Synergy of Systems Combining Plug-in Hybrid Electric Vehicles (PHEV) with Electricity Grid: Potential Effects on Energy Savings

and CO2 Reduction

Imperial College London The University of Tokyo 2nd Joint Symposium on Innovation in Energy Systems Sept. 24, 2009, @The University of Tokyo

Kenji Yamaji School of Engineering, CEE, and APET, The University of Tokyo

Related Papers
K. Yamaji, A. Hashimoto, H. Yamamoto, R. Hiwatari, and K. Okano: An Analysis of Market Potential and CO2 Mitigation Effects of Plug-in Hybrid Electric Vehicle in Japan, IAEE International Conference, Istanbul, June 2008 M. Takagi, H. Yamamoto, K. Yamaji, Evaluation of Expanded Allowable Capacity of Wind Power in Power Systems by Charge Control for Plug-in Hybrid Electric Vehicles, IEEJ Trans. PE, Vol. 128, No. 12, pp.1513-1521, 2008 (in Japanese) K. Yamaji, M. Takagi, and H. Yamamoto: Synergies in Energy System: A Case Study of Plug-in Hybrid Electric Vehicle and Wind Power Combination, GCOE Symposium, University of Tokyo, January 2009 S. Nakaue, H. Yamamoto, K. Yamaji, M. Takagi, and A. Hashimoto: Evaluation of Plug-in Hybrid Electric Vehicle Considering Driving Pattern and Optimal Power Generation Mix, 5th Dubrovnik Conference on Sustainable Development of Energy, Water, and Environment Systems, Dubrovnik, September, 2009
2

Two Case Studies in Energy Systems: the order is reversed from the abstract

First Case Study: PHEV-Power Sector Combination in reducing CO2 emissions Second Case Study: PHEV-Wind Power Combination in promoting renewable energies
3

First Case Study:

PHEV and Generation Mix Change

1. Driving Characteristics of Passenger Cars 2. Economics and Market Diffusion of PHEV 3. Electricity Demand for PHEV and Optimal Generation Mix 4. Gasoline Demands and CO2 Emissions 5. Conclusion of the First Case Study
4

Importance of Transport Sector in Energy Systems

- Transport sector represents around one fourth of total energy demand - Oil products almost monopolize the energy for transport - Thus, transport sector is most challenging area for CO2 mitigation Alternatives to Reduce CO2 Emissions from Passenger Cars - Improvements of fuel economy of ICE vehicle (ICV) - Hybrid Electric Vehicle (HEV) - Fuel Cell Vehicle (FCV) - Electric Vehicle (EV) - Plug-in Hybrid Electric Vehicle (PHEV): strategic importance - Biofuels
5

Hybrid Electric Vehicle (HEV)

motor/ generator

Plug-in Hybrid Ele. Vehicle (PHEV)

motor/ generator engine

Electric Vehicle EV)

motor/ generator

engine

battery

battery

battery

fuel tank

fuel tank

Outline of Method
Around 57.1 million passenger cars in Japan in future as well; 571 cohorts with a cohort of 100,000 cars which move in a same driving pattern Categorize the cars into several groups of specific driving patterns Setting the capacity of battery for PHEV through the analysis of relative fuel economy of PHEV to that of HEV Estimate of market share PHEV in future on the basis of the assumptions of the costs of battery and PHEV components Estimate the electricity demand for charging battery of PHEV Estimate the increase of CO2 emission for charging battery through calculating optimal generation mix taking into account the modified load profile by the introduction of PHEV Evaluate the total CO2 emissions from passenger cars including both direct emissions from fuel combustion and indirect emissions through power generation Results are compared with the cases without PHEV introduction Time horizon for the analysis is 2005-2050
7

Daily driving pattern ~OD survey~

Business

Leisure

Commuting

Source:

Categorization of Passenger Cars

Passenger cars in Japan 57,100,000 For commuting purpose

No
For business purpose

Yes
Cars for commuting 25,200,000 Used for weekend commuting

Yes
Cars for business 3,700,000 Used for weekend business

No

No
Cars for commuting only on weekdays 19,100,000

Yes
Cars for commuting on weekdays and weekends 6,100,000 Cars for business only on weekdays 2,900,000

No

Yes
Cars for business on weekdays and weekends 800,000 Cars not used for commuting nor business 28,200,000

Using a cohort of hundred thousand cars as a unit, passenger cars are represented by 571 cohorts. A cohort moves in a same driving pattern.
Source: 9

Characteristics of three groups

Commuting strong regularity
Trip length and the timings of return to home stay constant throughout the year.

Business and Leisureirregularity

Daily trip length and the timings of return to home are random for every cohort. Trip number per day for trip lengths that it makes corresponds with OD survey.
cohort No. travel pattern 273 274 275 276 277 278 279 280 281 282 Co mmu tin g Le isu re Co mmu tin g Le isu re B u sin e ss Le isu re Co mmu tin g B u sin e ss Le isu re Co mmu tin g 1 Mon. A B 16 35 0 17 25 18 14 20 6 2 .5 19 35 18 7 18 35 17 2 2 .5 16 15 2Tue. A B 0 0 17 25 16 15 0 0 18 7 17 15 0 16 15 3Wed. A B 16 35 0 17 25 19 14 10 14 0 18 7 15 7 0 16 15 day 4Thu. A B 0 15 1 7 .5 17 25 0 0 16 2 2 .5 18 7 16 30 0 16 15 5Fri. A B 16 35 0 17 25 0 12 1 7 .5 20 4 18 7 18 3 7 .5 0 16 15 6Sat. A B 16 35 0 0 23 100 17 30 0 22 75 0 18 5 2 .5 0 7Sun. A B 16 35 0 0 0 18 70 18 25 0 0 0 19 35

Atime of coming home or office (hour) Bdaily driving range (km)

Source:

10

Distribution of Driving Length

Using the driving patterns for 571 cohorts that represent distributions of the trip numbers and the trip schedules for 365 days, we obtained the distribution of driving length of passenger cars.
car number [100 thousand cars]
160 140 120 100 80 60 40 20 0

PHEV35 with 35 km of all-electric driving range

carnumber cumulativerate
100 90 80 70 60 50 40 30 20 10 0

carnumber(100thousandcars)

120 100 80 60 40 20 0

average annual kilometrage [km]

averagedailykilometrage

Distribution of annual driving length

Distribution of daily driving length

Both annual driving length and average daily driving length are consistent with the annual statistical survey of transportation.
Source:

100

100

10

15

20

25

30

35

40

50

60

70

80

90

11

cumulativerate []

140

Relative Gasoline and Electricity Demand

The savings of gasoline consumption and increases of electricity demand are calculated by comparing the gasoline or electricity demands for the case of all HEV cars with those for the case of all PHEV cars.
gasoline consumption ratio electricity demand

gasoline consumption ratio [%]

100 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100

60 50 40 30 20 10 0

electricity demand [TWh]

70

PHEV35 can save gasoline consumption by 58% compared with that of all HEV case. The incremental savings are getting smaller along with further increments of additional capacity of battery. If all passenger cars are replaced by PHEV35, electricity demand will increase by about 42 TWh in the 12 course of the year.
Source:

all-electric driving range [km]

100% in All HEV Case

Economics of PHEV
Taking into account future decline in battery cost, we assumed the parameters listed in the following table for the costs of vehicles.
ICV HEV PHEV
battery cost [JPY/kWh]
200000 150000

base cost for car 1,700,000 1,700,000 1,700,000 [JPY] HEV system cost 200,000 200,000 A [JPY] 1.5x 1.5x additional battery cost 7.1x B [JPY] x unit cost of battery (JPY/kWh) A electric motor drive system including control system + 1.5 kWh battery for HEV B additional 7.1kWh battery for PHEV

100000

50000 0 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 year

Source : The Proposals toward next-generation vehicle battery, METI

ICVinternal combustion engine vehicle

We assumed the lifetime of a battery is 12 years, same that of a vehicle.

Source:

13

We calculate total cost of the purchase of cars in the light of both vehicle cost and annual cost of fuel as shown in the following equation.

C: total cost V: vehicle value A: annual cost of fuel and electricity : discount rate (=0.05) Y: lifetime of vehicle (=12 years)

Source:

14

Market diffusion of PHEVs

By combining the breakeven analyses with future battery costs, we estimated the timing of market introduction of PHEV.
Assumption
The car stock of 57.1 million in Japan is gradually replaced with new cars at a rate of 4.8 million per year. Consumers purchase a vehicle which is the least expensive as for total cost (energy cost plus annualized capital cost of car).
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

HEV PHEV
ratio

ICV

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

HEV PHEV

ratio

ICV

2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 200100 30 10755 555 year(upperline) batterycost(lowerline,thousandyen/kWh)

2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 200100 30 10755 555
year(upperline) batterycost(lowerline,thousandyen/kWh)

Market shares in replaced car

Market shares in car stock

Source:

15

Battery charge for PHEV

The timing of charge (charging rate: 700W/PHEV)
bottom charge: it charges at night to raise the bottom demand of electricity.
25
180

maximum3days summerweekdays winterweekdays interimperiod weekdays summerweekends winterweekends

0 2 4 6 8 10 12 14 16 18 20 22

chargedpower[GW]

20 15

weekdays weekends

160 140 120 100

GW

10 5 0

80 60 40 20 0

10

12

14

16

18

20

22

time[h]

time[h]

interimperiod weekends

Daily load profile for charging batteries of PHEVs

Daily electricity demand

The created night load can

improve load factor of the total electricity demand. enjoy low electricity price at night in Japan.
16 Source:

Source:

17

Daily Electricity Demand

180 160 140 120

maximum3days summerweekdays winterweekdays interimperiod weekdays summerweekends winterweekends interimperiod weekends

GW

100 80 60 40 20 0 0 1 2 3 4 5 6 7 8 9 1011121314151617181920212223

Night Charge

time
180 160 140 120 100

maximum3days summerweekdays winterweekdays interimperiod weekdays summerweekends winterweekends

0 1 2 3 4 5 6 7 8 9 1011121314151617181920212223

GW

80 60 40 20 0

Bottom Charge 18

time

interimperiod weekends

Source:

Case
3 cases (+ ICV only case)
1. Reference case (without PHEVs) it optimizes generation mix without the additional electricity load for PHEV. 2. PHEV case (64GW ceiling case; case1): Consideration of the additional electricity load for charging batteries of the PHEVs. Besides, upper limit of generation capacity of nuclear plant is 64 GW. 3. PHEV case (80% ceiling case: case2): upper limit of generation capacity of nuclear plant is 80% of the least amount of hourly electricity demands. We calculated the difference between the reference case and PHEV cases such as 64GW ceiling case and 80% ceiling case.
Electricity generation CO2 emissions
Source: 19

Result electricity generation

water
40000 30000 20000

nuclear

coal

IGCC

LNG
100000 80000 60000 40000

LNGCC

oil

pumpedhydro

GWh

10000 0 10000 20000 30000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

20000 0 20000 40000 60000 80000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

year

100000

year

64GW ceiling case (case 1)

80% ceiling case (case 2)

Increased electricity demands caused by the introduction of PHEVs are supplied

by fossil power generations in 64GW ceiling case. by nuclear power generations in 80% ceiling case. Relaxation of the upper limit of 20 generation capacity of nuclear plant.

Source:

Result CO2 emissions from passenger cars

bygasoline(ICVs) bygasoline(HEVs) bygasoline(PHEVs) fromelectricitysectorbyPHEVs

CO2emissions[thousandtCO2]

120000 90000 60000 30000 0 30000 60000

33% 49%
71%

105%

Effect of total CO2 reduction from passenger cars in 2050.

HEV case (HEVs are available but PHEVs are not)

Reduction by 33% compared with that in only ICVs.

PHEV case (80% ceiling)

Reduction by 105% compared with that in only ICVs. The direct CO2 emission from the combustion of gasoline and the oil 21 product consumption are reduced by 71% of those in 2005. Source:

Case1: nuclear 64GW,

Case2: nuclear80% of bottom electricity demand 22 Source:

Economic Comparison
(case2 (bottom charge) ref. case (ICV+HEV))
600

additionalcost[billionyen]

400 200 0 200 400 600 800 1,000 1,200 1,400

2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

year

The cost of reducing CO2 emissions by the diffusion of PHEVs goes negative. The cost reduction of fuel is much larger than additional cost of electricity sector by the introduction of PHEVs.

The introduction of PHEVs produces both economic and Source: environmental benefits.

23

Economic Comparison
(case2 (night charge) ref. case (ICV+HEV))
600 400 200 0 200 400 600 800 1,000 1,200 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050

additionalcost[billionyen]

electricity sector

total vehiclerelated

The cost of reducing CO2 emissions by the diffusion of PHEVs goes negative. The cost reduction of fuel is much larger than additional cost of electricity sector by the introduction of PHEVs.

The introduction of PHEVs produces both economic and environmental benefits. Source:

24

Conclusion
PHEV will start market penetration in 2010 (the price of battery is 100 thousand JPY/kWh), and the share of PHEV in total passenger car stock will be around 80% after 2030 (the price of battery is 5 thousand JPY/kWh). Introduction of PHEVs with 35 km of all-electric driving range has a potential to decrease oil consumption by 71% and decrease CO2 emissions substantially, over 100% in 80% ceiling case compared with that in only ICVs case. The introduction of PHEVs produces both economic and environmental benefits.
Source: 25

Second Case Study:

PHEV-Wind Combination
With taking VtG (Vehicle-to-Grid) into perspective

Source: M. Takagi, H. Yamamoto, and K. Yamaji: Evaluation of Expanded Allowable Capacity of Wind Power in Power Systems by Charge Control for Plug-in Hybrid Electric Vehicles, IEEJ Trans. PE, Vol. 128, No.12 (2008) (in Japanese) K. Yamaji, M. Takagi, and H. Yamamoto: Synergies in Energy System: A Case Study of Plug-in Hybrid Electric Vehicle and Wind Power Combination, GCOE Symposium, University of Tokyo, January 2009 M. Takagi, H. Yamamoto, and K. Yamaji: Power System Stabilization by Charging Power Management of Plug-in Hybrid Electric Vehicles with LFC Signal, VPPC2009, Dearborn, Michigan, September 2009

26

Introduction
In power sector
Total capacity of wind power has been increasing drastically.
Wind power affects the frequency adversely because of the instability of output. Problem of frequency fluctuation becomes significant at nighttime when the capacity of Load Frequency Control (LFC) is insufficient.

In transport sector
Plug-in Hybrid Electric Vehicle (PHEV) is developed as environmentally friendly vehicle.
Electric energy of PHEVs will be charged mainly during nighttime when the electricity price is low.

Idea
Charging power control of PHEVs to compensate the LFC capacity.

Source:

27

Charging power control in this study

50.2 Frequency [Hz] 50.1

Charging power point at standard frequency (50Hz).

50

49.9

49.8 0 100 200 300 400 500 600 700 800 900 1000 Charging power [W]

PHEVs are charged at 500W per car when the frequency is 50Hz. PHEVs increase the charging power in case of increasing frequency and decrease the charging power in case of decreasing frequency. Control band of charging power ranges from 0 to 1000W Capacity of household outlets are enough for this control band. There is no voltage rise problem. There is no reverse-current.
Source: 28

Modeling -Three area models in JapanNumber of passenger vehicles and condition of power systems in each area
LFC capacity Area A B C Number of passenger cars [1,000] 2,400 7,640 5,410 Total demand Thermal Hydro [MW] [MW] [MW] 2,513 6,320 6,085 37 90 90 4 10 10 Total [MW] 41 101 99 Governor free capacity Ratio of the Thermal Hydro system [MW] [MW] capacity [%] 1.61 1.59 1.63 64 158 157 7 18 17 Total [MW] 71 176 174 Ratio of the system capacity [%] 2.82 2.79 2.86

We simulated the driving pattern of passenger cars with the following assumption. On-board battery of PHEV is 4.3kWh (working capacity), which corresponds to 30km of all electric-driving range. Batteries are charged when PHEVs are parked at home. Standard charging power is 500W with 10% energy loss. Batteries are charged during nighttime (from 23:00 to 7:00) using a timer. After the running distance exceeds the all electric-driving range, PHEVs will run on the hybrid mode. 29
Source:

Simulation result -daily charge curve3500 3000 Charging power [MW] 2500 2000 1500 1000 500 0 0 2 4 10 12 14 16 18 20 Time [h] Daily charge curve in Area C. 6 8 22 weekdays weekends

Assumption: PHEVs take 100% share of the market.

Charging power of weekdays is larger than that of weekends over the nighttime. We chose the charge curve of weekends for severe evaluation. 30
Source:

Simulation result frequency change0.2 Frequencydeviationf [Hz] 0.1 0.0 -0.1 -0.2 -0.3 -0.4 -0.5 0 500
Wind power generation capacity1200MW

150MW 50MW

Assumption: 1200MW of wind power generation is introduced in Area C. Consideration: Three cases, 1) without charge control 2) 50MW of charge control 3) 150MW of charge control

W/O charge control

1000 Time [sec] 50MW (100,000cars) 1500 2000

without charge control

150MW (300,000cars)
(Number of cars)

Frequency change in Area C

Control of charging power suppresses the frequency deviation effectively.

Source:

31

Evaluation approach for allowable capacity of wind power

Index
Standard deviation (SD) from nominal frequency Root mean square of deviation from 50Hz.
0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 0 100
(200)
without wind power

SD = ( f (t ) 50) 2

SD of frequency [Hz]

SD of frequency in reference case0.042Hz

200
(400)
900MW

300
(600)
1200MW

400

500

600

Level of no problem

Charging power [MW]

(800) (1,000) (1,200) (Number of cars[1,000] )

1500MW 1800MW

Method

SD of frequency to charging power.

1. SD of frequency without wind power is set as reference case. 2. When the SD of frequency becomes same level as reference case, we would judge the introduction of wind power as no problem.
Source: 32

Result allowable capacity of wind power 1400 1200 2800

3500

Charging power [MW]

Charging power [MW]

1100MW

2400 2000 1600 1200 800 400

Number of cars [1000]

3000 2500 2000 1500 1000 500 0

weekdays

weekends

1000 800 600 400 200 0 0

Gradient = about 0.5

More than

1100MW is

reserved during nighttime

16 18 20 22

600

1200

1800

2400

3000

Wind power generation [MW]

Needed charging power to keep the same level as reference case.

2800MW

10 12 14 Time [h]

Daily charge curve in Area C.

Needed charging power is almost proportional to the wind power capacity. Gradient of the straight-line is about 0.5 Current frequency quality can be maintained by the control of charging power with the 50% of wind power capacity. Minimum value of the charging power during nighttime is 1100MW Allowable capacity of wind power is 2800MW
Source: 33

Discussion
Result of allowable capacities of wind power

Area

Allowable capacity of wind power generations (market share of PHEV100%) [MW] Without charge control Charging power control

A B C Total

150 500 500 1,150

1,100 2,000 2,800 5,900

Total capacities of three areas are 1150MW without charging power control and 5900MW with the control. Capacity with the control is about 5.1 times bigger than that of without the control. Control of charging power for PHEV can contribute significantly to the introduction of wind power generation.
Source: 34