particulates can be collected by an electrostatic
precipitator or by fabric filtration before discharging the flue
gas into the atmosphere. Normally, such filtration devices are
part of the boiler. An FGD plant at the exit of the boiler can also
remove the suspended particulates when the flue gas is
washed by the slurry of the reagent.
3.2. The effectiveness and cost of emission-reductionequipment
According to the World Bank study on clean coal technologies
(Tavoulareas and Charpentier, 1995), the effectiveness of
emission-reduction equipment is measured by its removal
efficiency, which is the proportion of the emissions removed
by emission-reduction equipment relative to the amount at
the inlet. The typical removal efficiencies of different
emission-reduction technologies vary according to the type
of emission. The removal efficiency of an FGD plant is about
90% for SO2, that of a low NOx burner is about 60% for NOx, and
that of electrostatic precipitation is about 99.9% for RSP
(Takahashi, 1997; US EPA, 2007). A selective catalytic reactor
can also remove about 70% of the NOx from the flue gas (US
DoE, 1997).
The capital cost of emission-control equipment depends on
the technology, design, materials, and size of the unit. The
operating and maintenance (O&M) cost of such equipment is
dominated by the cost of the materials and labor required to
operate the machines. This cost also depends on the amount
of emission reduction that the equipment can achieve.
Tavoulareas and Charpentier (1995) reported that the capital
and O&M costs for an FGD plant were US$ 100–150/kW of
capacity and US¢ 0.15–0.33/kWh of electricity generated,
respectively. The capital and O&M costs for a low NOx burner
were US$ 20–50/kW of capacity and US¢ 0.5–1/kWh of
electricity generated, respectively, while those for an SCR
plant were US$ 45/kW of capacity and US$ 2165/tons of NOx
respectively. In the PRD, the equipment price for retrofitting
FGD to a group of power plants was about US$ 50/kW of
capacity, derived from the ‘‘Shandong flue gas desulfuriza-
tion’’ project in China (World Bank, 2007).
4. Air pollution control and emissions tradingframework in the region
In HK, emission control is enforced under the Air Pollution
Control Ordinance (Chapter 311 of the laws of the Hong Kong
Special Administrative Region) (Ordinance HK, 1997) and the
Air Pollution Control (Specified Processes) Regulation (Regula-
tion HK, 2007). The major stationary air polluters, such as
power plants, are subject to stringent emission controls. Their
licenses specify all of the emission limits for their operations,
which are normally given in terms of allowable concentrations
of emissions. Emission caps were imposed on power plants in
2006.
In 1997, the Hong Kong government ceased to allow the
construction of new coal-fired generating units in favor of
natural gas plants (EPD HK, 2011). Only renewable energy or
natural gas-fired combined cycle plants were considered
thereafter. The renewable energy options for electricity
generation include hydropower, solar, wind, geothermal,
biomass, nuclear, or ethanol (US Energy, 2005). The emissions
from these types of plants are zero (EMSD, 2011).
In the PRD, several laws have been enacted to control
emissions. The principal law is the ‘‘Law of the People’s
Republic of China on the prevention and control of atmo-
spheric pollution’’ (Law of PRC, 2000). According to this law,
the central government sets nationwide standards for emis-
sions. Local governments can set their own requirements for
local standards of emission control. The central authorities
delegate power to local governments for setting any standards
not covered by the national regulations, or for making local
standards more stringent than their national counterparts. In
addition to these emission-control laws, there are national
standards to govern emissions such as the ‘‘GB13223-1996
emission standard of air pollutants for thermal power plants’’
(GB China, 1996). Local governments have also set up various
regulations, such as the city of Shenzhen’s ‘‘Regulations on
controlling acid rain and emissions of sulfur dioxide in
Shenzhen’’ (Regulation SZ, 1998).
An emissions trading agreement, entitled the ‘‘Emissions
trading pilot scheme for thermal power plants in the Pearl
River Delta region,’’ has been signed by the Hong Kong and
Guangdong provincial governments (EPD HK, 2007a). Since
then, emissions trading scheme has been executed in HK and
PRD. However, both the cleaner fuel and the retrofitting emission-
reduction equipment approaches can meet emission targets
from 2010 to 2016 in Hong Kong. Emissions trading is only
considered as a market-based fall-back option, in case of an
Demand forecast
Genera�on mach ines
Coal -firedNatural gas -fired
Emiss ions
Emiss ion caps
Compare with cap
Sell
Emiss ions trading market
Improve the exis�ng equ ipment
Buy
Produ ce electricity
Cost of bu ying emiss ion all owanc e
Cost of capital and O&M
Fig. 1 – The details of the emissions trading model (ETM).
e n v i r o n m e n t a l s c i e n c e & p o l i c y 3 1 ( 2 0 1 3 ) 1 – 1 24
unexpected emergency in normal operations. This scheme is
intended to support the trading of SO2 emission credits and
also includes the trading of NOx and RSP emissions. This is a
‘‘cap and trade’’ system on a project basis. An emissions permit
holder who is interested and eligible may propose emission
reduction plans that would then be deemed acceptable by the
environmental protection authorities. The surplus between
the base emissions targets before and after the completion of
an emission reduction plan can be sold as ‘‘project-based
emission credits.’’ All participating power stations must
install suitable systems to monitor their emissions. Emission
credit trading is a market activity in the form of a business
contract that must be submitted to the environmental
protection authorities for approval. Once the authorities verify
the actual total emissions for a year, any emission credits can
then be transferred. If the seller fails to transfer the emission
credits, then he/she must compensate the buyer according to
the contract. To enforce overall compliance with regional
pollution limits, the central and local governments prosecute
violators of emission caps in accordance with their laws and
regulations.
5. Business environment of Hong Kong’selectricity generation industry
Electricity generation in Hong Kong is controlled by a business
model referred to as the Scheme of Control Agreement, which
applies to emissions permit holders. The government controls
the investments and profits of the emissions permit holders in
accordance with the Scheme, which is designed to ensure a
safe and reliable supply of electricity at an affordable price. A
new agreement was signed on January 1, 2009 to cover the next
10 years, with a possible extension for a further five years (EB
HK, 2008a, 2008b). Linear depreciation of fixed assets is
adopted. The permitted rate of return on electricity generation
is 9.99% on average net fixed assets and 11% on average
renewable net fixed assets. Incentives or penalties on the
permitted return are also given, based on environmental
performance, customer performance, energy efficiency, and
renewable energy production. Hence, HK’s electricity market
is still regulated, and will remain so for at least several years.
6. Emissions trading model (ETM) forelectricity generation with multi-fuel operationsin Hong Kong
To analyze the complicated emissions and electricity genera-
tion industry in Hong Kong, the characteristics of HK’s
electricity industry and the emissions trading program are
integrated into the Emissions Trading Model (ETM) for Hong
Kong, which applies mainly to natural gas-fired and coal-fired
power plants.
The driving force of the industry is the demand for
electricity. Power plants try to meet the demand for electricity
while controlling the emissions they produce in the process.
The emissions are reduced either by using clean coal
technologies or by trading emission credits to meet the
emission caps. The ETM is designed to capture the factors
involved, as Fig. 1 highlights.
6.1. The structure of the ETM
The ETM is separated into four modules: (1) an equipment
module, (2) a load dispatching module, (3) an emission
module, and (4) a financial module.
6.1.1. The equipment moduleThe equipment module is primarily based on the technology
of electricity generation. Two major types of electricity
generation equipment are considered: the combined cycle
gas turbine plant burning natural gas and the coal-fired plant.
The modules created for combined cycle gas turbine and coal-
fired power plants are shown in Fig. 2.
The emission performance of a power plant is measured by
its emission factor. When the emission performance is better,
the emission factor is lower. The emission factor for a specific
emission is measured by the amount of emission produced
divided by the amount of electricity generated (US EPA, 2011).
The emission factors for SO2, NOx and RSP for coal-fired and
combined cycle power plants can be calculated using this
equipment module.
Combined cycle power plants use natural gas, which
contains very little sulfur and ash, hence SO2 and RSP
emissions are close to zero. NOx emissions can also be
controlled by using a dry low NOx burner in the combustion
chamber. The energy conversion efficiency of a combined
cycle power plant is high, at around 55–59%. The emission
factors are low and the emission performance of the power
plant is high (GE Energy, 2011).
Coal-fired power plants use coal, which contains higher
sulfur and ash, thus SO2 and RSP are produced after
combustion. The combustion process also produces NOx.
The energy conversion efficiency of this type of power plant is
lower than that of a combined cycle plant, at around 35–42%.
The emission factors are higher and the emission perfor-
mance is lower than for a combined cycle plant.
Fig. 2 – The combined cycle gas turbine and coal-fired power plant modules.
e n v i r o n m e n t a l s c i e n c e & p o l i c y 3 1 ( 2 0 1 3 ) 1 – 1 2 5
6.1.2. The load dispatching moduleThe demand for electricity is allocated to different power
plants according to priority, which can be determined either
by the efficiency of the power plant or by its emissions
performance. Under stringent emission control, the electricity
demand will be logically dispatched according to emission
performance: a gas-fired combined cycle power plant will have
priority over a coal-fired power plant. For power plants with
the same emission performance, a plant with higher efficiency
will have priority over a less efficient one.
Each machine bears a load factor, which is the ratio of the
actual annual electricity generated to the total annual possible
production capacity. Hence, the actual electricity production
can be estimated by multiplying the total annual possible
capacity by the load factor.
The total emissions can be predicted by multiplying the
actual electricity generated by the emission factor, which can
Fig. 3 – The load disp
be obtained from the ‘‘equipment module.’’ Each emission for
a power plant is then summed to find the total emission of the
power station. In this module, the total emission of each
pollutant is estimated according to the electricity generated,
the emission factors of the machines and the mode of load
dispatching.
The module created for load dispatching is shown in Fig. 3.
6.1.3. The emissions module
In the emissions module, the total emission of each pollutant
is compared with the emission cap. If the emission caps are
not exceeded, then nothing needs to be done and the surplus
emission credit can also be sold in the emissions trading
market. If the total emissions exceed the emission caps, then
the emissions permit holder can either buy emission credits from
the emissions market, or try to improve its emissions
performance. The emissions module is shown in Fig. 4.
atching module.
Emission cap s
SO2, NOx, RSPIf actual
> cap
Do no thing or
sell emiss ion s
allowance
yes
Improve the
existing
equipmen t
Trade
emiss ion
credits
no
The emiss ion s modu le
Emissions
SO2, NOx, RSP
Emiss ions
SO2, NOx, RSP
Emiss ions
SO2, NOx, RSP
Sum of each emiss ion , SO2, NOx, RSP
Fig. 4 – The emissions module.
e n v i r o n m e n t a l s c i e n c e & p o l i c y 3 1 ( 2 0 1 3 ) 1 – 1 26
6.1.4. The financial moduleIf the actual emissions exceed the emission caps, the emissions
permit holder must either improve its equipment or buy
emission credits. Both options involve financial costs. Im-
proving the equipment includes upgrading the existing
installation or adding new emission-reduction equipment.
This option requires capital investment and the future O&M
cost of the equipment, hence the financial module of the ETM
is separated into two parts: (1) improving the existing plant by
installing emission-reduction equipment and (2) trading
emission credits with other emissions permit holders.
6.1.4.1. Financial module for improving existing equipment. In-
stalling emission-reduction equipment requires capital
Fig. 5 – The finan
investment. This capital cost can be obtained by multiplying
the capacity of the power plant by the capital cost of the
equipment per unit capacity as given by Tavoulareas and
Charpentier (1995). This capital investment generates two
capital expenses, the depreciation of the equipment and the
O&M costs incur to operate and maintain the new equipment.
A linear depreciation of capital cost is adopted. The O&M costs
are calculated by multiplying the amount of electricity
produced by the O&M costs per unit electricity produced as
given by Tavoulareas and Charpentier. The operating cost for
emission-reduction is the sum of the depreciation cost and the
O&M costs of the additional emission-reduction equipment.
The predicted price of an emission is calculated from these
costs which can be used as the price to trade the emission
credits with the other power stations. Since different power
plant has different ‘‘predicted price of the emission credit’’,
this difference creates the incentive to trade among the power
stations. In determining whether to ‘‘trade’’ or not in the
emissions trading market, the emissions permit holder would
compare the predicted price of the emission credit to the price
of buying the emission credit in the market for a period of
time.
6.1.4.2. Financial module for trading emission credits. The only
operating cost for this module is the cost of buying the
emission credits. If there is an emission trading market, the
price of emission credit can be determined by the trading
market. Since the actual emissions trading market was not
exercised, the price of the emission credit had to be
estimated. The price of emission credit is based on the
principle of the pricing in the emissions market, i.e. the
difference average operating cost of an emission reduction
equipment between HK and PRD according to the cap and
trade system (US EPA, 2006). In Hong Kong, the international
price of the emission reduction equipment is used, which is
derived from the research data published by the World Bank
(Tavoulareas and Charpentier, 1995). However, in PRD the
cial module.
z = total cost of emission
reduction for a power station
i = an emission
F = cost of buying emission
credits
I = all emissions
C = the cost of operating the
emission-reduction equipment
m = price of emission credit
E = electricity generated
e = quantity of emissions
a = additional emission-
reduction equipment
i (cap) = emission cap
T = the total years of study
after the base year
a = operating and
maintenance cost per
unit of emission reduction
1 = the base year
b = rate of depreciation
of the capital cost
t = a year after the
base year
C = capital cost of emission-
reduction equipment
r = discount rate
j = removal efficiency
NPV = net present value
s = emission per unit of
electricity generated
(emission factor)
n = power plant of the
power station
F = fuel specification and
N = all power plants
of the power station
h = conversion efficiency
of the power plant
e n v i r o n m e n t a l s c i e n c e & p o l i c y 3 1 ( 2 0 1 3 ) 1 – 1 2 7
home market price is used, which is derived from a project on
emission reduction submitted by the Chinese Government to
the World Bank (2007). The price in HK is higher than that in
PRD. Hence, it is possible to trade the emissions between HK
and PRD.
For a power station, the total operating cost of the emission
reduction is the sum of the cost of operating the additional
emission equipment and the cost of buying the emission
credits if the limit is exceeded.
The financial module is shown in Fig. 5.
6.2. Linear programming method for optimization
The ETM model applies to the multi-variables to meet multi-
targets, i.e. the emission caps (e(cap)) on the emissions (i) for
electricity generation. The linear programming (LP) method
can be used to describe the cost/benefit optimization process.
The objective function of the LP model is the total operating
cost of the emission reduction z(ET, aT) for (N) power plants
with two variables of electricity generated (Et) and the
additional emission-reduction equipment for each power
plant (at) in a period (T). In this model, Et in each period t is
given based on electricity demand forecast. The total operat-
ing cost z(Et, at) is composed of the net present value (NPV) at a
discount rate (r) for two elements: the cost of buying emission
credits (F) and the cost of operating the emission-reduction
equipment (C) to meet the emission caps (e(cap)) from the base
year (1) to year (T). Both F and C are functions of the electricity
generated (Et) and the additional emission-reduction equip-
ment (at).
The cost of buying emission credits F (Et, at) is the sum of
the price of emission credits m (ai,t) times the emission
exceedance ([P
ei (En,t, an,i,t) � ei (cap),t]) for each emission. The
cost of operating the emission-reduction equipment C (Et, at) is
the sum of O&M cost which is calculated by multiplying
electricity produced (Et) by the O&M cost per unit electricity a
(an,i,t) plus the depreciation cost which is calculated by
multiplying the capital cost C (an,i,t) with the rate of deprecia-
tion b (an,i,t).
To determine whether to purchase emission credits or not,
it is necessary to minimize the operating cost of emission
reduction for a period (T). However, it has to meet the
constraint that there is no exceedance of emission caps with
the period (T) for all emissions (I). The following is the LP model
function for the ETM.
To minimize
zðET; aTÞ ¼XT
t¼1
NPVfrt; ½FðEt; atÞ þ CðEt; atÞ�g
with function F (Et, at) is equal to
with FðEt; atÞ ¼XI
i¼1
mðai;tÞ �XN
n¼1
eiðEn;t; an;i;tÞ � eiðca pÞ;t
" #
and CðEt; atÞ ¼XI
i¼1
XN
n¼1
f½aðan;i;tÞ � En;t� þ ½bðan;i;tÞ � Cðan;i;tÞ�g
while eiðEn;t; an;i;tÞ ¼ ½1 � jðan;i;tÞ� � sðFn; hnÞ � En;t
provided that ½eiðca pÞ;t � eiðEt; ai;tÞ� 3 0 for
ð1 � t � TÞ \ ð1 � i � IÞ
where
6.3. Different business strategies for the emissionstrading program
The electricity generation in the study period is based on the
predicted growth rate of the electricity demand starting from
the base year. For the provision of additional emission-
reduction equipment, there are many possible combina-
tions, i.e. a multiple of the number of power plants and
emissions. To simplify the simulation, four possible business
strategies are considered to meet the new emissions
requirements.
Option 1: Emissions trading strategy: rely on the emissions
trading market to buy and sell emission credits to
meet the emission caps. It is not necessary to install
any emission-reduction equipment.
Option 2: Mixing strategy: install limited emission-reduction
equipment for SO2 and NOx and buy emission
credits to cover any exceedance of emission limits.
This is a compromise between Option 1 and
Option 3.
Option 3: Emission reduction strategy: install just enough emis-
sion-reduction equipment to meet the emission
caps with normal outage for maintenance.
Option 4: Secured emission reduction strategy: install enough
emission-reduction equipment to meet the emis-
sion caps with allowance for a single serious
equipment failure per year.
6.4. Simulation of the ETM
The four modules of the ETM are simulated in a computer
program to calculate the total operating cost, emissions, and
the capital investment for each option. The boundary
conditions of the simulation are as follows:
e n v i r o n m e n t a l s c i e n c e & p o l i c y 3 1 ( 2 0 1 3 ) 1 – 1 28
(1) The period of simulation: 2006–2016.
(2) The model simulates SO2, NOx, and RSP emissions because
these are the pollutants of concern in the region.
(3) The characteristics of the power plants as an emission
permit holder.
The input variables are (1) the electricity demand forecast
from 2006 to 2016 and (2) the additional emission-reduction
equipment to be installed in a specific year for each option.
The outputs from the computer program are (1) the total
emissions for each year, (2) the capital cost required for
additional emission-reduction equipment, and (3) the total
operating costs of the new equipment.
7. Simulation of the ETM, 2006–2016
The power plants of an emissions permit holder in Hong
Kong, Company A, are chosen to demonstrate the simula-
tion. Company A has a total installed capacity of 6908 MW.
The company operates four 350 MW and four 677 MW
conventional coal-fired units that have been fitted with
low NOx burners and electrostatic precipitators, and
eight 312.5 MW gas-fired combined cycle units. The total
electricity generated in 2006 was 26,400 GWh (CLP, 2007;
PAH, 2011).
The simulation starts in 2006, 10 years after the base year of
1997 when the new emission controls were introduced in HK.
The following parameters are input to the program: (1) the
characteristics of electricity generation machines including its
capacity, type, emission-reduction facilities if any, (2) the
depreciation requirement from the ‘‘scheme of control’’
agreement, (3) the emission caps based on the emission
reduction agreement of 2002 between HK and the PRD, and (4)
the emissions trading pilot scheme for thermal power plants
in the HK-PRD region.
Based on the historical data, the growth in electricity
generation was about 1.5% from 2000 to 2006 (CLP, 2007; PAH,
2011). This growth rate is used to project the electricity
demand from 2006 to 2016. The authority allocated the
emission caps for Company A up to 2009 (EPD, 2005, 2007b),
after which the emission caps were estimated according to the
agreement on air quality (CH2M HILL, 2002).
The electricity demand forecast and the predicted emission
caps for Company A are shown in Table 1.
The international prices were used for (1) the equipment
capital cost relating to the capacity of the unit and (2) the
operation and maintenance (O&M) costs per unit of emission
reduction for retrofitting emission-reduction equipment. The
Table 1 – The estimated electricity generation and the predict
Year 2006 2007 2008 2009
Electricity production (GWh) 26,408 26,804 27,204 27,608
SO2 emission cap (T/yr) 50,000 46,000 41,400 39,400
NOx emission cap (T/yr) 30,400 28,800 27,660 27,400
RSP emission cap (T/yr) 2030 1710 1165 1115
inflation rate and the discount rate in HK was about 3% for the
period (Trade Economics, 2012).
7.1. The simulation of the LP model for an emissionspermit holder from 2006 to 2016
Based on the above estimations, a simulation of the electricity
generation for Company A from 2006 to 2016 was conducted.
The results of the simulation are as follows.
For Option 1, emission allowances for SO2, NOx, and RSP are
purchased in the emissions market to meet the emission caps.
No other capital or operating costs are incurred. The
simulation is applied to SO2, NOx, and RSP emission reduction
separately.
The results of the simulation show that there is no
exceedance of the SO2, NOx, or RSP caps from 2006 to 2008
because the emissions are below the levels allowed by the
emission caps. However, beginning in 2009, the SO2 and RSP
emissions exceed their respective emission caps, and NOx
emissions exceed the emission cap from 2010 onwards. Hence,
Company A has to buy emission credits. The results of this
simulation are shown in Table 2.
For Option 2, only limited emission-reduction equipment
is provided to reduce emissions, with some allowance for
emission exceedance. Company A provides one FGD system
for one 677 MW coal-fired power plant per year in 2009 and
2010, and one selective catalytic reactor for one 677 MW
coal-fired power plant in 2010. The emission exceedance is
covered by buying credits from the emissions trading
market.
The results of the simulation for Option 2 are summarized
in Table 3.
In Option 3, no exceedance of emissions is allowed.
Company A needs to install one FGD system in one of its
677 MW coal-fired units in 2009, two FGD systems in 2010, and
one in 2011, with one selective catalytic reactor for each of two
677 MW coal-fired units, the first in 2010 and the second in
2011.
The results for the simulation for Option 3 are summarized
in Table 4.
In Option 4, the company has to provide one more
emission-reduction system in addition to those installed in
Option 3. Therefore, the company needs to install two FGD
systems in two of its 677 MW coal-fired units in 2009 and 2010,
respectively, one FGD in one of its 350 MW coal-fired units in
2011, and two selective catalytic reactors in two 677 MW coal-
fired units in 2010.
The simulation results for Option 4 are summarized in
Table 5.
ed emission caps for Company A (2006–2016).
2010 2011 2012 2013 2014 2015 2016
28,016 28,428 28,844 29,264 29,688 30,115 30,547
14,850 14,850 14,850 14,850 14,850 14,850 14,850
26,400 26,400 26,400 26,400 26,400 26,400 26,400
745 745 745 745 745 745 745
Table 3 – The simulation results on emission exceedance costs for Company A, 2006–2016 (Option 2).
Emission Items Unit Year Total
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
SO2 Produced Tons 36,591 40,830 41,070 32,741 23,416 26,642 29,436 31,008 32,538 34,139 35,754 363,857
Exceedance Tons 0 0 0 0 8531 11,757 14,551 16,123 17,654 19,254 20,869 108,430
Capital Cost HK$ million 0 0 0 316.8 316.8 0 0 0 0 0 0 633.7
Total operating cost HK$ million 0.0 0.0 0.0 62.1 117.7 133.0 145.7 154.4 163.0 172.8 181.8 1448.7
NOx Produced Tons 23,897 24,499 25,104 25,716 22,541 23,167 23,799 24,434 25,052 25,700 26,354 270,263
Exceedance Tons 0 0 0 0 0 0 0 0 0 0 0 0
Capital Cost HK$ million 0.0 0.0 0.0 0.0 237.6 0.0 0.0 0.0 0.0 0.0 0.0 237.6
Total operating cost HK$ million 0.0 0.0 0.0 0.0 72.1 70.3 68.5 66.6 64.8 63.0 61.1 466.4
RSP Produced Tons 1492 1391 1104 909 715 750 786 821 856 892 929 10,745
Exceedance Tons 0 0 0 0 0 8 43 78 113 149 186 577
Total operating cost HK$ million 0.0 0.0 0.0 0.0 0.0 0.1 0.4 0.7 1.0 1.3 1.7 5.2
Table 2 – The simulation results of emission exceedance costs for Company A, 2006–2016 (Option 1).
Emission Items Unit Year Total
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
SO2 Produced Tons 36,591 40,830 41,070 43,425 44,741 49,521 53,331 54,855 56,338 57,891 59,459 538,051
Exceedance Tons 0 0 0 4025 29,856 34,636 38,446 39,970 41,453 43,006 44,574 275,390
Cost of exceedance HK$ million 0.0 0.0 0.0 23.6 173.6 185.7 195.3 201.4 207.4 215.2 224.0 1426.3
NOx Produced Tons 23,897 24,499 25,104 25,716 26,333 26,957 27,591 28,227 28,845 29,493 30,146 296,811
Exceedance Tons 0 0 0 0 0 556 1189 1824 2442 3090 3744 12,840
Cost of exceedance HK$ million 0.0 0.0 0.0 0.0 0.0 10.3 21.9 33.7 45.1 56.4 67.5 234.9
RSP Produced Tons 1492 1391 1104 1138 1173 1208 1243 1278 1313 1349 1386 14,074
Exceedance Tons 0 0 0 23 430 465 500 536 570 607 643 3774
Cost of exceedance HK$ million 0.0 0.0 0.0 0.2 3.9 4.2 4.5 4.8 5.1 5.5 5.8 34.0
e n
v i
r o
n m
e n
t a
l s
c i
e n
c e
&
p
o l
i c
y 3
1 (
2 0
1 3
) 1
– 1
2
9
Table 4 – The simulation results on emission exceedance costs for Company A, 2006–2016 (Option 3).
Emission Items Unit Year Total
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
SO2 Produced Tons 36,591 40,830 41,070 32,741 23,416 26,642 29,436 31,008 32,538 34,139 35,754 363,857
Capital Cost HK$ million 0 0 0 316.8 633.7 316.8 0 0 0 0 0 1267.3
Total operating cost HK$ million 0.0 0.0 0.0 30.4 104.0 147.7 148.4 149.0 149.6 149.6 149.6 1028.4
NOx Produced Tons 23,897 24,499 25,104 25,716 22,541 23,167 23,799 24,434 25,052 25,700 26,354 270,263
Capital Cost HK$ million 0.0 0.0 0.0 0.0 237.6 0.0 0.0 0.0 0.0 0.0 0.0 237.6
Total operating cost HK$ million 0.0 0.0 0.0 0.0 48.3 48.3 48.3 48.3 48.3 48.3 48.3 133.7
RSP Produced Tons 1492 1391 1104 909 487 366 376 387 399 435 471 7818
Total operating cost HK$ million 0.0 0.0 0.0 24.5 105.6 105.6 151.0 151.0 151.0 151.0 151.0 990.7
Table 5 – The simulation results on emission exceedance costs for Company A, 2006–2016 (Option 4).
Emission Items Unit Year Total
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
SO2 Produced Tons 36,591 40,830 41,070 22,057 6623 7357 8005 8322 8637 8967 9305 197,764
Capital Cost HK$ million 0.0 0.0 0.0 633.7 633.7 163.8 0.0 0.0 0.0 0.0 0.0 1431.1
Total operating cost HK$ million 0.0 0.0 0.0 60.8 140.7 164.7 165.4 166.0 166.7 167.3 168.0 1199.5
NOx Produced Tons 23,897 24,499 25,104 25,716 18,750 19,375 20,005 20,641 21,260 21,907 22,561 243,715
Capital Cost HK$ million 0.0 0.0 0.0 0.0 475.3 0.0 0.0 0.0 0.0 0.0 0.0 475.3
Total operating cost HK$ million 0.0 0.0 0.0 0.0 96.7 96.7 96.7 96.7 96.7 96.7 96.7 676.9
RSP Produced Tons 1492 1391 1104 681 355 365 376 386 397 407 418 7373
Total operating cost HK$ million 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
e n
v i
r o
n m
e n
t a
l s
c i
e n
c e
&
p
o l
i c
y 3
1 (
2 0
1 3
) 1
– 1
21
0
Table 6 – The total emissions, capital cost and operating costs for Company A (2006–2016) for all four options.
Item (unit) SO2 produced(tons)
NOx produced(tons)
RSP produced(tons)
Capital expenditure(HK$ million)
Total operatingcost (HK$ million)
Option 1 538,051 296,811 14,074 0.0 1695.2
Option 2 364,166 270,263 10,645 871.3 1441.8
Option 3 218,987 270,263 7818 1505.0 1366.9
Option 4 197,764 243,715 7373 1906.4 1876.4
0.0
500.0
1,000 .0
1,500 .0
2,000 .0
2,500 .0
0
100,000
200,000
300,000
400,000
500,000
600,000
Op�on 1 Op �on 2 Op �on 3 Op �on 4
Expe
nses
(HK$
mill
ion)
Emis
sion
s (to
ns)
Total Emissions and E xpenses fo r 4 Op�ons for 10 Ye ars
SO2 NOx RSP Capital Expend iture
Fig. 6 – The total emissions, capital cost, and operating costs for Company A, 2006–2016, for all four options.
e n v i r o n m e n t a l s c i e n c e & p o l i c y 3 1 ( 2 0 1 3 ) 1 – 1 2 11
7.2. Summary of results for the four options from 2006 to2016
Based on the simulation results for the four options, the total
operating cost of the equipment, the total capital investment,
and the total emissions of SO2, NOx, and RSP from 2006 to 2016
are summarized in Table 6.
These results are plotted in Fig. 6.
8. Discussion and conclusion
According to the objective function graph in Fig. 6, the total
operating cost decreases from Option 1 to Option 3 and
increases in Option 4. The operating cost of Option 3 is the
lowest among the four. Option 3 produces a reduction of about
60% in SO2 and 9% in NOx compared with Option 1. The capital
costs increase from Option 1 to Option 4 as more emission-
reduction equipment is installed. In Option 4, both the capital
and operating costs are higher than those of Option 3, but the
reduction in emissions is not significant. Hence, Option 3 is the
most reasonable choice.
In early 2005, Company A proposed (1) to commission new
combined cycle power plants in 2005 and 2006 and (2) to
retrofit the four 677 MW coal-fired units with FGD and SCR
plants to reduce atmospheric emissions (EDLB, 2005). The
company announced that it had achieved its environment-
related goals. The company recorded a material reduction of
60% for SO2, NOx, and RSP. Therefore, the company’s emission
control efforts are consistent with the recommendation of
this study.
With the market forces and the flexibility of emission
trading, power plants can use this emission trading model
proactively to work out different cost-effective emission
reduction strategies to benefit the whole region. In reviewing
the trade or no trade decision, power stations could use the
model as reference guides to monitor the effectiveness of their
emission reduction strategies so that they could make proper
and timely investment decisions in reducing emission. The
emission trading model can be used to simulate various
business alternatives for power stations with multi-fuel
operations to meet the stringent emission controls at the
lowest operating costs. The research can also be extended to
analyze a complicated multivariable and multi-constraint
scenario for different large electricity companies with multi
power stations and plants. This in term would facilitate the
implementation of an effective emissions trading scheme in a
region.
Leung et al. (2009) commented that simplicity, flexibility,
periodic review, and broad participation are the important
factors for the successful implementation of an emission
trading scheme. The government policy makers can use this
emission trading model to evaluate the effectiveness of the
emission trading scheme in a region. With the simplicity of the
model, many different scenarios, including multi-power
stations and plants in a region, can be simulated to provide
useful information for policy makers to assess the practicality
of the emission trading scheme. This would help them to
formulate realistic emission trading strategies for the whole
region.
While the emission trading simulation model can provide
results that are difficult to be experimentally measurable, yet,
e n v i r o n m e n t a l s c i e n c e & p o l i c y 3 1 ( 2 0 1 3 ) 1 – 1 212
in applying it to the complicated emission trading scenario,
the model is required to be interpreted with cautions. The
simulation of the emission trading model involves the
interaction of a number of variables; the relationship of these
variables could be quite complicated. In a steady electricity
growth situation, the linear functions could be approximated.
However, under a highly fluctuated environment, the adop-
tion of the model may require special care. To enhance the
model, future research could try to investigate situations
where non-linear functions may apply.
r e f e r e n c e s
Alstom Power, 2011a. Key reference of Belchatow pulverizedcoal steam unit of 858 MW, website of Alstom Power.
Alstom Power, 2011b. KA26 combined cycle power plant,website of Alstom Power.
Census and Statistics Department, HK, 2007. Statistical data onGDP from census and statistics department of HKSAR,website of the Census and Statistics Department of HKSAR.
CH2M HILL (China) Limited, 2002. Final report, Agreement No.CE 06/98. Study of air quality in the pearl river delta region,CH2M Hill (China) Limited in association with ENSRInternational, Inc.; Rowan William Davies and Irwin, Inc.;Guangdong Province Environmental Monitoring Centre;Guangzhou Research Institute of Environmental Protection.
CLP Power (CLP), 2007. CLP group – about us, website of CLP.Eastop, T.D., McConkey, A., 1986. Applied Thermodynamics for
Engineering Technologists. Longman Group, UK, 232–242.EC Directive, 2003. Directive 2003/87/EC of the European
Parliament and of the Council of 13th October 2003establishing a scheme for greenhouse gas emissionallowance trading within the Community and amendingCouncil Directive 96/61/EC, Official Journal of the EuropeanUnion, 25th October 2003 Edition.
Economic Development and Labour Bureau (EDLB), HKSAR,2005. The CLP Power Hong Kong Limited Castle Peak PowerCompany Limited 2005 Financial Plan, the LegislativeCouncil Panel on Economic Services, Legislative Council,Hong Kong.
Electrical and Mechanical Services Department (EMSD) ofHKSAR, 2011. HK RE Net – overview, website of HK RE Net,EMSD, HKSAR.
Environment Bureau (EB HK), HKSAR, 2008a. The Scheme ofControl Agreement between HKSAR and CLP, 2008, websiteof the Environment Bureau, HKSAR.
Environment Bureau (EB HK), HKSAR, 2008b. The Scheme ofControl Agreement between HKSAR and HEC, 2008, websiteof the Environment Bureau, HKSAR.
Environmental Protection Department (EPD HK), HKSAR, 2005.Press Release – Power Station’s Licence Renewed withEmission Caps, website of Environment ProtectionDepartment, HKSAR.
Environmental Protection Department (EPD) of HKSAR, 2007a.Implementation Framework of the Emission Trading PilotScheme for Thermal Power Plants, website of EnvironmentalProtection Department, HKSAR, website of EnvironmentProtection Department, HKSAR.
Environmental Protection Department (EPD) of HKSAR, 2007b.Press Release – Power Station’s Licence Renewed withTightened Emission Caps, website of EnvironmentProtection Department, HKSAR.
Environmental Protection Department (EPD) of HKSAR, 2011.Climate Change – Electricity Generation by Fossil Fuel, thewebsite of Environmental Protection Department, HKSAR.
GB China (GB), 1996. GB13223-1996, Emission Standard on AirPollutants for Thermal Power Plants. National Standard ofthe People’s Republic of China, GB.
General Electric Energy (GE Energy), 2011. General ElectricEnergy on Advanced Technology of Combined Cycle, thewebsite of General Electric Energy.
HK Government news (Hong Kong SAR press release), 2002. JointStatement on Improving Air Quality in the Pearl River DeltaRegion, website of the HKSAR.
Hong Kong Observatory (HKO), 2007. Meteorological Statistics –Number of hours of Reduced Visibility (Hong KongObservatory), website of the Hong Kong Observatory.
Law of PRC, 2000. Law of the People’s Republic of China on thePrevention and Control of Atmospheric Pollution, 2000, Lawsof the People’s Republic of China; adopted 1995; effective 2000.
Leung, D.Y.C., Daniel, Y., Amanda, N., Leung, M.K.H., Alan, C.,2009. An overview of emissions trading and its prospects inHong Kong. Environmental Science & Policy 12 (February (1))92–101.
Liao, S., 2002. Speech in Legislative Council on EmissionsTrading. In: Minutes of Legislative Council Meeting on 6thNovember 2002, Legislative Council, Hong Kong.
Ordinance HK (Laws of Hong Kong), 1997. Air Pollution Control(Specified Processes) Regulations, Hong Kong Ordinance,Version date: 30 June 1997.
Regulation HK (Laws of Hong Kong), 2007. Chapter 311, AirPollution Control Ordinance, website of the Department ofJustice, HKSAR.
Takahashi, M., 1997 August. Technologies for reducingemissions in coal-fired power plants. In: Energy Issue 3,World Bank.
Regulation SZ (Environmental Protection Department ofShenzhen), 1998. Regulation on Controlling Acid Rain andEmission of sulfur Dioxide in Shenzhen.
Tavoulareas, E.S., Charpentier, J.-P., 1995. World Bank, CleanCoal Technologies for Developing Countries. In: The WorldBank Technical Paper Number 286.
The Power Assets Holdings Limited (PAH), 2011. Power AssetsHolding, website of Power Assets.
Trade Economics, 2012. China Inflation Rate, website of TradeEconomics.
US Department of Energy and Southern Company Services, Inc.,1997. Control of Nitrogen Oxide Emission: Selective CatalyticReduction (SCR), website of the US Department of Energy, USGovernment.
US Energy (US Energy Information Administration), 2005.Official Energy Statistics from the U.S. Government, websiteof Official Energy Statistics from the US Government.
US EPA, 2006. Acid Rain Program SO2 Allowance Fact Sheet,website of the Office of Air and Radiation Clean Air MarketDivision.
US EPA, 2007. Electrostatic Precipitators, website of the US EPA,US Government.
US EPA, 2011. Emission Factors & AP42, website of the US EPA,US Government.
World Bank, 2007. China – Shandong Flue Gas DesulfurizationProject (2007), website of the World Bank.
Richard C.M. Yam is an associate professor in the Department ofSystem Engineering and Engineering Management at City Univer-sity of Hong Kong. His current research interests are in the areas ofproduct innovation and technology management.
W.H. Leung is a professional engineer who has worked in a HongKong power station for over 30 years. He is also an engineeringdoctoral student at the City University of Hong Kong. His currentresearch interests are in the areas of environmental protectionand engineering management.