Charging/Discharging Behaviors and Integration of Electric Vehicle

to Small-Scale Energy Management System

Muhammad Aziz

Institute of Innovative Research, Tokyo Institute of Technology, 2-12-1 Ookayama, Meguro-ku, Tokyo, Japan

Keywords: Electric Vehicle, Charging Behaviour, Vehicle to Grid, Energy Management System.

Abstract: Integration of electric vehicle (EV) to support a small-scale energy management system (EMS) was

demonstrated and studied. Initially, charging and discharging behaviours of electric vehicle in different

seasons were evaluated to clarify the impact of surrounding temperature to charging and discharging rates. It

was found that charging and discharging during summer results in higher rates than ones during winter. In

addition, the integration of EVs to small-scale EMS (office) for peak-load shifting showed a very positive

effect. Discharging of EVs during noon’s peak load can cut and shift the load. Therefore, higher contracted

capacity of electricity can be avoided leading to lower total electricity cost.

1 INTRODUCTION

A massive adoption of electric vehicle (EV) replacing

the conventional internal combustion engine vehicle

(ICEV) is potential to reduce both greenhouse gases

emission and fossil fuel consumption (Oda et al.,

2016). Therefore, better environmental impacts can

be achieved. Rapid development of EV was

accelerated by some factors including rising oil and

gas prices, enhancement in battery technology, and

policies related to environment and transportation

(Aziz et al., 2015a; Oda et al., 2017). However, EV

has some barriers in its deployment such as high

initial cost, long charging time, and limited cruising

range. Although the operating cost of EV is relatively

lower than conventional ICEV, the production cost of

EV is significantly higher (Thiel et al., 2010).

Therefore, a value-added utilization of EV is crucially

required to improve the economic performance of

EV. Hence, sustainable deployment of EV can be

achieved. However, a massive deployment of EV can

give a significant impact to the grid, especially in case

that uncontrolled charging and discharging take place

massively. To minimize the impact of this problem,

as well as increase the economic performance of EV,

the concept of vehicle to grid (V2G) has been studied

(Kempton and Letendre, 1997).

EV utilization in supporting the grid has been

evaluated by some researchers previously (Kempton

and Kubo, 2000; Tomic and Kempton, 2007; White

and Zhang, 2011; Aziz et al., 2014; Aziz et al.,

2015b). The integration of EV to grid (V2G) can be

realized because of the character of EV in both

charging and discharging behaviours. As these can be

controlled, scheduling and rate control become

possible.

The parked and connected EVs can be considered

as a potential battery which is capable to absorb, store

and deliver back the electricity from and to the grid

following the given schedule and control value. The

realization of V2G can be achieved when minimally

three essential requirements are fully satisfied: (1)

electricity connection between EV and grid, (2)

communication facilitating control flows between EV

and operator, and (3) metering system providing fair

measurement (Drude et al. 2014). In V2G system,

certain grid operator or energy management system

(EMS) can send a request for electricity to a number

of parked and plugged EVs to discharge or absorb the

electricity to and from the grid.

V2G leads to possibility of several ancillary

services to grid such as load levelling, frequency

regulation, spinning reserve, and storage (Kempton

and Tomic, 2005; Gao et al., 2014). The distributed

EVs in large number and area are potential as massive

energy storage that can be used to balance

responsively the fluctuating supply such as PV and

wind. Furthermore, because EVs are mobile, they can

be utilized as energy carrier carrying the electricity

from and to different places and time because of some

factors including price difference and emergency

condition. EV utilization in V2G is considered

346

Aziz, M.

Charging/Discharging Behaviors and Integration of Electric Vehicle to Small-Scale Energy Management System.

DOI: 10.5220/0006377303460351

In Proceedings of the 6th International Conference on Smart Cities and Green ICT Systems (SMARTGREENS 2017), pages 346-351

ISBN: 978-989-758-241-7

feasible because the estimated profit is higher than the

current market price of EV batteries although

considering the wear of battery.

This paper focuses mainly on two issues: 1)

evaluation on charging and discharging behaviour of

EVs, and 2) impact of peak-load cutting and shifting

in a small-scale EMS based on demonstration test.

2 VEHICLE TO GRID

2.1 V2G in Community Energy

Management System

A community EMS (CEMS) has been proposed in

Japan with the main aim of integrating the energy

utilization, covering both demand and supply sides,

therefore, more efficient energy utilization and

reduction of CO

emission can be realized. The

initiatives to propose and adopt CEMS came from the

willingness to harmonize energy services, minimize

environmental impacts, and maximize economic

benefit. CEMS manages both demand and supply of

energy, especially electricity, across the whole

community. It is responsible in maintaining the

balance and harmonization in the community, hence,

the comfort, security, and safety of the members are

improved. In addition, environmental parameters are

also considered in parallel with the living quality.

Therefore, in CEMS, both information and energy

are flowing simultaneously across the community.

CEMS receives and manages the information and

then delivers it to the community members according

to their function. CEMS must be sufficiently robust

and secure because it deals with personal and

authentication information from the community.

Figure 1 represents a schematic structure of

CEMS including electricity and information flows,

especially its relation with the EV utilization inside

CEMS. CEMS collects and manages the information

from smaller EMS such as building EMS (BEMS),

house EMS (HEMS), and factory EMS (FEMS). In

addition, CEMS also has a communication with the

outside of community and also upper energy supply

such as electric utilities. CEMS predicts and

maintains both energy supply and demand based on

available previous historical data for certain time and

forecasted weather information. Furthermore, CEMS

also calculates and optimizes the energy balance with

the aim of achieving the lowest energy cost

throughout its community. In addition, in case of

emergency such as disaster, CEMS evaluates and

controls the energy conditions and communicates to

its lower EMSs and negotiates with other CEMSs or

electric utilities to cover its energy demand and

recover the conditions.

Figure 1: Utilization of EVs to support CEMS.

2.2 EV Utilization System

Utilization of EVs to support the grid, including

ancillary services and storage, is possible because of

EVs characteristics. In energy storage utilization,

charging of EVs can be scheduled when the price of

electricity drops because of electricity surplus in the

grid. In addition, when the electricity price increases,

electricity can be discharged and delivered back to the

grid, leading to economic margin for the owner. The

ancillary services from EV to grid includes frequency

regulation (both up and down) and spinning reserve.

Ancillary services are important to maintain the

quality of the electricity. As the responses of EV in

both charging and discharging are very fast, the

ancillary service by EV is considered potential.

Figure 2 represents the schematic utilization of

EVs for grid support. In general, there are two

schemes in this utilization: direct and aggregator-

based schemes. The collection of real time data in a

certain interval from EVs includes battery state of

charge (SOC), EVs position and predicted arrival

time. This data collection is performed by a vehicle

information system (VIS). In the reality, VIS can be

owned and operated directly by EMS, aggregator or

independent service operator.

Figure 2: Schemes in utilization of EV supporting the grid:

(a) direct scheme, (b) aggregator-based scheme.

In direct scheme, EV owners have the service contract

Charging/Discharging Behaviors and Integration of Electric Vehicle to Small-Scale Energy Management System

347

directly with electricity-related entities. In this

scheme, both the electricity and information are

handled privately. This scheme is well suited for a

relatively small-scale EMS and in where EVs are

parked and connected in relatively long time (such as

office). Its main advantage is the potential to

maximize the profit of the involved entities.

Furthermore, both charging and discharging controls

are easier as EVs are directly connected and fully

controlled by EMS. The current study focuses on this

type of utilization scheme.

On the other hand, in aggregator-based contract,

EV owners have service contracts with the

aggregator. The information including its position

and battery SOC are handled by aggregator via VIS.

Aggregator negotiates for electricity business with

the electricity-related entities (EMS or electricity

utilities). This kind of utilization scheme is prevalent

for relatively large-scale EMS or electricity utilities.

EVs may be distributed in different location, such as

charging stations and parking areas. The electricity to

and from EVs may be transferred via power wheeling

system through the available grids. Aggregator offers

some possible ancillary services to the EV owners. In

turn, EV owners can choose them and receive their

profit payment from aggregator.

Load levelling correlates strongly with the

management of both demand and supply of

electricity. Its aim is lowering the total power

consumption in peak hours by shifting the load from

peak to off-peak hours. Load levelling can be

performed through peak-load shifting and peak-load

cutting. The former is defined as moving the

electricity load during peak time to off-peak time. It

could be achieved through utilization of stationary

battery or other storages. The latter deals with the

effort to reduce the electricity purchased from the grid

by generating or purchasing the electricity. In reality,

it can be performed through harvesting the energy

especially during peak hours, such as RE, or by

purchasing the electricity from other entities

including EVs. In this case, EVs are considered as

energy storage and carrier storing and transporting the

electricity from different time and place. Hence, the

economic performance of EV can be increased by

joining this kind of ancillary program.

3 CHARGING AND

DISCHARGING BEHAVIOURS

EVs largely adopt li-ion batteries to store the

electricity as power source because of high energy

density, stable electrochemical properties, longer

lifetime, and low environmental impacts (Aziz et al.,

2016). Temperature is considered as one factors

influencing charging and discharging behaviours of

li-ion batteries. Generally, lower temperature leads to

poor charging and discharging performance because

of electrolyte limitation (Xiao et al., 2004) and

changes in electrolyte/electrode interface properties

including viscosity, density, electrolyte components,

dielectric strength, and ion diffusion capability

(Jansen et al., 2007). Liao et al. (2012) found that as

the temperature decreases, the charge transfer

resistance increases significantly, higher than bulk

resistance and solid-state interface resistance.

Unfortunately, lack of study deals with the effort

to clarify the charging behaviours in different

temperature or season. In this study, to clarify the

effect of temperature (ambient temperature), to the

charging behaviour of EV, charging in different

seasons: winter and summer, were conducted

initially.

Figure 3: Charging performances of EV in different

seasons: (a) winter, (b) summer.

Figure 3 shows the relation among charging rate,

charging time, and SOC of EV battery both in winter

(a) and summer (b). Generally, although the rated

capacity of the charger is 50 kW, the charging power

absorbed by EV battery is relatively lower, especially

in winter. Compared to charging in winter, charging

in summer leads to higher charging rate and shorter

SMARTGREENS 2017 - 6th International Conference on Smart Cities and Green ICT Systems

348

charging time. Numerically, to reach SOC of 80%,

the required charging times in winter and summer

were 35 and 20 min, respectively. In summer, higher

charging rate (about 40 kW) could be achieved up to

SOC of 50%. It decreased gradually following the

increase of SOC and it showed the charging rate of 16

kW when SOC reached 80%. On the other hand, in

winter, the charging rate reached about 35 kW

instantaneously in very short time and decreased

following the increase of SOC. The charging rate

when SOC reached 80% was about 10 kW.

4 V2G DEMONSTRATION IN

SMALL-SCALE EMS

The schematic diagram of EV integration test to

support the electricity in a small-scale EMS is

presented in Figure 4. EMS basically controls all the

electricity demand and supply. It requests, manages,

and integrates some information: electricity load,

weather information from meteorological agency, EV

information from VIS, electricity condition from

CEMS and utilities. The meteorological agency sends

periodically the weather information. It is used by

EMS to calculate the next coming load and the

possibly generated electricity from RE. Building load

is classified into base and fluctuating loads. The

former is the minimum demand consumed by the

system to operate for 24 hours continuously.

Therefore, it is generally constant throughout the time

and almost not affected by the weather condition. In

addition, the latter depends strongly to the behaviours

of the residents. Human behaviour is influenced by

the weather condition including temperature and

humidity. In addition, the generable power from RE,

such as wind and solar, is predicted also by EMS

based on the received weather information.

EMS also receives information from VIS

including EV position, battery SOC, and estimated

arrival time. VIS collects the information from EV

routinely. The collected data is utilized to coordinate

both charging and discharging of EV and to keep the

balance of electricity distribution and avoid any peak-

load in EMS. In addition, VIS also can provide

additional services to the driver regarding the

availability of ancillary service programs offered by

EMS or aggregator. Therefore, EMS also requests

from electric utility the electricity condition and price

information. These will be used to calculate the

demand as well as the charging and discharging

behaviours of both EVs and used EV batteries.

Figure 4: Schematic diagram of developed V2G in small-

scale EMS.

4.1 Developed V2G Concept

Figure 5 represents the concept of peak-load cutting

and shifting (load levelling) developed for a small-

scale EMS. Four main subsequent steps were

proposed: (1) forecasting of load and power from RE,

(2) forecasting the amount of load levelling, (3)

correction of calculated value, and (4) charging and

discharging controls of EVs. Forecasting of load and

power from RE is conducted for 24 hours-ahead.

During forecasting of load and generable power

from RE, EMS initially requests a weather

information from meteorological agency for

calculating the generable electricity from RE

including PV and wind. In this study, the generated

electricity from RE is completely used for peak-load

cutting and will be consumed entirely with no storage.

Therefore, the weather information is also utilized to

predict the fluctuating load mainly due to human

behaviour inside the building, especially related to air

conditioning and lighting. The outputs from this first

step are forecasted RE generation and load curves for

the next 24 hours ahead.

Figure 5: Load-levelling concept used in the V2G

demonstration.

Once the load and RE generation have been

forecasted, EMS calculates the possible load levelling

which can be achieved in the next 24 hours. For this

purpose, EMS communicates with VIS to estimate

the number of EVs and their SOC states which are

Charging/Discharging Behaviors and Integration of Electric Vehicle to Small-Scale Energy Management System

349

available to join the program. VIS initially receives

the travelling schedule from the drivers and,

subsequently, VIS transfers this information to EMS

including EV’s ID, planned departure time and

estimated arrival time. Moreover, the registration of

travelling schedule by the driver should be done up to

24 hours before the departure. As the available

resources for peak-load cutting and shifting and load

curves are estimated, EMS can calculate the peak-

load cutting threshold for the next day. It is defined as

the maximum amount of electricity purchased from

the grid by EMS. It is theoretically calculated based

on some factors including electricity price, contracted

capacity, available power generation, and storage.

When the electricity consumed by the building

increases and the purchased electricity from the grid

reaches the peak-load cutting threshold, EMS sends

the command to EVs and used batteries to discharge

their electricity. Hence, the electricity purchased from

the grid is same to of lower than the calculated peak-

load cutting threshold. In the real application, it may

avoid a higher price of electricity during peak time.

When EVs are in motion, the information of EVs

is transmitted to VIS. VIS sends the data to EMS

which is used to recalculate the available electricity

from EV. EMS also recalculates the building load

based on the real weather and the real load of

building. Next, EMS modifies its energy management

plan, especially the peak-load cutting threshold.

When EVs arrive and are connected to the

chargers, they communicate directly with EMS.

Hence, all the information is updated including the

available electricity from EVs. From this moment,

EVs are ready to take part in the ancillary service

program offered by EMS. EVs are fully controlled by

EMS, especially their charging and discharging

behaviours. Finally, EMS calculates and sends the

control command to each EVs and used batteries to

keep its previously calculated peak-load cutting

threshold.

4.2 Demonstration Test

The demonstration test facility was constructed in the

factory area of Mitsubishi Motors Corp., Okazaki,

Japan. It was connected and utilized to support the

electricity of the main office building. As RE

generator, 20 kW PV panels were installed on the roof

top of test bed. Five EVs (battery capacity of 16

kWh), Mitsubishi i-Miev G, were taking part in the

program and the drivers were also the employee.

Hence, EVs were mostly parked and plugged to the

charging poles during working hours. Therefore, five

used batteries, used for about 1 year from the same

type of EV, were also installed and directly owned by

EMS. EMS are managing all the demand and supply

sides of the test bed. On the other hand, VIS was also

developed as a standalone system to communicate

with EVs and transfer the information to EMS.

Used batteries were practically used for peak-load

shifting. They were charged during the night time

(off-peak hours) when the price of electricity was

cheaper. In this study, this charging was performed

from 00:00 to 06:00. The SOC thresholds during

charging and discharging of both EV and used battery

were fixed at 90% and 40%, respectively. In addition,

as the total capacity amount of EVs and used batteries

was very small than the total load of the office, peak-

load cutting was designed to start from 12:00 until

18:00 (targeting mainly on the afternoon peak).

Figure 6 represents the results of load levelling

test in a representative weekday consisting of total

grid load, building load, generated power from PV,

and charged/discharged electricity to and from EVs

and used EV batteries. The grid load is the net

electricity from the grid. Light green blocks in

positive and negative sides represent discharge and

charged electricity amount from and to EVs and used

batteries, respectively. The main objective of peak-

load cutting and shifting is to reduce the grid load,

especially during peak hours, to reduce the total

electricity cost of the building.

Figure 6: Schematic diagram of developed V2G in small-

scale EMS.

PV generated the electricity during a day and it was

directly consumed without being stored. The used EV

batteries were charged during the night with a lower

electricity price, starting from 00:00 to 06:00. As the

result, the grid load during the night slightly

increased. In the morning, around 08:00, EVs reached

the office and they were plugged in designated

charging poles. From this moment, charging and

discharging behaviours were fully controlled by

EMS. Because the building load was smaller than the

calculated peak-load cutting threshold, charging for

EVs was performed until the building load reaches

nearly the peak-load cutting threshold. Additional

SMARTGREENS 2017 - 6th International Conference on Smart Cities and Green ICT Systems

350

charging started again during noon break (12:00 to

13:00) because the building load dropped drastically.

Peak-load takes place twice in a weekday: before

and afternoon. Peak at the morning is lower than one

in afternoon. The afternoon peak usually starts from

13:00 following the end of noon break. As the peak-

load is higher than the calculated peak-load cutting

threshold, EMS sends the control command to both

EVs and used batteries to discharge their electricity.

As the results, the grid load can be reduced. However,

because the total capacity of EVs and used batteries

is very small compared to the total building load,

peak-load cutting can only be performed in a

relatively short duration . If the number of EVs

participating in the load levelling program increases,

the effect of load levelling becomes more significant.

Therefore, longer peak-load cutting and lower peak-

load cutting threshold can be achieved.

5 CONCLUSIONS

Integration of EV to small-scale EMS is studied and

demonstrated in this study. Charging and discharging

behaviours of EV were initially clarified. It was found

that charging and discharging rates during summer is

higher than ones during winter. The demonstration

test was performed to measure the application of EVs

for peak-load cutting and shifting. It was clarified that

the utilization of EVs for peak-load cutting and

shifting in small-scale EMS is very feasible.

ACKNOWLEDGEMENTS

The author expresses his deep thanks to Mitsubishi

Corp. and Mitsubishi Motors Corp. for their

collaboration during demonstration test.

REFERENCES

Aziz, M., Oda, T., Mitani, T., Watanabe, Y., Kashiwagi, T.,

2015a. Utilization of electric vehicles and their used

batteries for peak-load shifting. Energies, 8(5):3720-

3738.

Aziz, M., Oda, T., Morihara, A., Murakami, T., Momose,

N., 2014. Utilization of EVs and their used batteries in

factory load leveling. In Proceedings of 2014 IEEE PES

Innovative Smart Grid Technologies Conference

(ISGT), doi: 10.1109/ISGT.2014.6816370.

Aziz, M., Oda, T., Kashiwagi, T., 2015b. Extended

utilization of electric vehicles and their re-used batteries

to support the building energy management system.

Energy Procedia, 75:1938-1943.

Aziz, M., Oda, T., 2016. Load levelling utilizing electric

vehicles and their used batteries. Modeling and

Simulation for Electric Vehicle Applications, InTech,

Rijeka. pp. 125-147.

Aziz, M., Oda, T., Ito, M., 2016. Battery-assisted charging

system for simultaneous charging of electric vehicles.

Energy, 100:82-90.

Drude, L., Pereira, L.P., Ruther, R., 2014. Photovoltaics

(PV) and electric vehicle-to-grid (V2G) strategies for

peak demand reduction in urban regions in Brail in a

smart grid environment. Renewable Energy, 68:443-

451.

Gao, B., Zhang, W., Tang, Y., Hu, M., Zhu, M., Zhan, H.,

2014. Game-theoretic energy management for

residential users with dischargeable plug-in electric

vehicles. Energies, 7:7499-7518.

Jansen, A.N., Dees, D.W., Abraham, D.P., Amine, K.,

Henriksen, G.L., 2007. Low-temperature study of

lithium-ion cells using a Li

Sn micro-reference

electrode. Journal of Power Sources, 174:373-379.

Kempton, W., Kubo, T., 2000. Electric-drive vehicles for

peak power in Japan. Energy Policy, 28:9–18.

Kempton, W., Letendre, S.E., 1997. Electric vehicles as a

new power source for electric utilities. Transportation

Research Part D, 2:157–175.

Kempton, W., Tomic, J., 2005. Vehicle-to-grid power

fundamentals: Calculating capacity and net revenue.

Journal of Power Sources, 144:268-279.

Liao, L., Zuo, P., Ma, Y., Chen, X., An, Y., Gao, Y., Yin,

G., 2012. Effects of temperature on charge/discharge

behaviors of LiFePO

cathode for Li-ion batteries,

Electrochimica Acta, 60:269-273.

Oda, T., Aziz, M., Mitani, T., Watanabe, Y., Kashiwagi, T.,

2016. Number of quick charging stations and their

quick chargers for electric vehicles ─a case study

considering the traffic density of the road type─.

Journal of Japan Society of Energy and Resources,

37(6):7-12. (in Japanese).

Oda, T., Aziz, M., Mitani, T., Watanabe, Y., Kashiwagi, T.,

2017. Actual congestion and effect of charger addition

in the quick charger station —Case study based on the

records of expressway—. IEEJ Transactions on Power

and Energy, 136(2):198-204.

Thiel, C., Perujo, A., Mercier, A., 2010. Cost and CO

aspects of future vehicle options in Europe under new

energy policy scenarios. Energy Policy, 38:7142-7151.

Tomic, J.; Kempton, W., 2007. Using fleets of electric-

drive vehicles for grid support. Journal of Power

Sources, 168, 459-468.

White, C.D., Zhang, K.M., 2011. Using vehicle-to-grid

technology for frequency regulation and peak-load

reduction. Journal of Power Sources, 196:3972-3980.

Xiao, L.F., Cao, Y.L., Ai, X.P., Yang, H.X., 2004.

Optimization of EC-based multi-solvent electrolytes for

low temperature applications of lithium-ion batteries,

Electrochimica Acta, 49:4857-4863.

Charging/Discharging Behaviors and Integration of Electric Vehicle to Small-Scale Energy Management System

351