Plug-in Electric Vehicle Charging Demand Forecasting, Modeling and Facility Planning --On seminar of V2G-Sim team, LBNL

Hongcai Zhang, PhD Candidate SGOOL, Tsinghua University eCAL, University of California, Berkeley [email protected] February 12, 2016

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

1

Content

Introduction of SGOOL Electric Vehicle in China Previous Research Work Future Research Plan

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

2

SGOOL o Smart Grid Operation and Optimization Laboratory (SGOOL) n

Founded in Sept. 2009 at Department of EE, Tsinghua University

n

Led by Prof. Yonghua Song Prof. Yonghua Song Fellow of the Royal Academy of Engineering (UK) Fellow of IEEE (USA) Fellow of IET (UK) Professor and Pro-Vice-Chancellor in Brunel University and University of Liverpool (UK) Professor of Tsinghua University (PRC) Executive Vice-President of Zhejiang University (PRC)

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

3

Faculty o 1 professor, 1 associate professor, 1 lecturer, 2 postdoctoral fellows

Prof. Yonghua Song, director Fellow of RAE, IEEE and IET Executive Vice-President of Zhejiang University

Dr. Zechun Hu Associate Professor

Dr. Jin Lin Lecturer, Tenure-track

Research Interests: Electric Vehicle, Energy Storage

Research Interests: Renewable Energy Integration and Control, Active Distribution Network

Research Interests: Power System Security and Optimization, Electricity Markets, Energy Internet

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

4

Student o >20 graduate students

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

5

People affiliated with UCB o 2 exchange PhDs, 1 master o Hosted 1 exchange PhD from UCB (Froylan Sifuentes, ERG)

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

6

Research in SGOOL

Electricity market, energy policy

Electric vehicle, energy storage system

Active distribution network

Renewable energy operation & control

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

7

Hongcai Zhang, PhD Candidate Education Visiting student researcher, UC Berkeley, Jan. 2016-Dec. 2016 (expected) PhD in EE (Prof. Yonghua Song), Tsinghua Univ., Aug. 2013-Jul. 2018 (expected) B.E. in EE (graduation with honors), Tsinghua Univ., Aug. 2009-Jul. 2013

Research Interest PEV load forecasting & modeling, grid integration, charging facility planning

Selected Publication [1] H. Zhang, W. Tang, Z. Hu, Y. Song, et. al., “A Method for Forecasting the Spatial and Temporal Distribution of PEV Charging Load,” in Pro. IEEE Power & Energy Soc. Gen. Meeting, 2014, pp. 1–5. [2] H. Zhang, Z. Hu, Z. Xu, and Y. Song, “Evaluation of Achievable Vehicle-to-Grid Capacity Using Aggregate PEV Model,” Submitted to IEEE Transactions on Power Systems. (under review) [3] H. Zhang, Z. Hu, Z. Xu, and Y. Song, “Optimal Planning of PEV Charging Station with Single Output Multiple Cables Charging Spots,” to appear in IEEE Transactions on Smart Grid. [4] H. Zhang, Z. Hu, Z. Xu, and Y. Song, “An Integrated Planning Framework for Different Types of PEV Charging Facilities in Urban Area,” to appear in IEEE Transactions on Smart Grid. ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

8

Content

Introduction of SGOOL Electric Vehicle in China Previous Research Work Future Research Plan

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

9

PEV market in China is booming o 447,200 sold since 2011 through 2015 o Goal: 500 thousand by 2015, 5 million by 2020 o PEV charging load by 2020: ~40 TWh/year 331092

350000 300000 250000 200000 150000 78499

100000 50000

8159

12791

17642

2011

2012

2013

0 2014

2015

EV sales volume China* EV sales volume ininChina*

Best EV sellers in China

*Data source: China Association of Automobile Manufacturers ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

10

China’s ambitious goal to promote PEV charging facility o 49,000 public spots, 3,600 charging/swapping stations deployed o 4.8 million charging spots by 2020 o 12 thousand fast charging/swapping stations by 2020

Charging facility operators/investors in China

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

11

SGOOL’s research on plug-in electric vehicles On system operation

Temporal distribution

Integration of EVs into power system

Temporal & spacial distribution

Control Objectives

On system planning

Impacts of EV charging on power systems

Coordinated charging strategies

Control strategies of V2G

Peak shaving and valley filling Frequency regulation Providing reserve With renewable generation

Charging demand forecast Optimal planning of charging facilities Urban area

Operation of charging facitlities and market mechanism

Highway network

Research framework of plug-in electric vehicles in SGOOL

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

12

Content

Introduction of SGOOL Electric Vehicle in China Previous Research Work Future Research Plan

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

13

Plug-in electric vehicle charging demand forecasting, modeling and facility planning o R1. PEV charging demand forecasting considering spatial and temporal distribution o R2. Aggregated V2G (charging/discharging) model and flexibility evaluation of large-scale PEV fleet o R3. Station level charging facility planning: planning of a charging station equipped with SOMC spots o R4. System level charging facility planning: integrated planning of various types of charging facilities in urban area o Co-authors: Prof. Yonghua Song, Prof. Zechun Hu, Zhiwei Xu

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

14

Plug-in electric vehicle charging demand forecasting, modeling and facility planning o R1. PEV charging demand forecasting considering spatial and temporal distribution o R2. Aggregated V2G (charging/discharging) model and flexibility evaluation of large-scale PEV fleet o R3. Station level charging facility planning: planning of a charging station equipped with SOMC spots o R4. System level charging facility planning: integrated planning of various types of charging facilities in urban area

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

15

Challenges (1) o PEV load forecasting without recorded PEV load data n

Evaluation of PEV charging load’s effects on power grid

n

Verify effectiveness of grid-integration technologies

n

PEV charging facility planning

Travel survey data of internal combustion engine vehicles*

Land use of Longgang District, 2020

*A. Santos, N. McGuckin, H. Y. Nakamoto, D. Gray, and S. Liss, “Summary of travel trends: 2009 national household travel survey,” U.S. Dept. Transp., Federal Highway Admin., Washington, DC, USA, Tech. Rep. FHWA-PL-ll-022, 2011. ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

16

R1. Charging demand forecasting o Framework of the PEV charging demand forecasting method Parking generation rate method

Temporal distribution of PEV parking number*

Distribution of parking places (land use data)

Parking and driving behaviors

Temporal distribution of PEV arrival number*

Spatial & temporal distribution of PEV parking

Distribution of PEV parking duration*

Dynamic process of PEV parking and driving Monte-Carlo simulation

Spatial & temporal distribution of PEV charging demands *A. Santos, N. McGuckin, H. Y. Nakamoto, D. Gray, and S. Liss, “Summary of travel trends: 2009 national household travel survey,” U.S. Dept. Transp., Federal Highway Admin., Washington, DC, USA, Tech. Rep. FHWA-PL-ll-022, 2011. ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

17

R1. Charging demand forecasting o Framework of the PEV charging demand forecasting method

©Hongcai Zhang

No.

Arrive time

Departure time

Arrive SoC

Leave SoC

Power

Type

Location

1

18:00

22:00

0.4

0.9

6.6

H

Block 1

















SGOOL, Tsinghua and eCAL, UC Berkeley

18

R1. Charging demand forecasting o Case studies in Longgang District, Shenzhen n

195.88 km2 with a PEV population of 16,000 predicted in 2020.

Land use planning of Longgang District, 2020

©Hongcai Zhang

PEV charging load (uncoordinated charging scenario)

SGOOL, Tsinghua and eCAL, UC Berkeley

19

R1. Charging demand forecasting o Simulation software for PEV charging demand forecasting*

* Developers: Zechun Hu, Hongcai Zhang, Xiaoshuang Chen, Zhiwei Xu, Guannan Qu, Xiang Lin, Jiangwei Yang ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

20

Plug-in electric vehicle charging demand forecasting, modeling and facility planning o R1. PEV charging demand forecasting considering spatial and temporal distribution o R2. Aggregated V2G (charging/discharging) model and flexibility evaluation of large-scale PEV fleet o R3. Station level charging facility planning: planning of a charging station equipped with SOMC spots o R4. System level charging facility planning: integrated planning of various types of charging facilities in urban area

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

21

Challenges (2) o Model a fleet of PEVs/batteries as simply as a single grid-scale storage system*

A fleet of individual PEVs/batteries Low capacity, hard to forecast and control

Grid-scale storage High capacity, easy to forecast and control

* Z. Xu, Z. Hu, Y. Song, and J. Wang, “Risk-Averse Optimal Bidding Strategy for Demand-Side Resource Aggregators in DayAhead Electricity Markets Under Uncertainty,” IEEE Trans. Smart Grid, pp. 1–10, 2015. ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

22

R2. Charging/discharging modelling o Individual PEV model n

Use energy & power boundaries to model PEV charging demand

Controllable charging demand

Uncontrollable charging demand

* Z. Xu, Z. Hu, Y. Song, and J. Wang, “Risk-Averse Optimal Bidding Strategy for Demand-Side Resource Aggregators in DayAhead Electricity Markets Under Uncertainty,” IEEE Trans. Smart Grid, pp. 1–10, 2015. ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

23

R2. Charging/discharging modelling o Aggregated PEV fleet model E

P t t Energy boundary

n

Power boundary constraints

©Hongcai Zhang

Power boundary

n

Energy boundary constraints

n

Power injected to the grid

SGOOL, Tsinghua and eCAL, UC Berkeley

24

R2. Charging/discharging modelling o Aggregated charging/discharging parameter forecasting No.

Arrive time

Leave time

Arrive soc

Leave soc

power

1

18:00

22:00

0.4

0.9

6.6

2

15:00

17:00

0.6

0.8

3.3

3

11:00

14:00

0.7

0.6

6.6













N

19:00

07:00

0.3

1

6.6

E

P

t

Individual charging demand D1

D2

t

Storage-like aggregate model

D3



Dm-1

Dm

P t

P t

E

P

P t

t

E t E

E

t

t

t

Storing and forecasting aggregated energy and power boundaries ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

25

R2. Charging/discharging modelling o Implementation in optimal charging and discharging scheduling Electricity Battery costs degradation

Storagelike constraints

Flexibility income

E

t Energy boundary

P t

flexibility

Power boundary ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

26

R2. Charging/discharging modelling o Laxity-SoC-based heuristic smart charging strategy n

Charging laxity*:

* A. Subramanian, M. J. Garcia, D. S. Callaway, K. Poolla, and P. Varaiya, “Real-time scheduling of distributed resources,” IEEE Transactions on Smart Grid, vol. 4, no. 4, pp. 2122–2130, 2013. ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

27

R2. Charging/discharging modelling o Case studies & performance evaluation 1: accuracy n

5000 Leafs, V2G power: 50% 6.6kW, 50% 3.3kW, PJM, 214 days

Average power profiles

Average reserve profiles

n

Average shortage ratio, < 2%

n

Outliers, < 8%

Box-plot of reserve shortage ratio ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

28

R2. Charging/discharging modelling o Case studies & performance evaluation 2: benefit n

5000 Leafs, V2G power: 50% 6.6kW, 50% 3.3kW, PJM, 214 days

Average reserve profiles

Energy Average power profiles

Benefit Comparison Utilizing different strategies

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

n

Gross revenue, 92.7%

n

Benefit, 96.2%

29

Plug-in electric vehicle charging demand forecasting, modeling and facility planning o R1. PEV charging demand forecasting considering spatial and temporal distribution o R2. Aggregated V2G (charging/discharging) model and flexibility evaluation of large-scale PEV fleet o R3. Station level charging facility planning: planning of a charging station equipped with SOMC spots o R4. System level charging facility planning: integrated planning of various types of charging facilities in urban area

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

30

Challenges (3) o Enhance the utilization of destination charging spots

A residential parking lot

A community in China

©Hongcai Zhang

A PEV’s parking and charging process in a public parking lot

SGOOL, Tsinghua and eCAL, UC Berkeley

31

R3. Planning of an SOMC station o Smart charging can affect charging facility demands

A schematic diagram of two PEVs sharing one charging spot

©Hongcai Zhang

Manual intervention is inconvenient

SGOOL, Tsinghua and eCAL, UC Berkeley

32

R3. Planning of an SOMC station o Single-output-multiple-cables (SOMC) charging spot n

High utilization, low costs for destination charging

*Chung, Ching-Yen. “Electric Vehicle Smart Charging Infrastructure”, University of California, Los Angeles, 2014. ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

33

R3. Planning of an SOMC station o Two-stage stochastic programming planning model

n Objective: investment + operation costs n Variables: spot number (N) and cable number (O)

Coordinated charging mechanism

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

34

R3. Planning of an SOMC station o Case studies: 100 PEVs, 6.6 kW spot, residential parking lot, charge 1 time/day, TOU electricity costs Planning results

Charging power profiles of PEVs ©Hongcai Zhang

Number of PEVs getting recharged

SGOOL, Tsinghua and eCAL, UC Berkeley

35

Plug-in electric vehicle charging demand forecasting, modeling and facility planning o R1. PEV charging demand forecasting considering spatial and temporal distribution o R2. Aggregated V2G (charging/discharging) model and flexibility evaluation of large-scale PEV fleet o R3. Station level charging facility planning: planning of a charging station equipped with SOMC spots o R4. System level charging facility planning: integrated planning of various types of charging facilities in urban area

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

36

Challenges (4) o Minimize the social costs for providing charging services n

Fast charging station + destination charging spot

8 gasoline stations around Tsinghua ©Hongcai Zhang

621 parking lots around Tsinghua

SGOOL, Tsinghua and eCAL, UC Berkeley

37

R4. Integrated planning of various types of charging facilities o The substitution effect between different facilities Characteristic comparison of different types of charging facilities type

Power

Private spot (HCS1)

Investment costs

User costs

facility

Grid

Land

Buildings

Electricity

Time

Low













Residential public spot (HCS2)

Normal













Other public spot (PCS)

Normal













Fast-charging station (FCS)

High













Distributed charging spot ©Hongcai Zhang

Centralized fast-charging station

SGOOL, Tsinghua and eCAL, UC Berkeley

38

R4. Integrated planning of various types of charging facilities o The substitution effect between different facilities Characteristic comparison of different types of charging facilities type

Power

Private spot (HCS1)

Investment costs

User costs

facility

Grid

Land

Buildings

Electricity

Time

Low













Residential public spot (HCS2)

Normal













Other public spot (PCS)

Normal













Fast-charging station (FCS)

High













Costs related to FCS ©Hongcai Zhang

Total costs of the charging system

SGOOL, Tsinghua and eCAL, UC Berkeley

39

R4. Integrated planning of various types of charging facilities o Case studies in Longgang District, Shenzhen n

195.88 km2 with a PEV population of 16,000 predicted in 2020. Scenarios of the planning

Planning results of FCSs Planning results

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

40

Content

Introduction of SGOOL Electric Vehicle in China Previous Research Work Future Research Plan

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

41

By 2050, renewable energy could meet more than 60% of China’s primary energy demand

*Data source: Energy Research Institute, National Development and Reform Commission of China ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

42

Integration of renewable energy in China is suffering serious challenges o Wind, 105 GW (Jun.30, 2015); Solar, 38 GW (Sept.30, 2015) Wind Power, Jan.-Jun., 2015 Curtailed, 17.5 TWh, 15%

Solar Power, Jan.-Sep., 2015 Curtailed, 3.0 TWh, 10%

Integrat ed, 97.7 TWh, 85%

In Heilongjiang Province, 50% curtailed

Integrat ed, 27.6 TWh, 90%

In Gansu Province, 28% curtailed

*Data source: National Energy Administration of China ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

43

Future research plan o PEV charging system planning considering integration of renewable resources IV. Charging system operation

III. Charging network planning

Centralized PVs

Charging network

Centralized Winds

Charging demand forecasting

Charging station

Distributed PVs

Charging demand modelling

II. Charging station planning

Distributed storage

I. Analysis & modeling

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

44

PEV charging system planning considering integration of renewable resources o Passively satisfy demands n

initiatively utilize flexibility

Trade-off between investment costs and operation revenue (flexibility)

o Interdisciplinary constraints: transportation + power grid

EV charging station*

EV charging network

* Y.-T. Liao and C.-N. Lu, “Dispatch of EV Charging Station Energy Resources for Sustainable Mobility,” IEEE Trans. Transp. Electrif., vol. 1, no. 1, pp. 86–93, 2015. ©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

45

Thank You!

©Hongcai Zhang

SGOOL, Tsinghua and eCAL, UC Berkeley

46

Feb-12-2016 Plug-in Electric Vehicle Charging Load Forecasting ...

Feb-12-2016 Plug-in Electric Vehicle Charging Load Forecasting, Modeling and Facility Planning.pdf. Feb-12-2016 Plug-in Electric Vehicle Charging Load ...

14MB Sizes 2 Downloads 57 Views

Recommend Documents

Long-Term Load Forecasting of Jordanian - ijeecs.org
May 1, 2013 - natural resources, his imports from energy represent around ... [3] . (ii) Nonlinear models such as Artificial. Neural Networks (ANN) [4], Support Vector ..... Load Forecasting, Renewable Energy, Nuclear Energy, Economic and.

Long-Term Load Forecasting of Jordanian - ijeecs.org
May 1, 2013 - 95% of his consumption [2] , around every ten years has a waves of refugees come in, and with the big fluctuation in energy prices, long-term load forecasting represent really a challenging ... 1; one input layer (source nodes with ....

[PDF] Advancements in Electric and Hybrid Electric Vehicle Technology
Electric Vehicle Technology (S P (Society of Automotive ..... hyperlinks last name of 1st author authors without affiliation title subtitle series ed year pages arabic.

The Load Flow Calculation in Unbalanced Radial Electric Networks ...
one source node (infinite bus), to which the specified quantities are the components of voltage, and the unknown quantities are the components of loads.

ePUB The Bhutan Electric Vehicle Initiative
(Directions in Development) Da Zhu Read Online. 1 I celebrate .... ongoing improvements in technology and resulting cost decreases in the global market.This.