INTEGRATED EXPANSION STRATEGIES FOR PUBLIC CHARGING ...
Transcript of INTEGRATED EXPANSION STRATEGIES FOR PUBLIC CHARGING ...
©Fraunhofer ISE/Foto: Guido Kirsch
© Fraunhofer ISE
INTEGRATED EXPANSION STRATEGIES FOR PUBLIC CHARGING INFRASTRUCTURE IN CITIES
3rd E-Mobility Power System Integration Symposium
Matti Sprengeler
Research associate and Ph.D. candidate
Fraunhofer Institute for Solar Energy Systems ISE
Group Smart Cities
Dublin, October 14th 2019
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
2
AGENDA
Motivation
Methodology
Case study: The Pfaff area
Results
Conclusion
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
3
Motivation
Cities facing numerous transport-related challenges
E.g. noise pollution, climate change, traffic-induced smog
One element to tackle these challenges is electric mobility
Total share of electric vehicles (EVs) is low in the EU countries
Several issues are discouraging people to purchase an EV, i.a. the absence of charging possibilities
A demand covering charging infrastructure (CI) is necessary
The following questions need to be answered:
How much charging infrastructure is needed?
Where should charging points be placed to serve the demand?
Are local distribution grids prepared for a considerable increase of electric mobility?
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
4
Methodology – Overview
Estimation of traffic volume
Quantification of charging
infrastructure demand
Location planning of
charging infrastructure
Examination of electrical loads and power grid
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
5
Methodology
Traffic volume by user group and economical use
Residential, employee, customer, economic EVs
E.g. offices, restaurants, retail sale, housing …
Bosserhoff method and mobility behaviour data
Quantification of CI demand on a district level
Method: queuing theory
Assumption: all points in area within walking distance
Service quota: 85 %
Normal and fast charging, separately
Location according to utility analysis
Points of interest (POIs) are scored acc. to number of vehicles, frequency and length of stay
Score assigns number of CPs to POIs proportionately
Fig.: Queuing theory
Fig.: Normal and fast charging potential
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
6
Methodology – Electrical loads and power grid
Power grid overload: transformer stations
Electrical loads, e.g. from buildings and local electricity generation are considered
Probabilistic driving profiles generated from mobility behaviour data using the tool synPRO
Model is formulated as a linear program (LP)
State of charge (SoC) of each EV battery
Charging management transformer station-wise
Different charging strategies
Restriction: EVs fully charged at departure (if possible)
Most critical days of the year are examined
Fig.: Electrical loads at different charging strategies
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
7
Case study – Pfaff area
The area is located in Kaiserslautern, Germany
Formerly a production site for sewing machines
Revitalization process of the industrial wasteland will last until 2029
Ongoing area development project shall create a smart and climate neutral quarter with support by the research project EnStadt:Pfaff
Achievement of goals through innovations
digitisation
energy
buildings
mobility
© astoc/mess (Images)
Figs.: Vision of the Pfaff area in 2029
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
8
Case study – Characteristics of the area
Area is situated in 1 km distance to city centre
Mixed-used quarter
Area of approx. 19 ha
1,500 inhabitants
2,700 employees
Gross floor area housing: 41.000 m²
Fig.: Use of the area
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
9
Case study – Traffic volume
Fig.: Parking space occupancy
Fig.: Traffic originating and terminating in the area per day
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
10
Case study – Public charging infrastructure
Assumption:
Share of EVs is 30% in 2029 (constant over the day)
Public CI based on customer traffic
Normal CI demand higher than fast CI demand (64 vs. 24)
CI demand in building plot III higher than in VII
Fig.: Charging station demand
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
11
Case study – Power grid
The area is served by six transformer stations
Every transformer station consists of two transformers with an output of 630 kVA each
There is no overload, even without a charge management
EVs increase the peak load by up to 10.7 %
Using a charge management, the peak load increases by up to 2.1 %
High peak shaving potential by controlled charging
I. Yearly peak load without EV charging
II. Yearly peak load with uncontrolled EV charging
III. Yearly peak load with controlled EV charging
IV. Relative increase by uncontrolled EV charging
V. Relative increase by controlled EV charging
VI. Relative peak shaving potential by controlled charging
Fig.: KPIs concerning power grid
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
12
Conclusion
CI demand and location planning
64 normal CPs and 24 fast CPs are necessary to serve 85% of customer EVs
Eq. to 9.35 EVs per CP
Besides the number of EVs per day, frequency and length of stay over the day are decisive
Power grid
Load increase of up to 10.7 % by EV charging, with charge management up to 2.1 %
No overload resulting
Overload is a matter of the power grid dimensioning
On average 5.7 % peak shaving potential
Case-by-case analysis is necessary
The presented method is capable of quantifying the CI demand, identifying demand serving CP locations as well as avoiding grid overload issues by smart planning of the CI expansion
R 23 G 156 B 125
R 242 G 148 B 0
R 31 G 130 B 192
R 226 G 0 B 26
R 177 G 200 B 0
R 254 G 239 B 214
R 225 G 227 B 227
© Fraunhofer ISE
13
THANK YOU VERY MUCH FOR YOUR ATTENTION!
Matti Sprengeler
Research associate and Ph.D. candidate
Mail: [email protected]
Phone: +49 761 4588 5455
Fraunhofer Institute for Solar Energy Systems ISE
Group Smart Cities