Optimized Integration of Electric Vehicles with Lithium Iron

Phosphate Batteries into the Regulation Service Market of Smart

Grids

Enhanced Vehicle-to-Grid Business Model

Marco Roscher, Jonas Fluhr and Theo Lutz

FIR at RWTH Aachen, Pontdriesch 14/16, Aachen, Germany

Keywords: Regulation Service Market, Integration of Electric Vehicles into Smart Grids, Vehicle-to-Grid (V2G)

Business Model, Prediction of Electric Vehicles Batteries Degradation, Optimization Model.

Abstract: In the “Li-Mobility” project, a battery management system (BMS) for lithium-ion batteries was developed.

It particularly aims to perform online diagnoses while driving as well as to predict the impact of additional

cyclization through grid control and balancing functions on the degradation of the battery. The information

about the predicted degradation and the overall decision whether a car battery is providing balancing power

or not is based on an optimization model run by an aggregator- The aggregator’s ICT system uses

information based on the number of available electric vehicles, their configuration, battery characteristics

and information about market prices for supply of balancing power in order to optimized offers. Hence, it is,

possible to make realistic cost estimations for vehicle-to-grid (V2G) scenarios and to determine

economically profitable ones for each vehicle individually. The optimization model was tested on the basic

principles of conventional lithium-ion cells in general and with lithium iron phosphate (LiFePO4) cells in

particular.

1 INTRODUCTION

The increasing shortage of fossil resources during

the last decades makes the search for sustainable

energy resources more and more essential

PPELRATH 2012). The pertinence of renewable

energy source (RES) like wind power or

photovoltaic solar power plants is constantly

growing during the last years in most European

countries. Though, the energy generation through

RES indeed is complicated due to fluctuations over

time. Volatile power producers increase the grid´s

demand for balancing power because the overall grid

stability depends on the balance between production

and consumption at every point in time. Hence,

options are needed to use the electricty produced

during times of low consumption on one hand and to

obtain addition electricty during times of low

production on the other hand.

An excellent cost-effective alternative is perhaps

the Vehicle-to-Grid (V2G) principle. The concept of

V2G not only provides additional value to the

batteries of the electric vehicles by serving for

balancing power services, but also contributes to the

reduction of global CO

production by decreasing

the need of fossil fueled regulation plants. As a

result of additional charging cycles due to balancing

services, additional degradation of the battery

occurs. This consequently incur costs. It is important

to account this factor in existing V2G business

models and establish the necessary requirements in

the electric vehicles systems, in particular the battery

management system (BMS).

The subsequent sections are structured as

follows. Section 2 provides a short overview of the

state of the art in the energy market and electric

mobility. In Section 3, an enhanced V2G business

model and its key market are described. Section 4

discusses the business model underlying

optimization model. Section 5 offers a brief

summary and conclusion of the paper.

2 STATE OF THE ART IN V2G

Due to the attractiveness of the idea to combine the

Roscher M., Fluhr J. and Lutz T..

Optimized Integration of Electric Vehicles with Lithium Iron Phosphate Batteries into the Regulation Service Market of Smart Grids - Enhanced

Vehicle-to-Grid Business Model.

DOI: 10.5220/0004551600880092

In Proceedings of the 2nd International Conference on Smart Grids and Green IT Systems (SMARTGREENS-2013), pages 88-92

ISBN: 978-989-8565-55-6

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

two large energy conversion systems - “electricity

grid” and “mobility” - with each other (KEMPTON;

LETENDRE 1997), several studies from various

perspectives with regard to the realisation of V2G

have been done so far. For instance, studies have

been performed on the potential availability of

electric vehicles (F

LUHR; AHLERT; WEINHARDT

2010), the target market (base load, peak load,

regulation) and the profitability including other

aspects like the influence of the car type (BEV,

FCEV, HEV) for (D

ALLINGER; KRAMPE;

WIETSCHEL 2011).

The effect of V2G on the German electricity

grid has also been investigated (B

IRNBAUM;

LINSSEN; MARKEWITZ [et. al.] 2009). In addition,

there are studies with regard to the integration of

RES by V2G (K

EMPTON; TOMIC 2005).

In summary, while intelligent charging of

batteries might be applicable in Germany´s

electricity market model without problems even at

present day, the application of V2G would however

need serveral modifications (G

UDERGAN; ANSORGE;

FROMBACH 2011). A complete and optimized

integration of electric vehicles into a Smart Grid

requires a bi-directional energy and information

flow. This accounts particularly for the regulation

market, because of high requirements with respect of

reliability and response time of the ICT system.

3 BUSINESS MODEL

There are different concepts of V2G business

models that can be applied for regulation market

services. Every business model can be assigned to

one of two types of models. The first model focuses

on vehicle specific aspects and the second focuses

more on the information respective infrastructural

perspective. The business model described in this

paper is the second one (infrastructure-focused). It

was further investigated for the application for

regulation market services. It is an extension of

existing business models by the very important

factor of battery degradation through additional

cycles because of balancing power services. In this

context, every cycle incurs costs by decreasing the

electric vehicles battery monetary value. Only by

knowing these costs allows to deduct adequate fees

for the supply of balancing power services as a basis

for a robust business model.

3.1 Aggregator of Electric Vehicles

The business model "aggregator of electric vehicles"

is developed exemplary and is based on the

optimization model and the additional information

about degradation available from the car’s BMS.

It is very important to get a deep and clear

understanding of the business model especially from

the very beginning of the conception and planning

stages to enable and facilitate an optimized

integration of the electric vehicle´s batteries into the

regulation service market of smart grids. To comply

with the requirements of smart grid integration, a

structured scientific substantiation and

documentation of the business model would also be

essential. For this the strategic management template

Business Model Canvas was adapted

STERWALDER; PIGNEUR 2010) delivering a broad

overview of the business model´s key issues (Table

1) as a start and subsequently offering a base of

operation for the deduction of specific requirements

regarding a smart grid integration.

Table 1: Key Issues of Electric Vehicles Aggregator’s

Business model.

Key Partners Key Activities

 Individuals

 Organisations

 Public and private

infrastructure operators

 Backup storage for

renewable energy sources

 Regulation market

services

Key Resources Value Propositions

 Electric vehicle´s

battery capacity

 Arbitrage profits due to

variable prices

Customer Segments Revenue Streams

 System Operators

 Individuals

 Organisations

 Public and private

infrastructure operators

 Arbitrage profits due to

variable prices

 Shares of arbitrage profits

 User fees

Within this business model the aggregator

represents a crucial instance, which initially divides

the required total power into smaller packages, so-

called load profiles. It then translates them into a

format of time and performance vectors that is

interpreted by the BMS of electric vehicles. Finally,

it sends corresponding inquiries to pool of electric

vehicles. The primary objective is to optimize the

use of the battery by additional cycles for the

provision of balancing power, aside from the single

application for driving, so that the battery´s life is

optimally utilized.

Arbitrage profits due to variable prices are in no

doubt the business model´s crucial revenue stream

with regard to smart grid regulation service market.

The realization of arbitrage profits could allow at

least a partial refinancing of the relatively high

investment costs of the vehicles batteries and as a

OptimizedIntegrationofElectricVehicleswithLithiumIronPhosphateBatteriesintotheRegulationServiceMarketof

SmartGrids-EnhancedVehicle-to-GridBusinessModel

consequence the reduction of the overall investment

risks.

3.2 Regulation Service Market

In order to maintain the security of energy supply,

the transmission network operators is obliged to

guarantee the balance between energy production

and consumption in every point in time.

Occasionally, there is a deviation from the predicted

demand or supply of electric power in the grid, e.g.

because of failures or volatile energy producers.

Transmission network operators are keeping

additional equipment available to be able to react at

all times.

Three different types of balancing power are

established in Germany. The overall amount of each

balancing power type for the four German

transmission network operators is advertised online

for bidding on Germany´s national regulation market

platform. The regulation market place is open to all

technical entities which have successfully carried

out the pre-qualification by the responsible

transmission network operator of the control area.

Every balancing power type is characterized by a set

of properties listed in Table 3 to Table 5 for each

type and outlined in the following subsections.

3.2.1 Primary Balancing Power

Primary balancing power constitutes the first and

immediate action that can be taken to restore the

balance within the grid. It possesses the highest

value for the grid and has to be activated

automatically within 30 seconds after the request

call. The adduction may take up to 15 minutes. It is

advertised weekly on the regulation market online

platform with a minimum power quantity of 1

Megawatt (MW) positive or negative. There are no

time slices such that the balancing power which was

sold on the regulation market has to be available at

all times during the week it was paid for. Only a fix

rate per MW is paid for primary balancing power

services, regardless of the service delivery.

Regarding the required minimum size of

aggregators´ electric vehicle fleet, the primary

balancing power is the type that needs the fewest

electric vehicles, since transferred energy volumes

are normally low.

Table 2: Characteristics of primary balancing power.

Bidding frequency Minimum quantity

Weekly + / - 1 MW

3.2.2 Secondary Balancing Power

Secondary balancing power substitutes the primary

balancing power in the event of longer technical

disturbances. The activation of the balancing power

adducted has to start maximum 5 minutes after the

request call and can last up to 1 hour to stabilize the

power grid. Secondary balancing power is advertised

weekly on the regulation market online platform

with a minimum power quantity of 5 MW positive

or negative for each of the two time slices, primary

processing time and secondary processing time.

In contrast to primary balancing power, the price

paid for secondary balancing power services consists

of a rate per MW and a rate per MWh. The MW-rate

takes account the provision of balancing power,

while the MWh-rate comprises the de facto delivery

of balancing power to the grid.

Table 3: Characteristics of secondary balancing power.

Bidding frequency Minimum quantity

Weekly + / - 5 MW

3.2.3 Minute Reserve Power

In case of long term failures which take longer than

15 minutes, the minute reserve power is utilized. Its

complete activation and delivery is compulsory

within 15 minutes after the request call. Compared

to the other balancing power types, minute reserve

power has the lowest value and the longest possible

time period of balancing power delivery. As a

consequence, the duration of balancing power

supply can take several hours. The compensation for

the supply of minute reserve power consists of a

demand rate and a MWh rate similar to the

secondary balancing power. Regarding the limited

battery capacity of electric vehicles, it is very

unlikely that one vehicle will be able to deliver

minute reserve power for several hours.

Table 4: Characteristics of minute reserve power.

Bidding frequency Minimum quantity

Daily + / - 5 MW

4 OPTIMIZATION MODEL

In order to understand the optimization model, the

aggregator´s planning process for the supply of

balancing power is described in Figure 1.

At first, the aggregator has to be informed about

the probable availability of its electric vehicles

SMARTGREENS2013-2ndInternationalConferenceonSmartGridsandGreenITSystems

Figure 1: Flow chart of the aggregator´s planning process

for supply of balancing power.

within its vehicle pool at least for the following day,

since it wants to participate for the bidding of minute

balancing power and up to a whole week as for

primary and secondary balancing power biddings.

As soon as a vehicle determined the amount of

balancing power it is willing to offer, it can

participate (assuming prior pre-qualification) in the

biddings. After being awarded to supply balancing

power by the transmission network operator´s

bidding, the actual optimization starts. The

aggregator translates the requested balancing power

into three predefined load profiles. The balancing

power ranges to minimum 1 MW up to several MW

and is dependent on the number of electric vehicles

available in the pool. A load profile consists of a

power dimension (kW) and a time dimension

(seconds).

As a function of available and suitable electric

vehicles, the overall balancing power is split in to

partial balancing power amounts for each of the

three load profiles. One request consists of up to

three load profiles and is sent to each electric vehicle

BMS which is connected by the aggregator to the

pool’s ICT systems. For that, the load profiles have

to be initially translated to a BMS proprietary

format. In this way, every load profile´s power and

time dimension is encoded in a CAN message of 8

bytes which is interpretable by the BMS. The BMS´

of the electric vehicles then process the request and

send the aggregator the predicted degradations as a

change of the battery´s state of health (deltaSOH)

for every load profile received. The new state of

health of the battery is calculated according to

equation (1):

SOH

new

= SOH

actual

– deltaSOH (1)

By this, the aggregator has up to three different

deltaSOH values with associated balancing power

amounts for each electric vehicle. At this point, the

aggregator concentrates all the information of each

connected electric vehicle, e.g. availability, the

battery´s state of charge, capacity, value, deltaSOH

etc. Then in the first step of optimization model´s

algorithm all vehicles with the lowest costs (€ per

kWh) are added to the bulk of selected electric

vehicles for balancing power services as long as the

required overall balancing power is achieved. In

case the required overall power is not achieved and,

all available vehicles are already selected, the

algorithm deselects the electric vehicles which do

not supply their maximum balancing power possible

yet. Thereafter, the electric vehicle with the

minimum costs in € per kWh is selected with the

restriction that the selected load profiles amount of

balancing power has to be higher than the amount of

the previously selected one. This procedure is

repeated until the required overall power is finally

achieved. The aggregators profit (G) is the

difference between the earnings (e) and the costs (c)

for the balancing power services of every electric

vehicle (index k) and the load profile (index i)

selected according to equation (2):



,

max

,







,







,∀

(2)

The aggregators profit is the difference between

the earnings and the costs for the balancing power

services of every electric vehicle selected according

to equation (2): The supply of negative balancing

power is exceedingly interesting for an aggregator

especially during the times where its electric

vehicles need to be charged e.g. overnight or when

the price is under the regular rate for electricity from

the grid. This optimization model is currently

implemented in a test environment of a lab including

a real BMS and its degradation. The aggregator side

and its vehicle pool were simulated via Matlab.

5 SUMMARY AND CONCLUSION

The optimization model developed can be adapted

for a variety of business models in principle. The

OptimizedIntegrationofElectricVehicleswithLithiumIronPhosphateBatteriesintotheRegulationServiceMarketof

SmartGrids-EnhancedVehicle-to-GridBusinessModel

critical success factors and conditions are considered

in the modelling of boundary scenarios and their

impact on business success.

Since the future of electric vehicles depends to a

great extent on the profitability of developed

solutions, the selected and specified business model

"aggregator of electric vehicles“ addresses initially

in particular large fleet operators with hundreds or

thousands of vehicles. Only in this magnitude,

reliability for balancing services can be guaranteed if

some vehicles are not available. Moreover, only then

needed scaling effects can be reached. On the other

side, the potential revenues from sales of balancing

power could lead to an additional incentive for the

electrification of vehicles and thus, represent an

appropriate leverage and multiplier for further

distribution and market acceptance of electric

vehicles in general.

Furthermore, the model is also of interest for

energy utilities and transmission system operators,

since it is a new source for the supply of balancing

power.

ACKNOWLEDGEMENTS

The authors wish to thank all project partners for

their support. This work was partly funded by the

Federal Ministry of Education and Research, based

on a decision of the German Bundestag (Li-

Mobility: Grant 03X4614B).

REFERENCES

Appelrath, H.-J., 2012. (Hrsg.): Future energy grid.

Migrationspfade ins Internet der Energie. Acatech-

Studie. Springer, Berlin, Heidelberg.

Birnbaum, K. et al., 2009. Elektromobilität -

Auswirkungen auf die elektrische Energieversorgung.

In: BWK - Das Energie-Fachmagazin 61, 1/2, S. 67–

74.

Dallinger, D.; Krampe, D.; Wietschel, M., 2011. Vehicle-

to-Grid Regulation Reserves Based on a Dynamic

Simulation of Mobility Behavior. In: IEEE

Transactions on Smart Grid 2, S. 302–313.

Fluhr, J.; Ahlert, K. H.; Weinhardt, C., 2010. A Stochastic

Model for Simulating the Availability of Electric

Vehicles for Services to the Power Grid. In: 43rd

Hawaii International Conference on System Sciences

(HICSS), Honolulu, Hawaii, 5 - 8 Jan. 2010. Hrsg.: R.

H. Sprague. IEEE, Piscataway, NJ, S. 10.

Gudergan, G.; Ansorge, B.; Frombach, R., 2011.

Exploring Interdependencies of Innovation Systems.

In: POM 2011. 22nd Annual Conference of the

Production and Operations Management Society.

Hrsg.: H. Correa. o.V., Reno

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

new power source for electric utilities. In:

Transportation Research 2, S. 157–175.

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

implementation: From stabilizing the grid to

supporting large-scale renewable energy. In: Journal

of Power Sources 144, S. 280–294.

Osterwalder, A.; Pigneur, Y., 2010. Business model

generation. A handbook for visionaries, game

changers, and challengers. Flash Reproductions,

Toronto.

SMARTGREENS2013-2ndInternationalConferenceonSmartGridsandGreenITSystems