Implementation of cost benefit analysis of vehicle to grid coupled real Micro-Grid by considering battery energy wear: Practical study case

Abstract

The proposed research represents a spin-off of the Malta College of Arts, Science and Technology (MCAST) Micro-Grid (MG) project. Particularly, economic impact of Electric Vehicles (EV) integration into the MG is investigated in this paper. The MCAST MG consists of photovoltaic generation unit, a diesel generator and a battery storage system. In this paper, a Vehicle-to grid (V2G) concept is considered where utilities can profit from controlled energy trading operations according to EVs availability. EVs are categorized under different profiles considering energy and time availability of owners typical work hours. V2G energy cost is estimated based on battery energy wear due V2G extra cycling and refunded to EVs owners. As most of developed V2G studies don’t consider real world input data or/and EV battery aging cost in system modeling and evaluation, the present paper presents a reliable study as it considers a real life MG with in field measurement input data and appropriate battery degradation model. The adopted model represents a linear approximation with a minimum error value to make a suitable tradeoff of computational complexity and accuracy of obtained results. Economic assessment of the system according to the proposed energy management is performed, where results indicate that the V2G system assisted the MG operation during high electricity price period and achieved economic profit to EVs owners. According to numerical results, V2G energy trading achieved 29.90 EUR of gross selling revenues with only 4.46 EUR as battery degradation cost which makes a 16.41% average cost reduction of daily MG operation cost.

Keywords

Micro-grid electric vehicle operation cost vehicle to grid battery wear

Introduction

Due to continuously growing electric energy demands, policy makers, utilities and research communities have been constantly seeking to increase production levels. Renewable energy sources (RES) are currently one of the most promising solutions to support electric energy production, however several aspects have to be considered regarding their integration into the power system.^1–4 Owing their intermittent nature, RESs are not flexible and their production rates depend strongly on weather conditions, which is not convenient for grid operations in terms of power balance and quality.⁵ Thereby, energy storage represents a key feature of nowadays power system since it ensures a more efficient RES integration.^6–8

Electric Vehicles (EVs) production and utilization are having increasing rates, their integration into the power system is strongly enhanced by stakeholders.^9–11 Through unidirectional charging, EVs are considered as flexible loads, where some services can be drawn by controlling the batteries charging rate. Furthermore, aggregated EVs can be considered as means of mass energy storage under the Vehicle-to-Grid (V2G) concept. In this case, energy can circulate from and to the vehicles according to controlled bidirectional charging scheme.^12–14 As vehicles are parked 95% of the time daily, the main idea is that the power grid can profit from the available energy during the long hours of parking time.¹⁵ Research communities are making a lot of efforts in studying the V2G capability in supporting the grid regarding power production and ancillary services. However, EV owner satisfaction, stochastic nature of EVs characteristics and battery degradation issues are the main challenges for real life implementation.^16–22

Yet, the management objectives of a V2G system can be classified under three levels. The first and basic priority has to be addressed to the EV owner, like providing the desirable amount of battery energy when leaving the charging station, or a minimum battery state of charge (SOC) not to be exceeded at each time. In addition, EV time availability has to be considered in the system planning. At a second level, EVs bidirectional power exchange has to be controlled in a way to prevent grid instabilities, i.e. avoiding EVs charge during peak load demand periods. At a final level, V2G energy can be used for grid support services. According to related research, many papers have proven the ability of V2G to provide ancillary services and peak power to the grid.

The modeling of photovoltaic power generation and EVs charging on city-scale was reviewed in.²³ The integration of photovoltaic and wind energy associated under the V2G concept was investigated in.^24,25 The study of hydrogen fuel cell EVs integration with V2G technology, photovoltaic power and a residential building was proposed in.²⁶ Voltage and frequency stability by means of V2G technology were considered by authors in.^27–30 Controlled power exchange to reduce peak load and load variance were also studied in.^31,32 On the other hand, several research studies were interested in the economic value of EVs integration into the power system. Authors in^11,33 investigated current policies and economic value of EVs integration as future electric energy source in China and South Korea. V2G system simulation associated with peak shaving uncertainty management and battery degradation mitigation are considered in.^34–36 Economic evaluation of V2G energy considering impact of charging demand on the grid operation and EV sharing are studied in.^37,38

In this context, the present study proposes a cost effective EVs integration study in the campus Micro-Grid (MG) of Malta College of Arts, Science and Technology (MCAST). The MG under study takes part of the 3DMICROGRID (3DMG) project. It represents a real life MG dedicated for research activities including a photovoltaic (PV) generation and a diesel generator as renewable and conventional energy sources and a battery as a storage system. The bidirectional charging station components modeling and control are described and implemented. A control algorithm is developed to manage the charging/discharging power flows in a cost effective way according to batteries available time and energy. The main contribution of the proposed study is to perform reliable management and cost analysis of V2G energy trading using real world input data and appropriate model of EVs batteries aging. A V2G system is characterized by a high amount of data exchange, yet in order to reduce computational complexity, previous studies ignore or use a simple estimate of battery wear. Consequently, operation cost is under evaluated and system economic assessment present non reliable results. Energy wear coefficient is used to estimate battery wear due to V2G cycling with a minimum prediction error and related cost is supposed to be refunded to EVs owners when the aggregator sells energy to the MG. Furthermore, EVs owners satisfaction level is considered, where final state of charge is predefined to secure a sufficient amount of stored energy. The proposed management proves that V2G system integration is able to reduce MG daily energy trading cost considerably and to make an economic profit for EVs owners.

The rest of the paper is organized as follow. In the next section, the adopted MG architecture will be described. The design of the proposed V2G charging station and system components will be detailed in the Design of the V2G charging station section. The Day-ahead operation planning of the V2G coupled MG section presents the day ahead operation planning of the V2G coupled MG, where cost effective charging of EVs is considered. The economic assessment of the V2G system integration taking into consideration batteries wear cost is investigated in the Economic assessment of the V2G operation section. Simulation results and discussions are presented in the Simulation results and discussion section, and finally the conclusion is given in the last section.

MCAST micro-grid architecture

3DMICROGRID project framework

3DMICROGRID (3DMG) is an ENTRAMED funded research project that aims the Design, Development and Demonstration of two future-proof active smart MGs located in two university campus in Jordan and Malta. The proposed architecture proposes the integration of small to medium size energy sources and loads. The overall objective of this project is to provide a real life MG for developing new control and energy management techniques for optimal system performance. In field data measurements have been performed for component sizing and design and system models have been developed according to two techniques. Electromagnetic transients modeling (EMT) is implemented to investigate dynamic system response to transient events and phasor (RMS) modeling enables day-ahead power related performance studies. In this paper, the MCAST MG is adopted as a case study, where the general architecture is illustrated by Figure 1. Phasor simulation models are used to enable a daily operation study under 1-minute resolution.

Figure 1.

MCAST Micro-Grid single line diagram.

The MCAST campus consists mainly of three renovated buildings over an underground car park. Buildings are supplied via the 11 kV distribution network, where buildings D and F are connected to substation 2 (11/0.4 kV) and building J is connected through substation 1. The substation have a circular configuration for satisfying the N-1 criterion to ensure the MG resilience under component failure. The connection with the main grid is ensured by a single controllable breaker that represents the single point of common coupling (PCC).

Load demand

The load demand of the MCAST is based mainly on the consumption of the three campus buildings. The elaborated model enables load categorization according to their priority: essential, non-essential and thermal loads. Essential loads represents high priority loads that cannot be shed. Non-essential loads can be shed under crucial states, and thermal loads that mainly include AC conditioning represent the least essential category where curtailment can take place according to system operator request.

Energy generation

The studied MG includes conventional/flexible and renewable/non flexible distributed energy generation units. In particular considering the suitable solar irradiation, rooftop PV generation units are installed in each building. According to the actual in field measurements, the existent PV generation does not satisfy the load demand. Furthermore, considering the intermittent nature of renewable energy sources, conventional generation has to be also included. A diesel generator is installed in building J to ensure frequency stability and load satisfaction during islanded operation of the MG.

Energy storage

For an increased flexibility, a battery is installed in building D as a stationary storage system. The battery system absorbs the excess PV generation, and restores it in order to assist the diesel generator during islanded mode. Through proper control strategies, the battery system can also improve the MG operation profit by participating in energy trading according to variable electricity pricing.

Design of the V2G charging station

Bidirectional fast charging concept

Within V2G framework, EVs represent short-term energy storage systems, where energy can be exchanged among the grid and EV batteries to profit from aggregated stored energy during their parking time. Public and workplace fast charging stations provide energy from electric grid in AC or DC high voltage forms. With AC charging the DC/DC converter is included on-board the vehicle, whereas with DC charging, the DC/DC converter is installed in the charging station.³⁹ As the V2G system under study is included within a small scale system, and in order to minimize the infrastructure cost, AC fast charging is considered. Here, AC voltage is converted into DC voltage on-board the vehicle and thus no additional AC/DC converters are required in the charging station. This charging method is considered more efficient regarding overall V2G system cost since investment capital cost is significantly reduced.

V2G system components

A V2G system includes two key subsystems: EVs fleet block and a bidirectional charger which includes an AC/DC converter for grid interface (grid-side converter) and a bidirectional DC/DC converter (EV-side converter). The control scheme represents a key feature for proper system operation and includes a grid side controller (GSC) and a local controller usually know as aggregator in the EV research community.⁴⁰

EV fleet

The main role of the EV fleet is to provide the necessary input data for the aggregator, mainly the number of EVs of each profile, starting and ending parking times and the associated state of charge values. Since the present study represents a workplace charging station, different car profiles are considered according to daily work hours. Yet, EVs owners’ behavior is classified into three categories: owners that are present only during the morning working hours, owners that are present only during the afternoon working hours and owners that are present for the whole workday. Initial and desired batteries state of charge are specified according to each EV profile characteristics.

Grid-side converter

The grid side power converter represents the interface of the grid and converts the high voltage AC power into high voltage DC power. During battery charge mode, the converter acts as a rectifier that converts the AC power from the grid to DC power and vice-versa in discharge mode. The converter proper operation is secured by a grid side controller connected to the MG and performing two main roles. It assigns a proper amount of power for EVs charge/discharge modes. Moreover, it receives the reactive current reference signal generated by the aggregator as displayed by Figure 2 and enables reactive power support. Reactive power produced by the grid-side converter is injected to the PCC to support the grid, where the reference signal is generated by the EV-side controller.

Figure 2.

V2G system design.

EV-side converter

EV-side converters ensure the connection of EVs with the DC bus. Since both charging and discharging modes are considered within the V2G concept, a bidirectional DC/DC converter is required. In this study, a controlled current source is explored as an EV-side converter owing its ability to further integrate local controllers.

Aggregator. The aggregator which represents the EV side controller is a key component in the V2G system design. According to Figure 2, an aggregator receives data from the grid operator regarding reactive power request, estimated load demand and electricity price. It also receives signals from the EV fleet considering, availability, actual power, initial and desired final battery SOC.^41,42 By analyzing the data received from both grid operator and EV fleet, the aggregator determines a scheduling for the charge/discharge of the EVs, according to the system objectives. Furthermore, as the grid-side controller does not interact with the grid operator, reactive power request is received by the aggregator that generates the reference signal and send it to the grid-side controller. Daily operation profit, power quality and ancillary services can be drawn by efficient operation of the aggregated EVs power.

Day-ahead operation planning of the V2G coupled MG

MG operation modes

For the scope of this study, the interconnected MG operation mode is considered and the diesel generator is supposed to be in a standby mode, i.e. no power will be requested from the latter. The MG power $P_{M G}$ exchanged with main grid at the PCC point can be determined knowing different components power flows:

P_{M G} = P_{P V} + P_{bat} + P_{V 2 G} - P_{load}

(1)

Where $P_{P V}$ represents the total PV generation power, $P_{bat}$ is the battery storage system power, $P_{V 2 G}$ represents the sum of all EVs power included in the V2G system and $P_{load}$ is the load power demanded by the campus building.

During their parking time, EVs’ batteries, are supposed to provide the same services that the battery storage system provides to the power system. Furthermore, the total capacity of a V2G system highly exceeds the battery capacity due to the aggregated number of EVs. Thus, the battery and the V2G system are controlled not to operate simultaneously and day-ahead operation is classified into 3 main modes according to V2G availability:

Mode1 (Battery mode): This mode covers the early morning hours before the arrival of EVs in the charging station. During this period, only the battery system is available as a storage device assisting the MG operations.

Mode2 (V2G mode): This mode is activated during the presence of EVs in the charging station. The MG can profit from aggregated EVs stored energy, whereas the battery is in a standby mode and is not included in the system operation until the departure of EVs.

Mode3 (Battery mode): After the EVs leave the parking station, the battery mode is activated again for system services.

V2G system control scheme

As previously explained in the Energy generation section, the control of EVs active and reactive power flows are performed mainly by the aggregator according to utility grid requirements and V2G system operational limitations. Here, a control algorithm included in the aggregator is implemented to manage V2G energy trading aiming to achieve an economic benefit.

As an input for the aggregator block, plug state signal indicates the presence of a vehicle in the parking station. A plug state equal to ‘1’ indicates that the vehicle is parked and plug state equal to ‘0’ indicates that the EV is not present in the station. The SOC signal indicates the current value of each EV battery state of charge. In particular, when an EV enters the parking station, the initial SOC value must be sent to the aggregator for deciding the appropriate operation mode. According to daily distributions of load demand and electricity price sent from the MG operator, the aggregator manages the available power to maximize the economic profit achieved from energy trading. The control of active power flows is performed under different schemes according to each EV profile specifications. Namely, EVs of the same profile, having identical SOC and parking duration characteristics are managed evenly. It should be mentioned that at departure time, each vehicle is expected to have a sufficient amount of energy represented by the desired battery SOC predefined by EV owners.

The proposed power management algorithm illustrated by Figure 3 involves three basic modes: charge, discharge and standby according to instantaneous battery state of charge and electricity price. First of all, the algorithm verifies whether the EV is plugged or not. Afterwards, the battery SOC value is compared ${SOC}_{thr}$ which represent a threshold reference value of charge and discharge modes. This value depends on the time and energy availability of the vehicle. Hence, each EV profile admits a different value of ${SOC}_{thr}$ . The second decision making parameter is the electricity price $E P$ . Likewise, the electricity price is compared to a reference value ${E P}_{thr}$ to decide which mode is suitable. At the end of their parking time, owners’ satisfaction must fulfilled by ensuring a sufficient remaining energy for subsequent city driving needs.

Figure 3.

Active power management flowchart for EV profiles.

Discharge mode: EV discharge is performed when both battery SOC and electricity price $E P$ are greater than reference values ${SOC}_{thr}$ and ${E P}_{thr}$ .

Charge mode: Charge mode is activated when both current battery SOC and electricity price $E P$ are inferior to reference values ${SOC}_{thr}$ and ${E P}_{thr}$ , respectively. This mode is activated mainly to ensure that EVs remaining energy satisfy the owners’ needs at the end of parking time.

Standby mode: This mode is activated if charge and discharge mode conditions are not satisfied or when the EV is not parked in the charging station.

After deciding each EV profile operation mode, the corresponding charging/discharging powers of each EV and EV profile $P_{E V}$ and $P_{p r}$ are calculated as follows:

P_{E V} (i, t) = Charge (i, t) {c h}_{rate} + Discharge (i, t) {dis}_{rate}

(2)

P_{p r} (i, t) = \sum_{t = 1}^{t_{f}} P_{E V} (i, t)

(3)

Where $Charge$ and $Discharge$ are Boolean decision variables assigned by the control algorithm described in Figure 3, ${c h}_{rate}$ and ${dis}_{rate}$ indicate the charging and discharging rates of the batteries in kW/s. and $t_{s}$ is the final time step expressed in s.

Eventually the instantaneous V2G power exchange with the utility grid is calculated as the sum of the exchanged power of each EV profile:

P_{V 2 G} (t) = \sum_{i = 1}^{N_{p r}} P_{p r} (i, t)

(4)

Where $N_{p r}$ is the number of EV profiles.

The battery SOC of each EV is estimated according to the Coulomb counting method. The adopted formula integrates the battery exchanged power, which is useful for applications where the transferred energy is important such as EV applications⁴³:

{SOC}_{E V} (i, t) = {SOC}_{E V} (i, t - 1) - \frac{1}{C_{bat}} \int_{t - 1}^{t} P_{E V} (t) d t

(5)

Where $C_{bat}$ is the battery capacity in kWh and $P_{E V}$ is the exchanged power.

Economic assessment of the V2G operation

In order to evaluate the (economic profit of the V2G system operation/impact of the V2G system on MG operation cost), V2G energy selling revenues and buying cost have to be estimated in the first place.⁴⁴ In the present study, the price of V2G energy considering battery wear cost will be addressed.

V2G energy cost

Several studies have investigated the V2G electricity cost, and concluded that it represents mainly the sum of the rated electricity price and the battery degradation cost, although in some studies, the infrastructure or investment cost is also included.^45–47 In this study, infrastructure cost is not considered since the main objective of the MCAST project is to continuously provide a suitable experimental research environment. Furthermore, the adopted AC charging strategy does not require high infrastructure cost as AC/DC converters are integrated within the EV and not in the station. Yet, V2G electricity price $E P_{V 2 G}$ can be estimated as follows:

{E P}_{V 2 G} = E P + C_{d_b}

(6)

Where $E P$ represents the rated electricity price in (EUR/kWh) given by utilities and $C_{d_b}$ is the battery degradation cost in (EUR/kWh) due to extra cycling of V2G operations.

According to the model presented in,⁴⁸ the battery degradation cost can be expressed as the battery capital cost by the actual battery energy life ratio:

C_{d_b} = \frac{C_{C}}{L_{V 2 G}}

(7)

Where $C_{C}$ is the battery capital cost in (EUR) and $L_{V 2 G}$ is the battery actual life due to V2G extra cycling expressed in (kWh).

Battery energy wear cost estimation

Energy wear reduces the battery lifespan and occurs mainly due to heating, depth of discharge (DOD), peak charging currents and frequent charging/discharging. As batteries are nowadays equipped with thermal management and current control loop, the temperature and peak current effects can be neglected. On the other hand, rather than considering the DOD, it is more reliable to consider the overall amount of energy drawn from a battery.⁴⁹ Yet, battery degradation can be estimated giving the exchanged energy.

Energy wear can be expressed as the loss of energy life for a battery cycled from an initial DOD equal to $D_{i n}$ to a final DOD equal to $D_{f n}$ may be described by the general equation (8):

E_{w} = E || D_{i n} K_{D, i n} - D_{f n} K_{D, f n} ||

(8)

Where $E$ denotes the exploitable energy at the beginning of the charge/discharge cycle and $K_{D, i n}$ and $K_{D, f n}$ are discharge coefficient at initial and final DODs. These coefficients account for loss of energy life in terms of total processed energy and measured in (kWh/kWh).

According to the approximation performed in⁵⁰ and assuming the common ACC-DOD characteristic,⁵¹ capacity loss as a function of DOD is evaluated. The obtained characteristic is almost linear and the energy wear $E_{w}$ of a single charging or discharging process can be estimated as a linear function of initial and final DODs as follows:

E_{w} = E K_{w} || D_{i n} - D_{f n} ||

(9)

Where $K_{w}$ is the global wear coefficient expressed in (kWh/kWh). For each charged or discharged kWh, the battery loses $K_{w}$ kWh of its capacity.

Eventually, the actual battery life throughput energy in kWh due to V2G cycling can be estimated as:

L_{V 2 G} = L_{E} - {2 E}_{w}

(10)

It should be mentioned here that since the amount of energy drawn from the V2G system will be compensated later whether in the studied charging station or in another facility (public charging station, home.), i.e. each V2G discharging cycle is associated with a charging cycle. Hence, a factor of 2 is added in equation (10) to calculate the energy wear as equation (9) considers a single charge or discharge mode.

$L_{E}$ is the nominal battery life throughput energy in kWh and represents the energy that it can deliver during its lifespan measured at a specific depth of discharge, which is generally considered equal to 1. The manufacturer usually provides the battery cycle life that represents the number of charge/discharge cycles the battery can deliver before losing 20% of its initial nominal capacity. Yet, the battery energy life is estimated as the following:

L_{E} = L_{c} C_{bat} DOD

(11)

Where $L_{c}$ is the battery cycle life in cycles, and $C_{bat}$ is the battery initial nominal capacity in kWh.

V2G operation benefit assessment

In order to evaluate the added economic value of the V2G system, the net benefit achieved through energy trading operations has to be calculated. In this study the net benefit given by equation (12) represents the gross benefit coming from energy trading subtracted by the total batteries degradation cost $C_{d_{V 2 G}}$ given by equation (13):

{benefit}_{net, V 2 G} = {benefit}_{gross, V 2 G} - C_{d_{V 2 G}}

(12)

C_{d_{V 2 G}} = \sum_{i = 1}^{N_{p r}} \sum_{j = 1}^{N_{E V}} C_{d_b} (i, j)

(13)

Where $C_{d_b}$ is the battery degradation cost calculated from equation (7), $N_{p r}$ and $N_{E V}$ are the number of EV profiles and the number of EVs in each profile, respectively.

The gross benefits of the V2G system operation can be calculated as the revenues from selling energy minus the cost of buying it from the grid:

{benefit}_{gross, V 2 G} = {revenues}_{V 2 G} - {cost}_{V 2 G}

(14)

The total energy trading revenues and cost of the V2G system comes from the selling revenues and the buying cost of each EV profile i and are expressed by equations (15) and (16), respectively:

{revenues}_{V 2 G} = \sum_{i = 1}^{N_{p r}} {rev}_{sel} (i)

(15)

{cost}_{V 2 G} = \sum_{i = 1}^{N_{p r}} {Cost}_{buy} (i)

(16)

{rev}_{sel} (i)

are the selling revenues of each EV profile i. It is expressed as the product of the injected energy to the grid

E_{inj}

by the V2G electricity price

{E P}_{V 2 G}

{rev}_{sel} (i) = \sum_{t = t i}^{t f} E_{inj} (i, t) {E P}_{V 2 G} (t)

(17)

{Cost}_{buy} (i)

is the cost of buying energy from the grid of each EV profile i and is calculated as the product of the absorbed energy from the grid

E_{abs}

by the rated electricity price

E P

{Cost}_{buy} (i) = \sum_{t = t i}^{t f} E_{abs} (i, t) E P (t)

(18)

Simulation results and discussion

Preliminary data

In this section, MCAST MG preliminary data used for simulations are represented. Adopted data are collected based on in field measurements and design parameters represent real world values. Three PV generation units of 21 kW peak each are installed in each campus building, making a total of 63 kW peak renewable energy production. Figure 4 represents a typical daily PV generation (Figure 4(a)) and a typical workday load demand distribution (Figure 4(b)). The rated electricity price given by utilities is presented by Figure 5. The stationary battery storage system is characterized by a15kW rated power and 20 kWh capacity.

Figure 4.

Typical daily distributions of PV generation and load demand.

Figure 5.

Daily electricity price distribution.

In this study, the considered EV fleet will include a limited number of vehicles since it is associated with a small scale MG. The fleet is composed of a total number of 15 vehicles divided equally into three profiles, i.e. $N_{p r} = 3$ and $N_{E V} = 5$ according to the characteristics mentioned in Table 1. The initial battery SOC value of profiles 1 and 3 is supposed to be equal to 90%, since in the morning, EV owners are supposed to come from their homes with almost a fully charged vehicle. Whereas, the initial SOC of profile 2 is assumed to be equal to 60%, since owners are supposed to come later in the afternoon and may have done some activities.

Table 1.

EV profiles characteristics.

EV profile	Start of parking time	End of parking time	Initial SOC	Final SOC
Profile 1	8 h	12h	90%	≥60%
Profile 2	14 h	16h	60%	≥60%
Profile 3	8 h	18h	90%	≥60%

The initial battery SOC for profile 2 is supposed to be equal to 60% since EVs owners might have used their vehicles for morning activities. The final desired battery SOC is set to be greater or equal to 60% for all the car profiles. With this desired value, the remaining energy, is amply sufficient for after work city driving typical activities.

Renault Zoe model is adopted in this study, where AC fast charging and battery characteristics are summarized in Table 2.⁵² The considered EV model is equipped with an on-board charger and provides a flexible charging scheme with the ability to charge at AC slow and fast levels and DC rapid level.

Table 2

EV characteristics.

Model reference	Renault Zoe ZE50 R110
Battery usable capacity	52 kWh
Maximum Charger Power (fast charging)	22 kW AC
Time to charge (0 to 90%)	2 h 35 min
Vehicle Consumption	16.5 kWh/100 km
Electric Range	390 km
Life span (cycles)	1500

Simulation results

A dynamic phasor model has been created and tested within MATLAB/Simulink to simulate the day-ahead operation of the MCAST MG. Load demand, PV generation and electricity price represent input data to the V2G system under study, where their typical distributions are presented in the previous paragraph. The EV fleet model has been implemented according to the predefined profiles and EV parameters given by Tables 1 and 2. The day-ahead simulation of the MG associated with the proposed V2G system is performed according to the three operation modes described in the MG operation modes section. And the available energy of each EV profile is managed according to the flowchart given by Figure 3.

The MG day-ahead simulation results are displayed in Figure 6. As noted in Figure 6(a), the MG active power $P_{M G}$ presents both negative and positive operation modes. During early morning and late afternoon hours $P_{M G}$ is negative i.e the MG is not able to satisfy load demand. During midday hours, in particular when EVs are parked in the charging station, the MG is able to satisfy load demand and injects excess power to the utility grid. The MG voltage is regulated through reactive power control (Figure 6(b) and presented in p.u by Figure 6(c) Observed results display the stability of the MG during different operation modes, where voltage deviation is inferior to 0.1% of the nominal voltage.

Figure 6.

MG results: (a) Voltage, (b) Reactive power, (c) Active power.

Results for the battery power $P_{bat}$ and state of charge ${SOC}_{bat}$ are displayed in Figure 7. During early hours of the day, the battery will be charged since its initial battery state of charge reaches its minimum, i.e 20% and the electricity price presents the lowest energy cost. Once its SOC reaches 90% and during all the hours where EVs are parked, i.e. during [8h, 18 h], the battery operates in an idling mode where no power exchange is performed. The battery energy is discharged starting from 18 h to support the MG and the SOC reaches its minimum allowed value (20%) after only one hour, i.e 19 h. During this period, the MG power is still negative, which denotes that the capacity of the installed stationary storage system, i.e. the battery, is not sufficient for the MG load requirements.

Figure 7.

Battery results: (a) State of charge, (b) Active power.

Figure 8(a) displays the MG active power with and without V2G. The total V2G system active power distribution $P_{V 2 G}$ , which is the sum of the injected/absorbed powers of the EV profiles during their parking time is presented by Figure 8(b). In accordance with energy and time limitations of each profile, the V2G system injects power to the utility grid during high electricity price time interval following the predefined threshold value ${E P}_{thr}$ which is equal to 0.22 EUR/kWh in the present case study.

Figure 8.

(a) MG active power with and W/O V2G, (b) V2G system active power.

Figure 9.

EV profile 1 results: (a) State of charge, (b) Active power, (c) Plug state.

Figure 10.

EV profile 2 results: (a) State of charge, (b) Active power, (c) Plug state.

Figure 11.

EV profile 3 results: (a) State of charge, (b) Active power, (c) Plug state.

EVs included in profile 2 are supposed to start their parking time with a $SOC$ equal to 60% since owners may be doing some activities before coming to their work place. This profile injects power to the utility grid during the highest electricity price, i.e $E P = 0.24$ EUR/kWh. Later when electricity price reaches its local minimum during EVs availability time, i.e $E P = 0.21$ EUR/kWh, the batteries are controlled to restore energy back from utilities. The charging process stops when batteries $SOC$ reaches 60% fulfilling the desired final value that ensures owners’ satisfaction.

EV profile 3 presents the most profitable characteristics in terms of energy and time availability for energy trading. Here, EVs are supposed to start their parking phase with almost charged batteries and are available during the whole work hours. The profile sells V2G energy to utilities during the maximum electricity price period until batteries $SOC$ reaches 60%.

It should be mentioned that the final batteries SOC set at 60% for all EV profiles guarantees owners comfort while participating in V2G operations. Based on the parameters given in Table 2, the remaining energy, calculated up to 20% SOC, allows a range of 126 km, which is very adequate for city driving activities.

Numerical results analysis

In this section energy trading numerical results and V2G economic impact on the MG operation cost are calculated. Table 3 presents day-ahead MG accumulative trading results in terms of energy and cost without EVs integration. Similarly to the V2G scenario, PV generation, load demand and electricity price distributions are not changed, also the battery storage system is controlled in the same way. In this study, MG energy trading cost represents energy buying cost minus selling revenues. Recorded MG operation cost reaches 54.90 EUR for a day-ahead operation without EVs integration. This value will be used as a benchmark to calculate MG cost reduction owing to EVs integration.

Table 3.

MG operation results: energy and cost without V2G system.

Energy injected to the grid (kWh)	153.34
Selling revenues (EUR)	34.23
Energy absorbed from the grid (kWh)	440
Buying cost (EUR)	89.13
MG energy trading cost (EUR)	54.90

Table 4 displays V2G performance results regarding accumulative exchanged V2G energy, batteries wear cost and energy trading costs. Batteries wear cost is calculated in accordance with the amount of the injected energy as previously explained in the article. Gross and net selling revenues represent revenues before and after deducting the batteries wear cost. Yet the accumulative daily V2G operation net benefit is calculated as the net selling revenues minus the buying cost, which is equal to 22.48 EUR.

Table 4.

V2G operation results: energy and cost.

V2G Energy injected to the grid (kWh)	105.22
Batteries wear cost (EUR)	4.46
Gross selling revenues (EUR)	29.90
Net selling revenues (EUR)	25.44
V2G Energy absorbed from the grid (kWh)	14.22
Buying cost (EUR)	2.96
V2G operation net benefit (EUR)	22.48

The gained V2G profit will be shared among the V2G system represented by EV owners and the MG operator. It should be mentioned that batteries wear cost will be compensated to the EV owners, yet the MG operator will benefit from net profits. In Table 5 we present the proportion of V2G system share in % and the amount of V2G and MG share of net benefits. V2G gross benefit share represents the effective amount of money that the V2G system will get, which is the net benefit added by the battery wear cost. Finally, we present the MG cost reduction $C_{red}$ in % owing to V2G integration calculated according to the next equation:

C_{red} (%) = \frac{C_{w o, V 2 G} - C_{w, V 2 G}}{C_{w o, V 2 G}} * 100 = \frac{P_{M G, s h}}{C_{w o, V 2 G}} * 100

Table 5.

Benefit share and cost reduction.

V2G system net benefit share proportion (%)	80	70	60	50	40
V2G system net benefit share (EUR)	14.42	12.61	10.81	9.01	7.21
V2G system gross benefit share (EUR)	18.88	17.07	15.27	13.47	11.67
MG net benefit share (EUR)	3.60	5.41	7.21	9.01	10.81
MG operation cost reduction (%)	6.56	9.84	13.13	16.41	19.69

Where $C_{w o, V 2 G}$ represents the MG operation cost without V2G integration. According to the results shown in Table 3, $C_{w o, V 2 G} = 54.90$ EUR. $C_{w, V 2 G}$ denotes the MG operation cost with V2G integration and $P_{M G, s h}$ is the MG share from net benefit.

Obtained results displays the V2G system capability to improve economic performance of the studied MG according to different share ratios. In particular, with 50% benefit share, the average MG operation cost will be reduced by 9.01 EUR per day which is equivalent to 16.41% reduction.

Conclusion

In this paper, a day-ahead planning of the MCAST micro-grid is developed under real world data of energy production, demand and electricity price, and realistic EV profiles. Economic impact of EVs integration was highlighted under the so called V2G concept, where controlled energy trading proved to be able to reduce daily MG operation cost by an average value of 16.41% and 9.01 EUR net benefit for EVs owners. In particular, V2G energy trading is managed considering EVs batteries degradation cost in order to improve owners experience. Energy wear and its related cost were estimated using an appropriate degradation model with high accuracy and low computational burden where degradation cost was refunded to owners. As a future work, uncertainties associated with energy demand, PV production electricity price and EVs availability will be studied within an optimization framework.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors wish to thank the Scientific and Technological Research Council of Turkey (TUBITAK-215E373) and the Malta Council for Science and Technology (MCST) for their financial support and data sharing.

ORCID iD

Ahmet Onen

Rayhane Koubaa currently pursues her postdoctoral research at the LSEER laboratory (Electrical Systems and Renewable Energies) at the National Engineering School of Sfax, Tunisia. Rayhane does research in energy management and optimization for renewable energy systems.

Yeliz Yoldas is PhD student at AGU. Her research interest is AI, Optimization based power system.

Selcuk Goren is faculty member in Industrial Engineering, AGU. His Research Interests: Robust optimization, renewable energy systems, smart grids, production scheduling under uncertainty, stochastic modeling, simulation, applied probability, discrete optimization, metaheuristics.

Lotfi Krichen is a professor in Department of Electrical Engineering, National School of Engineering, University of Sfax, 3038 Sfax, Tunisia.

Ahmet Onen received the BSc degree in Electrical-Electronics Engineering from Gaziantep University in 2005. He received the MS degree in Electrical-Computer Engineering from Clemson University in 2010 and his PhD from Virginia Tech – Electrical and Computer Engineering Department in 2014. Prior to joining Abdullah Gul University as a faculty of Electrical-Electronics Engineering in 2014, he was a member of EDD company located in US to develop software for distribution companies. He is in the editor board of Electrical Engineering (Springer), Frontiers in Energy, El-Cezeri Journal of Science and Engineering. His research interests are Transmission Network and Smart Grids, Big Data in Power Systems, Renewables and Integrations, Power System Optimization, Cost-Benefit Analysis, Energy Storages, Demand Response, FACTs Devices and their application, Block Chain Applications, MicroGrid and Virtual Power Plants.

References

Perera

ATD

Nik

Mauree

et al. Electrical hubs: an effective way to integrate non-dispatchable renewable energy sources with minimum impact to the grid. Appl Energy 2017; 190: 232–248.

Kroposki

Johnson

Zhang

et al. Achieving a 100% renewable grid: operating electric power systems with extremely high levels of variable renewable energy. IEEE Power Energy Mag 2017; 15: 61–73.

Oree

Hassen

SZS

and Fleming

PJ.

Generation expansion planning optimization with renewable energy integration: a review. Renewable Sustainable Energy Rev 2017; 69: 790–803.

Srithiam

Somphop

and Tawatchai

Smart grid national pilot project in Mae Hong Son province, Thailand. Energy Environ 2015; 26: 23–34.

Janicek

Perny

Saly

et al. The role of smart grid in integrating the renewable energies in Slovakia. Energy Environ 2018; 29: 300–312.

McPherson

and Tahseen

Deploying storage assets to facilitate variable renewable energy integration: the impacts of grid flexibility, renewable penetration, and market structure. Energy 2018; 145: 856–870.

Argyrou

Christodoulides

and Kalogirou

SA.

Energy storage for electricity generation and related processes: technologies appraisal and grid scale applications. Renewable Sustainable Energy Rev 2018; 94: 804–821.

Zame

Brehm

Nitica

et al. Smart grid and energy storage: policy recommendations. Renewable Sustainable Energy Rev 2018; 82: 1646–1654.

Mahmud

Town

Morsalin

et al. Integration of electric vehicles and management in the internet of energy. Renewable Sustainable Energy Rev 2018; 82: 4179–4203.

10.

Knezovic

Marinelli

Zecchino

et al. Supporting involvement of electric vehicles in distribution grids: Lowering the barriers for a proactive integration. Energy 2017; 134: 458–468.

11.

Yang

and Sandu

Electric vehicles in China: a review of current policies. Energy Environ 2018; 29: 1512–1524.

12.

Zhou

Sun

Shi

et al. Robust energy scheduling in vehicle-to-grid networks. IEEE Network 2017; 31: 30–37.

13.

Aluisio

Conserva

Dicorato

et al. Optimal operation planning of V2G-equipped microgrid in the presence of EV aggregator. Electric Power Syst Res 2017; 152: 295–305.

14.

Shaukat

Khan

Ali

et al. A survey on electric vehicle transportation within smart grid system. Renewable Sustainable Energy Rev 2018; 81: 1329–1349.

15.

Liu

Kockelman

Boesch

et al. Tracking a system of shared autonomous vehicles across the Austin, Texas network using agent-based simulation. Transportation 2017; 44: 1261–1278.

16.

Choi

Han

et al. Reviews on grid-connected inverter, utility-scaled battery energy storage system, and vehicle-to-grid application-challenges and opportunities. In: IEEE transportation electrification conference and expo (ITEC), Chicago, IL, USA, 22–24 June 2017, pp.203–210. IEEE.

17.

Gough

Dickerson

Rowley

et al. Vehicle-to-grid feasibility: a techno-economic analysis of EV-based energy storage. Appl Energy 2017; 192: 12–23.

18.

Chen

Hardy

Harper

et al. Towards standardized vehicle grid integration: current status, challenges, and next steps. In: IEEE transportation electrification conference and expo (ITEC), Dearborn, MI, USA, 14–17 June 2015, pp.1–6. IEEE.

19.

Habib

Khan

Abbas

et al. A comprehensive study of implemented international standards, technical challenges, impacts and prospects for electric vehicles. IEEE Access 2018; 6: 13866–13890.

20.

Noel

De Rubens

Kester

et al. The technical Challenges to V2G. In: Vehicle-to-grid. Cham: Palgrave Macmillan, 2019, pp.65–89.

21.

Esther

Singh

Goswami

et al. Recent challenges in vehicle to grid integrated renewable energy system: a review. In: Second international conference on intelligent computing and control systems (ICICCS), Madurai, India, 14–15 June 2018, pp.427–435.

22.

Noel

De Rubens

Kester

et al. The economic and business challenges to V2G. In: Vehicle-to-grid. Cham: Palgrave Macmillan, 2019, pp.91–116.

23.

Shepero

Munkhammar

Widén

et al. Modeling of photovoltaic power generation and electric vehicles charging on city-scale: a review. Renewable Sustainable Energy Rev 2018; 89: 61–71.

24.

Rahbari

Vafaeipour

Omar

et al. An optimal versatile control approach for plug-in electric vehicles to integrate renewable energy sources and smart grids. Energy 2017; 134: 1053–1067.

25.

Zhao

Noori

and Tatari

Boosting the adoption and the reliability of renewable energy sources: Mitigating the large-scale wind power intermittency through vehicle to grid technology. Energy 2017; 120: 608–618.

26.

Robledo

Oldenbroek

Abbruzzese

et al. Integrating a hydrogen fuel cell electric vehicle with vehicle-to-grid technology, photovoltaic power and a residential building. Appl Energy 2018; 215: 615–629.

27.

De Forest

Mac Donald

and Black

DR.

Day ahead optimization of an electric vehicle fleet providing ancillary services in the Los Angeles air force base vehicle-to-grid demonstration. Appl Energy 2018; 210: 987–1001.

28.

Neofytou

Blazakis

Katsigiannis

et al. Modeling vehicles to grid as a source of distributed frequency regulation in isolated grids with significant RES penetration. Energies 2019; 12: 720.

29.

Peng

Zou

and Lian

Dispatching strategies of electric vehicles participating in frequency regulation on power grid: a review. Renewable Sustainable Energy Rev 2017; 68: 147–152.

30.

Kongjeen

and Bhumkittipich

Impact of plug-in electric vehicles integrated into power distribution system based on voltage-dependent power flow analysis. Energies 2018; 11: 1571.

31.

Tan

Ramachandaramurthy

Yong

et al. Minimization of load variance in power grids – investigation on optimal vehicle-to-grid scheduling. Energies 2017; 10: 1880.

32.

Mahmud

Hossain

and Town

GE.

Peak-load reduction by coordinated response of photovoltaics, battery storage, and electric vehicles. IEEE Access 2018; 6: 29353–29365.

33.

Lim

Kim

and Lee

Forecasting the future scale of vehicle to grid technology for electric vehicles and its economic value as future electric energy source: the case of South Korea. Energy Environment. Epub ahead of print 7 January 2020. DOI: 10.1177/0958305X19898283.

34.

Mehrjerdi

and Rakhshani

Vehicle-to-grid technology for cost reduction and uncertainty management integrated with solar power. J Clean Prod 2019; 229: 463–469.

35.

Martinez

and Yang

and Charging, power management, and battery degradation mitigation in plug-in hybrid electric vehicles: a unified cost-optimal approach. Mech Syst Signal Process 2017; 87: 4–16.

36.

Erden

Mithat

and Nuh

Adaptive V2G peak shaving and smart charging control for grid integration of PEVs. Electr Power Compon Syst 2018; 46: 1494–1508.

37.

Mozafar

Rezaei

Amini

MH.

and Moradi

Innovative appraisement of smart grid operation considering large-scale integration of electric vehicles enabling V2G and G2V systems. Electr Power Syst Res 2018; 154: 245–256.

38.

Zhang

and Shen

On the values of vehicle-to-grid electricity selling in electric vehicle sharing. Manuf Serv Oper Manag. Epub ahead of print 30 March 2020. DOI: 10.1287/msom.2019.0855.

39.

Levinson

and West

TH.

Impact of public electric vehicle charging infrastructure. Transp Res Part D Transp Environ 2018; 64: 158–177.

40.

Ul-Haq

Cecati

and Al-Ammar

EA.

Modeling of a photovoltaic-powered electric vehicle charging station with vehicle-to-grid implementation. Energies 2016; 10: 4.

41.

Honarmand

Zakariazadeh

and Jadid

Integrated scheduling of renewable generation and electric vehicles parking lot in a smart microgrid. Energy Convers Manage 2014; 86: 745–755.

42.

Sayed

and Gabbar

HA.

Electric vehicle to power grid integration using three-phase three-level AC/DC converter and PI-fuzzy controller. Energies 2016; 9: 532.

43.

Zhu

Xia

and Chiang

HD.

Coordinated sectional droop charging control for EV aggregator enhancing frequency stability of microgrid with high penetration of renewable energy sources. Appl Energy 2018; 210: 936–943.

44.

Kularatna

Rechargeable battery technologies: an electronic engineer’s view point. In: Energy storage devices for electronic systems. Rechargeable batteries and supercapacitors. 1st ed. 2015, pp. 29–61.

45.

Noori

Zhao

Onat

et al. Light-duty electric vehicles to improve the integrity of the electricity grid through vehicle-to-grid technology: analysis of regional net revenue and emissions savings. Appl Energy 2016; 168: 146–158.

46.

Ahmadian

Sedghi

Mohammadi-Ivatloo

et al. Cost-benefit analysis of V2G implementation in distribution networks considering PEVs battery degradation. IEEE Trans Sustain Energy 2018; 9: 961–970.

47.

Ghofrani

Arabali

Etezadi-Amoli

et al. Smart scheduling and cost-benefit analysis of grid-enabled electric vehicles for wind power integration. IEEE Trans Smart Grid 2014; 5: 2306–2313.

48.

Schuller

Dietz

Flath

et al. Charging strategies for battery electric vehicles: economic benchmark and V2G potential. IEEE Trans Power Syst 2014; 29: 2014–2022.

49.

Hoke

Brissette

Smith

et al. Accounting for lithium-ion battery degradation in electric vehicle charging optimization. IEEE J Emerg Sel Topics Power Electron 2014; 2: 691–700.

50.

Farzin

Fotuhi-Firuzabad

and Moeini-Aghtaie

A practical scheme to involve degradation cost of lithium-ion batteries in vehicle-to-grid applications. IEEE Trans Sustain Energy 2016; 7: 1730–1738.

51.

Han

and Aki

A practical battery wear model for electric vehicle charging applications. Appl Energy 2014; 113: 1100–1108.

52.

Renault. Renault Zoe technical specifications, http://cdn.group.renault.com/ren/gb/renault-new-cars/editorial/global-range/new-zoe/Renault-ZOE-eBrochure 2020.pdf.asset.pdf/1577096937232.pdf?modelAdminId=zoe-b10-ph2 (accessed 10 February 2020).