energies Article Techno Economic Analysis of Vehicle to Grid (V2G) Integration as Distributed Energy Resources in Indonesia Power System Muhammad Huda 1,2 , Tokimatsu Koji 1 and Muhammad Aziz 3, *,† 1 2 3 * † Department of Transdisciplinary Science and Engineering, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama 226-8503, Japan; hudz.muhammad@gmail.com (M.H.); tokimatsu.k.ac@m.titech.ac.jp (T.K.) PT Perusahaan Listrik Negara (PLN), Jl Trunojoyo Blok M 1/135 Kebayoran Baru, Jakarta 121XX, Indonesia Institute of Innovative Research, Tokyo Institute of Technology, 2-12-1 Ookayama, Tokyo 152-8550, Japan Correspondence: maziz@iis.u-tokyo.ac.jp; Tel.: +81-3-5452-6196 Current address: Institute of Industrial Science, The University of Tokyo, 4-6-1 Komaba, Tokyo 153-8505, Japan. Received: 27 January 2020; Accepted: 28 February 2020; Published: 4 March 2020 Abstract: High penetration of electric vehicles (EVs) leads to high stress on a power grid, especially when the supply cannot cover and actively respond to the unpredictable demand caused by charging EVs. In the Java-Madura-Bali (JAMALI) area, Indonesia, the capability of the grid to balance its supply and demand is very limited, and massive EV charging additionally worsens the condition because of unbalanced load profiles. Ancillary services of EVs have led to the idea of utilizing EV batteries for grid support, owing to their high-speed response to the fluctuating power system. In this study, a techno-economic analysis of the vehicle-to-grid (V2G) system in the JAMALI grid is conducted in terms of the changes in the feed-in tariff schemes, including regular, natural, and demand response tariffs. The results show that by utilizing EVs, the supply during peak hours can be reduced by up to 2.8% (for coal) and 8.8% (for gas). EVs owned by business entities as operating vehicles with a natural tariff show the highest feasibility for ancillary services, and can potentially reduce the cost of charging by up to 60.15%. From a power company perspective, V2G also potentially improves annual revenue by approximately 3.65%, owing to the replacement of the fuel. Keywords: vehicle-to-grid; load leveling; electric vehicles; driving pattern; frequency regulation 1. Introduction The use of renewable energy has been a growing trend in the world for many years, especially since the Paris Agreement, where all parties have pledged to keep the global temperature below 2 β¦ C [1]. However, the rising global population and the advancement of technology have led human beings to consume more energy, and therefore, there is an urgent need to find a sustainable solution [2]. A green scenario has been set up by the International Energy Agency (IEA). The scenario recommends mitigation, countermeasures, and actions to reduce CO2 emission in the atmosphere to below 450 ppm through several policies regarding the use of fossil fuels, and supporting the sales of hybrid cars and electric vehicles (EVs), which have been projected to increase by approximately 40% by 2030 [3]. The trend of EV adoption is believed to be one of the solutions for a green solution in the transportation sector. In general, EVs include plug-in hybrid electric vehicles (PHEVs), battery electric vehicles (BEVs), and fuel cell electric vehicles (FCEVs) [4]. The penetration of these vehicles increases because of their beneficial characteristics compared to internal combustion engine vehicles (ICEVs), including low maintenance, high performance and efficiency, and zero tailpipe emission, especially Energies 2020, 13, 1162; doi:10.3390/en13051162 www.mdpi.com/journal/energies Energies 2020, 13, 1162 2 of 16 when they are integrated with renewable energy resources [5]. Furthermore, some researchers have studied the utilization of an EV battery as energy storage in the power system, especially to balance demand and supply, while maintaining the quality and stability of the grid [6,7]. However, the high interest in EVs leads to a high demand in electricity when EVs are charged massively at the same time (dumb charging, uncoordinated charging), causing an imbalance in the power system [8]. Conversely, if both charging and discharging can be coordinated and scheduled, EVs can potentially support the grid, as they can be considered huge energy storage. Thus, the management of EV charging and discharging has become an important issue for developing a smart energy system utilizing mobile energy storage in the near future. The adoption of renewable energy in power grids has had a positive impact on emissions reduction. However, it also potentially leads to a significant problem due to instability, as most renewable energies have unpredictable fluctuations [9]. The Indonesian government has set a target of a high portion of renewable energy, of approximately 27% by 2030, including geothermal, wind, biomass, and solar energies. However, at the same time, the government has also accelerated the construction of coal-based power plants to match fast-growing economic growth [10,11]. Moreover, the peak of daily electricity demand still occurs at a certain time (evening to night), causing a large gap between peak and valley in the load profile [12]. The difference between peak and off-peak becomes an oversupply problem that can threaten the electricity market between power producers and transmission system operators (TSO) [13]. The utilization of EVs together with the adoption of renewable energy and feed-in tariffs (FIT) in power grid management are estimated to increase the popularity of the vehicle-to-grid (V2G) concept currently popular in several European countries and the US. Implementation of V2G in Indonesia is considered challenging and attractive, as Indonesia is a developing country with several limitations in the power grid [14]. V2G affects the possible synergetic gain for both the vehicles and the grid. For the grid utilities and operator, EVs are expected to supply the back-up electricity, shift the electricity load, and respond quickly to balance the grid. Compared to other EVs, the BEV generally has a significantly larger battery capacity. Batteries represent excellent energy storage, having high energy efficiency and a faster response time than a pumped hydropower plant [15]. Currently, BEVs commonly adopt lithium-ion batteries, owing to their high energy efficiency (minimum energy loss) and energy density for facilitating a longer travel distance [16]. There are several advantages of lithium-ion batteries compared to other types of batteries, including high energy density, effective dimensions, low self-discharge, and a longer lifespan [17]. However, the number of instances of charging and discharging potentially impacts battery degradation, leading to a decrease in battery lifetime. This is one of the most challenging issues in the application of V2G [18]. Uddin et al. [19] studied this issue and stated that with the advancement of technology and intelligent control systems of V2G, the battery degradation could be reduced, leading to a relatively long lifetime of lithium-ion batteries. Their study is a response to another study conducted by Dubarry et al. [20] that studied the durability and degradation of the battery caused by bi-directional charging and discharging, including ancillary services provision. In several countries, energy storage has already existed. For example, a pumped storage power plant pumps water from any low to high reservoirs when the demand is low, and then utilizes this water to generate electricity during high demand, as well as to balance the power grid when imbalances occur in the grid from fluctuating renewable energy, etc. [21]. However, there are issues related to pumped storage, as it is limited by geographical conditions, and only a few of the pumped storages can respond quickly to the grid imbalances [22]. A power system frequently requires a very fast response that is better than that of pumped hydroelectricity, which can be covered by other solutions, such as a flywheel [23]. Compressed-air energy storage (CAES) is also another idea to level the gaps in peak and off-peak demands [24]. However, CAES also faces some problems related to low energy efficiency and response time. Several developing countries have an insufficient capacity of installed power storage in addition to low capabilities for controlling power plants, leading to difficulties to balance the power grid responsively. The massive adoption of BEVs in the future opens an opportunity to utilize them to Energies 2020, 13, 1162 3 of 16 provide ancillary services (e.g., V2G). The aggregating and scheduling strategies for widely-distributed EVs are considered as important keys to the success of V2G [25]. Several possible ancillary services include load leveling, frequency regulation, energy storage, voltage regulation, and congestion mitigation at transmission and distribution lines. Nunes and Brito [26] found that EVs have great potential to replace the gas-based power plants to balance the grid. Many studies have evaluated ancillary services utilizing EVs, especially in developed countries, including France [27], Denmark [28], Germany [29], and the USA [30]. However, to the best of the authors’ knowledge, it is very hard to find a study focusing on efforts to evaluate the opportunity of V2G application in developing countries, including Indonesia. Dhar et al. [31] assessed the long-term benefits of EV adoption in the case of India. However, their study mainly focused on the environmental impacts, rather than the application of V2G. In addition, Li et al. [32] investigated the energy, environmental, and economic implications of EVs in China. They found that the controlled charging and discharging of EVs are advantageous in terms of environmental impacts. However, these impacts will be insignificant if not followed by any decarbonization in the energy sector. Although their study is very interesting in terms of the environmental and economic points of view, they did not cover the technological applications of ancillary services using EVs. This study focuses on the techno-economic analysis of V2G application in the Indonesian power grid as a representative of developing countries. The characteristics of the power grid in Indonesia are significantly different from those of many developed countries. Both load leveling and frequency regulation are evaluated in terms of applicability, environmental impacts (CO2 reduction), and economy. The actual data (load profile and frequency) used and analyzed is based on an Indonesian state-owned electric company (PLN). The conditions of the Indonesian electricity system, including the electricity growth, energy mix, frequency fluctuation, and V2G system, are also initially explained in Section 2. The methodology used in this study to analyze the techno-economic performance of V2G in the Indonesian power grid is described in Section 3. Next, Section 4 explains several conditions and parameters adopted in this study, including technical parameters, driving patterns, and electricity market. The results of this study are explained and discussed in Section 5. Lastly, Section 6 summarizes the findings from this study. In general, the main contributions of this work can be summarized as follows: • • • Analysis of the Indonesian power grid, especially the condition of frequency fluctuation, which clarifies the demand for balancing service with fast response. Evaluation of V2G opportunity in supporting or balancing the Indonesian power grid, especially the opportunity of load-leveling (including peak shaving) utilizing EVs. Environmental and economic assessments of load-leveling adopting EVs in the Indonesian power grid. This also includes how far the fuel consumption for balancing service can be reduced due to the employment of V2G for load-leveling. 2. Condition of Power System in Indonesia Electricity demand is strongly influenced by economic activities, population, and lifestyle (including seasonal conditions). Therefore, it can be considered as a sign of national economic growth [3]. In Indonesia, PLN is the main decision-maker for the entire development of electricity infrastructure, as it has a monopoly on electricity transmission and distribution, as well as most of the power generation. PLN has a plan to achieve security of supply by increasing the electrification ratio and setting a reserve margin of 25–30% in the Java-Madura-Bali (JAMALI) area, and of 35–40% in the rest of the country. In addition, a commitment to higher safety and lower production costs are also projected [33]. To meet the future demand and encourage investment from the private sector, PLN is offering future revenue through power purchase agreements (PPAs), in which PLN guarantees a long-term power demand to independent power providers (IPPs) [33]. The forecast of demand depends on the economic situation and other aspects that affect consumption behavior [16]. Indonesia has a Energies 2020, 13, 1162 4 of 16 Energies 2020, 13, x FOR PEER REVIEW 4 of 17 significantly growing investment in power generation, with a target that by 2027 approximately 63% of the the generation generation capacity capacity is is supplied supplied by by IPPs, IPPs,with withan anexpectation expectationof of6–7% 6–7%of ofeconomic economicgrowth growth[33]. [33]. of Unfortunately, after thethe economic situation diddid notnot meet expectations and Unfortunately, afterseveral severalyears yearsofofthis thisprogram, program, economic situation meet expectations PLN reduced the target. Still, it is obligated to pay for on-going contracts under the PPA’s take-or-pay and PLN reduced the target. Still, it is obligated to pay for on-going contracts under the PPA’s takeagreement. In this In agreement, the capacity payment is usedisasused a basis which IPPthe willIPP receive or-pay agreement. this agreement, the capacity payment as aonbasis on the which will payment as longas aslong the power capacity is available, even if the ifdemand is low dispatched [13]. receive payment as the power capacity is available, even the demand is or low or dispatched In contrast, the peak consumption is increasing gradually year by year, leading to significant load gaps [13]. In contrast, the peak consumption is increasing gradually year by year, leading to significant between and off-peak times, as shown Figure 1 [34–36]. This analysisThis shows the potential load gapspeak between peak and off-peak times,inas shown in Figure 1 [34–36]. analysis shows risk the in overpaying for unutilized power generation, and an increase in the peak follower investment cost to potential risk in overpaying for unutilized power generation, and an increase in the peak follower meet the demand to achieve the targeted energy security and reserve margin [13]. investment cost to meet the demand to achieve the targeted energy security and reserve margin [13]. Figure Figure 1. 1. Comparison Comparison of of power power consumption, consumption, peak peak load load and and economic economic growth. growth. As Indonesia Indonesia is is aa very very large country country having having approximately approximately 17,000 17,000 islands, islands, the the power power grid grid is is As divided into several local local power grids. Among Among them, them, the the JAMALI JAMALI grid grid is is the the biggest biggest power power grid grid with with divided approximately 70% of the total national capacity, owing to its larger population and higher economic approximately of the total national capacity, its larger population and higher economic activities as areas. In In thisthis study, the JAMALI power grid isgrid analyzed in the context activities as compared comparedwith withother other areas. study, the JAMALI power is analyzed in the of the possibility and feasibility to utilize for EVs loadfor leveling and secondary regulation control. context of the possibility and feasibility to EVs utilize load leveling and secondary regulation Currently, the power demand this power grid is mainly covered geothermal and coal-fired control. Currently, the power in demand in this power grid is mainlybycovered by geothermal andsteam coalpowersteam plantspower [36]. However, theHowever, load fluctuation must be covered by abe high ramp-rate generator fired plants [36]. the load fluctuation must covered by a power high ramp-rate to return the frequency to its value (50.00 Hz) value [34]. In addition, the peak load basically occurs power generator to return thenormal frequency to its normal (50.00 Hz) [34]. In addition, the peak load twice in a day, caused by high demand from several sectors, including industrial (noon peak) and basically occurs twice in a day, caused by high demand from several sectors, including industrial residential (night activities. As the primary regulation, governor-free regulationgovernor-free is adopted to (noon peak) and peak) residential (night peak) activities. As the primary regulation, stabilize the frequency, with a deadband range (βf ) of +36 mHz. However, there is a high possibility of regulation is adopted to stabilize the frequency, with a deadband range (Δπ) of +36 mHz. However, disturbance ofpossibility the power of system (higher of and than +36 mHz) owing to technical or non-technical there is a high disturbance thelower power system (higher and lower than +36 mHz) owing problems [10]. to technical or non-technical problems [10]. Figure 22 shows shows the the total number number of frequency fluctuations occurring occurring in corresponding corresponding times times Figure during weekdays weekdays for for two two months months (December (December 2017 2017 and and January January 2018). 2018). ItIt is is higher higher than than the the nominal nominal during frequency deadband (+36 than 1 min in frequency (+36 mHz), and and the the corresponding correspondingfluctuations fluctuationsoccurred occurredfor forlonger longer than 1 min the JAMALI power grid. In general, the number of frequency fluctuations is higher in the morning in the JAMALI power grid. In general, the number of frequency fluctuations is higher morning and evening, evening, owing owingto tohigh highdemand demandand andchanges changesininthe theresidential residential sector. This fluctuation leads and sector. This fluctuation leads to to several technical and economic problems, including low energy efficiency, high operating cost, several technical and economic problems, including low energy efficiency, high operating cost, and and high environmental impacts byfossil usingfuels fossil fuels suchand as coal. gas and AsIndonesia of today, high environmental impacts causedcaused by using such as gas As ofcoal. today, Indonesia has no large-scale powersuch storage, such ashydroelectricity, pumped hydroelectricity, which able tothe balance has no large-scale power storage, as pumped which is able to isbalance grid the grid responsively. Therefore, grid balancing depends heavily on the marginal capacity and peaking responsively. Therefore, grid balancing depends heavily on the marginal capacity and peaking power power generators. generators. The concept developed in this study is to utilize the batteries of EVs to both deliver and absorb electricity to and from the grid, respectively. EVs can deliver their electricity during the peak time (high demand), or when the grid frequency is lower than the nominal value. In addition, they also can absorb electricity from the grid when there is any surplus power in the grid (such as during the 130 130 110 110 90 90 Count Count generation to cover this fast-growing demand. As the current condition, although the intermittent power generation (solar and wind) still has a very limited contribution to the Indonesian power system, the Indonesian grid has suffered a power imbalance leading to very low power quality. These phenomena are caused by several problems, including insufficient power supply, poor supply-side protection system, significant distribution system disturbance by natural disaster, and slow responseEnergies 2020, 13, 1162 5 of 16 ability to adjust the demand. 70 70 50 50 30 30 0 3 6 9 12 Time (a) 15 18 21 23 0 3 6 9 12 15 18 21 23 Time (b) Figure Figure 2.2. Total Totalnumber numberofoffrequency frequencyfluctuations fluctuationsoccurred occurredininthe thecorresponding correspondingtime timeduring duringthe the weekdays monthsin inJava-Madura-Bali Java-Madura-Bali(JAMALI) (JAMALI) power grid (frequency fluctuation higher weekdays for for two months power grid (frequency fluctuation higher than than the nominal frequency deadband (+36 and mHz) and than longer than (a) 1 min). (a) βfHz < 0.036 (b) βfHz. > the nominal frequency deadband (+36 mHz) longer 1 min). βf < 0.036 (b) βfHz > 0.036 0.036 Hz. The concept developed in this study is to utilize the batteries of EVs to both deliver and absorb electricity from thethe grid, respectively. EVs can duringincluding the peak time Figureto2 and shows that frequency fluctuation is deliver caused their due electricity to some factors, the (high demand), or when the grid frequencyactivity, is lowerand than theofnominal value. In the addition, they The also changes in the usage behavior, uncontrolled use electricity during peak time. can absorb electricity from the change grid when there is any surplus power grid (such as during the purple bar indicates the largest of frequency (βf) compared to in thethe normal condition (50 Hz). night), or when the frequency is lower than its nominal value. Therefore, distributed EVs can be The increase of demand causes the frequency (f) to drop less than 50 Hz; therefore, the system needs considered as supply a bundled energy storage when theyIfare managed [12].the Hence, in case to increase its to normalize the frequency. theappropriately demand drops suddenly, system can that the is very the plants grid operator potentially dispatch the stand-by generators to reduce itsdemand supply from thelow, power throughcan several mechanisms, including a governor control stop spinning. In addition, one can allow the wind and solar power systems to deliver their optimum (primary regulation). power generation capacity to the grid. Energy storage can the manage the problems active power generated from V2G is considered as an innovative idea for solving current in the Indonesian intermittent renewable energy sources, such as wind andfor solar, to prevent any grid disturbance [37]. power grid. Figure 3 shows a conceptual diagram of V2G, providing ancillary services to the grid. Byaddition, pluggingseveral the EVsdrawbacks into bidirectional points,be it is possible toand introduce more unpredictable In of EVs charging can potentially minimized even evaded through the renewable energy into the electricity supply advent of improved technologies in the futuremix, [39].and therefore, reduce the carbon emission. The EV drivers can set both the maximum and minimum values of the battery’s state-of-charge (SOC) during these ancillary services. The ancillary service market can be conducted through any automatic auction technology, and an aggregator of EVs can calculate and optimize the possible revenue for the EVs taking part in the ancillary services. In the current condition of the Indonesian grid system, the adoption of renewable energy is still very limited. In the future, especially up to 2040, although the share of renewable energy is predicted to increase, the main contribution to power generation is still dominated by coal and natural gas [38]. This is due to the fast-growing economy, which is demanding a significant capacity of power generation to cover this fast-growing demand. As the current condition, although the intermittent power generation (solar and wind) still has a very limited contribution to the Indonesian power system, the Indonesian grid has suffered a power imbalance leading to very low power quality. These phenomena are caused by several problems, including insufficient power supply, poor supply-side protection system, significant distribution system disturbance by natural disaster, and slow response-ability to adjust the demand. Figure 2 shows that the frequency fluctuation is caused due to some factors, including the changes in the usage behavior, uncontrolled activity, and use of electricity during the peak time. The purple bar indicates the largest change of frequency (βf ) compared to the normal condition (50 Hz). The increase of demand causes the frequency (f ) to drop less than 50 Hz; therefore, the system needs to increase its supply to normalize the frequency. If the demand drops suddenly, the system can reduce its supply from the power plants through several mechanisms, including a governor control (primary regulation). V2G is considered as an innovative idea for solving the current problems in the Indonesian power grid. Figure 3 shows a conceptual diagram of V2G, for providing ancillary services to the grid. In addition, several drawbacks of EVs can potentially be minimized and even evaded through the advent of improved technologies in the future [39]. Energies 2020, 13, 1162 6 of 16 Energies 2020, 13, x FOR PEER REVIEW 6 of 17 Figure 3. Vehicle-to-grid (V2G) with electric vehicles (EVs), bi-directional charger, and aggregator. Figure 3. Vehicle-to-grid (V2G) with electric vehicles (EVs), bi-directional charger, and aggregator. 3. Methodology 3. Methodology To assess the technical and economic benefits of V2G, especially load-leveling (peak shaving), several have beenand additionally of the limitations of Indonesia’s policies To parameters assess the technical economic assumed, benefits ofbecause V2G, especially load-leveling (peak shaving), regarding EVs and their ancillary services. In general, frequency regulation is divided into three several parameters have been additionally assumed, because of the limitations of Indonesia’s policies categories based on response time, frequency deviation, duration, and capacity: primary, secondary, regarding EVs and their ancillary services. In general, frequency regulation is divided into three and tertiary reserves [40]. Primary regulationdeviation, is defined as the control that provides ansecondary, automatic categories based on response time, frequency duration, and capacity: primary, response from the governor within a few seconds, owing to sudden imbalances. It is performed to return and tertiary reserves [40]. Primary regulation is defined as the control that provides an automatic the grid frequency to its within nominal i.e., 50 Hz for the Indonesian power grid, to stabilize response from the back governor a value, few seconds, owing to sudden imbalances. It isand performed to quickly. In the JAMALI grid, primary frequency regulation is currently conducted governor-free return the grid frequency back to its nominal value, i.e., 50 Hz for the Indonesian power grid, and to within a deadband +36 mHz and a maximum speed drop of 5%. It is is currently automatically activated for any stabilize quickly. Inof the JAMALI grid, primary frequency regulation conducted governorfrequency within ±36mHz mHzand to the nominal frequency [14]. When primary regulation is free withindeviation a deadband of +36 a maximum speed drop of 5%. It isthis automatically activated not able to return the frequency back to the nominal value or the frequency deviation is larger than for any frequency deviation within ±36 mHz to the nominal frequency [14]. When this primary ±36 mHz, is the secondary frequency regulation is to activated. Secondary is always being regulation not able to return the frequency back the nominal value orregulation the frequency deviation is operated the peak period when high consumption is expected, and V2Gregulation may participate in larger thanduring ±36 mHz, the secondary frequency regulation is activated. Secondary is always providing a more responsive service This study focuses on potential of EVs being operated during the peak period[41]. when high consumption is the expected, andand V2Gfeasibility may participate utilization fora more secondary frequency regulation, grid on frequency profile data for every 10 s. in providing responsive service [41]. Thisusing studythe focuses the potential and feasibility of EVs The proposed forfrequency the techno-economic conducted in this profile study isdata shown in Figure 4. utilization for method secondary regulation, analysis using the grid frequency for every 10 s. Basically, there are four stages in this methodology, including the assumption of several technical The proposed method for the techno-economic analysis conducted in this study is shown in Figure parameters driving analysis, technicalthe analysis related grid balancing, 4. Basically, and therea are four pattern, stages inload this profile methodology, including assumption ofto several technical and economic analysis [42]. parameters and a driving pattern, load profile analysis, technical analysis related to grid balancing, and economic analysis [42]. Energies 2020, 13, 1162 Energies 2020, 13, x FOR PEER REVIEW 7 of 16 7 of 17 start Defined from forecast of EVs growth and survey of driving behavior based on private vehicles usage in Jakarta Technical parameter Driving pattern Technical Analysis Peak period Fluctuation Period Load shaving Spinning reserves Actual data from power grid that analyzed by calculation methods Fossil fuel substitution GHG emission reduction Economical assessment Economical Analysis Converted by carbon emission factor Optimized demand response analysis finish Figure Figure 4. 4. Conceptual Conceptual diagram diagram of of techno-economic techno-economic assessment assessment for for V2G V2G implementation. implementation. 4. Input Parameters 4. Input Parameters 4.1. Technical Parameters 4.1. Technical Parameters Based on data from Statistics Indonesia (BPS), the number of passenger cars in Java and Bali in Based on data from Statistics Indonesia (BPS), the number of passenger cars in Java and Bali in 2016 was approximately 9 million [14]. To measure the potential of EV utilization for load-leveling 2016 was approximately 9 million [14]. To measure the potential of EV utilization for load-leveling and economic feasibility, it is assumed that the number of EVs will reach 1 million in 2040. It is and economic feasibility, it is assumed that the number of EVs will reach 1 million in 2040. It is important to note that it is challenging to have a precise assumption related to the number of EVs in important to note that it is challenging to have a precise assumption related to the number of EVs in developing countries, such as Indonesia, as currently, the adoption of EVs in these countries is almost developing countries, such as Indonesia, as currently, the adoption of EVs in these countries is almost zero. In addition, Table 1 lists several assumptions used in the study for the analysis. The battery zero. In addition, Table 1 lists several assumptions used in the study for the analysis. The battery capacity is assumed to be the average capacity of EVs available in the market, such as owned by capacity is assumed to be the average capacity of EVs available in the market, such as owned by Nissan Leaf. Moreover, the V2G participation percentage is set to 30%considering an early and Nissan Leaf. Moreover, the V2G participation percentage is set to 30%considering an early and sufficiently-developed adoption period. sufficiently-developed adoption period. Table 1. Technical parameter for electric vehicles (EVs) and Charger [43,44]. Table 1. Technical parameter for electric vehicles (EVs) and Charger [43,44]. No. No. 1 1 2 2 3 3 4 4 5 5 6 7 6 8 7 8 Parameter Parameter Battery capacity Battery capacity Charger capacity Charger capacity Charging time Charging Charging time rate Discharging raterate for V2G Charging Depth of discharge Discharging rate for V2G Real energy consumption Depth of discharge Average distance for travel to work Real energy consumption Average distance for travel to work Value Value 24 kWh 24 kWh 3 kW (level 2) 3 kW (level 2) 7 h 15 min h 15 min 2.47kWh/h 2.85 kWh/h 2.4 kWh/h 80%kWh/h 2.85 9.26 kWh per 100 km 80% 20 km/trip 9.26 kWh per 100 km 20 km/trip Energies 2020, 13, x FOR PEER REVIEW Energies 2020, 13, 1162 8 of 17 8 of 16 4.2. Driving Patterns 4.2. Driving Patterns To compare the technical benefits and revenue that can be earned by EVs in a metropolitan area, five cases were assumed as driving behavior, based on the results of a survey conducted in the Jakarta To compare the technical benefits and revenue that can be earned by EVs in a metropolitan area, area.cases These cases were as projected on based a survey conducted 2018 with 219 correspondents five were assumed driving based behavior, on the results ofin a survey conducted in the Jakarta located in the Jakarta Figure 5 shows the assumed driving behaviors setcorrespondents in this study. In Case area. These cases werearea. projected based on a survey conducted in 2018 with 219 located 1, EVs are owned as commuting vehicles during the weekdays are only connected to in the Jakarta area. privately Figure 5 shows the assumed driving behaviors set in thisbut study. In Case 1, EVs are the V2Gprivately charger as during the night at home, as there is no V2Gbut connection the office/company. owned commuting vehicles during the weekdays are only at connected to the V2G During the weekend, the at EVs are not available of vacation or unplanned events. Case 2the is charger during the night home, as there is nobecause V2G connection at the office/company. During developedthe based Case with an additional being connected weekend. weekend, EVsonare not1,available because of availability vacation orfor unplanned events. during Case 2 the is developed In Caseon 3,Case an additional V2G connection is assumed noonconnected at the office/company. Moreover, Case based 1, with an additional availability for at being during the weekend. In in Case 3, 4, EVs are owned by the office or company as operating vehicles during weekdays and, therefore, an additional V2G connection is assumed at noon at the office/company. Moreover, in Case 4, EVs are they arebyusually moving around before returning toweekdays the office/company in the evening and owned the office or company as noon, operating vehicles during and, therefore, they are usually connecting to the V2G charger until the morning of the next day. During the weekend, they are fullymoving around noon, before returning to the office/company in the evening and connecting to the connected foruntil 24 h.the Finally, in Case 5, EVs notDuring used for commuting used during the V2G charger morning of the nextare day. the weekend, but theyare areonly fully-connected for weekend, which in metropolitan areas withbut established commuting 24 h. Finally, in usually Case 5, happens EVs are not used for commuting are only used duringtransportation the weekend, systems. which usually happens in metropolitan areas with established commuting transportation systems. Figure 5. Driving pattern of EVs based on private vehicle usage and availability for V2G connection. Figure 5. Driving pattern of EVs based on private vehicle usage and availability for V2G connection. All of these driving patterns affect the SoC of the battery, owing to both charging and discharging All of these driving patterns affect the SoC of the battery, owing to both charging and behaviors. In all cases, the depth-of-discharge (DoD) is set to 80%. Charging or discharging during V2G discharging behaviors. In all cases, the depth-of-discharge (DoD) is set to 80%. Charging or will be stopped when the battery is fully charged or its SoC reaches the DoD, respectively. In driving, discharging during V2G will be stopped when the battery is fully charged or its SoC reaches the DoD, SoC is strongly affected by the average distance (d) and energy consumption (Econs ). In addition, respectively. In driving, SoC is strongly affected by the average distance (π) and energy consumption the capacity of the battery (Bcap ), charging rate (Crate ), and discharging rate (Drate ) also affects the SoC (πΈππππ ). In addition, the capacity of the battery (π΅πππ ), charging rate (πΆπππ‘π ), and discharging rate (π·πππ‘π ) while t represents the status on time (h). Their correlations can be shown in the following equations: also affects the SoC while π‘ represents the status on time (h). Their correlations can be shown in the following equations: (Econs ∗ d) Trip discharging SoCt (trip) = SoCt−1 − (1) Bcap (πΈππππ ∗π) (1) Trip discharging πππΆπ‘ (π‘πππ) = πππΆπ‘−1 − Crate Charging and V2G charging SoCt = SoCt−1 + Bcap Charging and V2G charging πππΆπ‘ = πππΆπ‘−1 + Drate V2G discharging SoCt (V2G) = SoCt−1 − Bcap V2G discharging πππΆπ‘ (π2πΊ) = πππΆπ‘−1 − π΅πππ (2) πΆπππ‘π π΅πππ π·πππ‘π π΅πππ (2) (3) (3) Energies2020, 2020,13, 13,x1162 Energies FOR PEER REVIEW of17 16 99of 4.3. Electricity Market 4.3. Electricity Market The liberalization of the energy market, including electricity, has motivated all of the related Theincluding liberalization of suppliers, the energyconsumers, market, including electricity, has motivated all of the related entities, power and transmission operators, to monitor the grid entities, including power suppliers, consumers, and transmission operators, to monitor the grid condition actively and participate in its market, such as in Europe. For example, in several Nordic condition everyone actively and in its market, as in Europe. For example, in several Nordic countries, canparticipate generate electricity withsuch minimum requirements and can participate in countries, everyone can generate electricity with minimum requirements and can participate in ancillary ancillary support to the power grid in certain periods with a high rate of tariff [45]. In addition, the support power grid certain periods with a high rate tariff quality, [45]. In addition, the controlling TSO who is TSO whotoisthe responsible forin managing the grid frequency andof power and also for responsible managing gridlevel frequency and power quality, also forin controlling the ancillaryfor services, has the a high of concern regarding the and challenges balancingthe theancillary power services, has a high level of concern regarding the challenges in balancing the power system [46]. system [46]. Incontrast, contrast, PLN PLN still still implements implements aa fixed-rate In fixed-rate electricity electricity tariff tarifffor forhousehold householdconsumers. consumers.Figure Figure6 profile of of the theJAMALI JAMALIpower powergrid, grid, and share of fossil and natural 6shows showsthe the load load profile and thethe share of fossil fuel,fuel, coal,coal, and natural gasgas-based power generation. The TSO only provides incentives for both industrial and business based power generation. The TSO only provides incentives for both industrial and business consumersfor foroff-peak off-peak periods, periods, to to attract attract electricity electricity consumption consumption during during this this period. period. As Asititcan can be be consumers observed in Figure 6, there is an unusual supply fluctuation from fossil fuel-based power plants, observed in Figure 6, there is an unusual supply fluctuation from fossil fuel-based power plants, especiallycoal, coal,where wherethere thereisisaalarge largegap gapbetween betweenpeak peakand andoff-peak off-peakhours, hours,and andbetween betweenweekdays weekdays especially andthe theweekend. weekend.This Thisproblem problemisiscaused causedininpart partby bythe thelack lackof of peak peak follower followerpower powergenerators generatorsto to and provide a high-speed ramp to solve the sudden fluctuation in the peak period and oversupply from provide a high-speed ramp to solve the sudden fluctuation in the peak period and oversupply from coal-fired power power plants. plants. InInaddition, addition,ininthis thisstudy, study, analysis analysisof of the the frequency frequency fluctuation fluctuation occurring occurring coal-fired duringthe theday dayisisalso alsoconducted conductedto tostudy studythe thepotential potentialof ofthe theancillary ancillaryservices servicesmarket marketfor forfrequency frequency during regulation. For these purposes, four periods are assumed in a day, as shown in Table 2. regulation. For these purposes, four periods are assumed in a day, as shown in Table 2. 2.5 10 4 Energy (MWh) 2 1.5 Geothermal Coal Combine Cycle Hydro Gas Turbine Gas Engine 1 0.5 0 00:00 03:00 06:00 09:00 12:00 15:00 18:00 21:00 24:00 Time (a) (b) (c) (d) Figure Load profile profileofof JAMALI power (a) general load profile a representative Figure 6. 6. Load JAMALI power grid:grid: (a) general load profile during during a representative weekday, weekday, (b) demand covered by fossil fuels, (c) demand covered by coal, and (d) demand covered (b) demand covered by fossil fuels, (c) demand covered by coal, and (d) demand covered by natural gas. by natural gas. Table 2. Period of the electricity market for frequency regulation. Table 2. Period of the electricity market for frequency regulation. No. Period Time No. Period 1 Block 1 4 AM–6Time AM 1 Block 1 4 AM–6 AM 2 Block 2 10 AM–12 PM 2 Block 2 Block 3 10 AM–12 3 2 PM–4 PM PM 3 Block 3 Block 4 2 PM–4 4 6 PM–8 PM PM 4 Block 4 6 PM–8 PM Energies 2020, 13, x FOR PEER REVIEW 10 of 17 EnergiesThe 2020,electricity 13, 1162 market for load-leveling and frequency regulation is divided into three different 10 of 16 categories: fixed tariff, natural tariff, and demand response tariff (DRT). The fixed tariff provides flat charging and discharging tariffs to EV owners in accordance with the current PLN regular tariffs electricity market for load-leveling andthe frequency regulation is divided three different (IDRThe 1467). The natural tariff is obtained from fluctuation of the cost of the into composition of the categories: fixed tariff, natural tariff, and demand response tariff (DRT). The fixed tariff flatis energy mix, with an additional margin that is close to the average PLN regular tariffs.provides This tariff charging and discharging tariffs EVtotal owners in accordance withpower the current tariffs calculated from the fuel cost (ππ )toand supply (ππ ) from each plant PLN of theregular different fuel (IDR 1467). The natural tariff is obtained from the fluctuation of the cost of the composition of the resources, which depends on the energy mix for every 30 min. The total generation cost (Cππ’ππ ) from energy mix, with an additional margin that the average PLN tariffs.fuel Thisresources tariff is each fuel can be calculated by summing upisallclose the to power generators ofregular the different calculated from the supply (Si ) from each powerGE: plant the different fuel i ) and total (G: geothermal; CP: fuel coal;cost CC:(P combine cycle; H: Hydro; GT: gas turbine; gasofengine). resources, which depends on the energy mix for every 30 min. The total generation cost (C f uel ) from π π π π π π each fuel can be calculated by summing up all the power generators of the different fuel resources ππ‘ππ‘ = ∑ ππΊπ + ∑ ππΆππ + ∑ ππΆπΆπ + ∑ ππ»π + ∑ ππΊππ + ∑ ππΊπΈπ (G: geothermal; CP: coal; CC: combine cycle; H: Hydro; GT: gas turbine; GE: gas engine). π=1 Stot = π=1 n X SGi + i=1 π=1 π=1 π=1 π=1 n X n n n n X X X X ∑ππ=1 ππ ππ SCPi + + S + S + SGEi πΆππ’ππ S= CCi Hi GTi πi=π‘ππ‘ i=1 i=1 1 i=1 i=1 Pn πππ‘π’πππ πππππππ (π) = πΆπΊ + πΆπΆπ +i Si πΆπΆπΆ + πΆπ» + πΆπΊπ + πΆπΊπΈ P C f uel = i=1 In the above, π and π are the number of unitsStot for each corresponding power generation type Natural Tari f f s (N ) = CG + CCP + CCC + CH + CGT + CGE and fuel. Demand response tariffs (DRT) offer incentives according to the amount of electricity, in terms In the above, i and n are the number of units for each corresponding power generation type and fuel. of both supply and demand. This tariff will increase when the demand is high, and vice versa. To Demand response tariffs (DRT) offer incentives according to the amount of electricity, in terms of analyze DRT, data has been obtained and recorded for two months (December 2017 and January both supply and demand. This tariff will increase when the demand is high, and vice versa. To analyze 2018), resulting in a regression line that determines both supply and demand lines. To create this line, DRT, data has been obtained and recorded for two months (December 2017 and January 2018), resulting the minimum point is a result of deduction from the total minimum supply and V2G power supply in a regression line that determines both supply and demand lines. To create this line, the minimum for every 30 min in a month. The maximum point is taken from the highest demand in a month point is a result of deduction from the total minimum supply and V2G power supply for every 30 min supplied by a dispatched power plant, depending on the merit order (indexed by lowest pricing and in a month. The maximum point is taken from the highest demand in a month supplied by a dispatched highest efficiency). power plant, depending on the merit order (indexed by lowest pricing and highest efficiency). Figure 7 shows a comparison of regular tariffs, natural tariffs, and DRTs. There is a large gap Figure 7 shows a comparison of regular tariffs, natural tariffs, and DRTs. There is a large gap between weekends and weekdays for the generation cost because of lower supply from the coal-fired between weekends and weekdays for the generation cost because of lower supply from the coal-fired power plant, which has the most significant percentage in the energy mix. The natural tariff has one power plant, which has the most significant percentage in the energy mix. The natural tariff has one fluctuation during the representative weekday, while DRT, shown as a blue dashed line, is relatively fluctuation during the representative weekday, while DRT, shown as a blue dashed line, is relatively high during most of the day, except for a part of the night. On the other hand, the fluctuation of high during most of the day, except for a part of the night. On the other hand, the fluctuation of natural tariff during the weekend, shown as a red line, is higher than the weekday. It means that the natural tariff during the weekend, shown as a red line, is higher than the weekday. It means that composition of energy mix during the weekend is less efficient because most of the energy mix the composition of energy mix during the weekend is less efficient because most of the energy mix generated from coal is shut down to reduce the supply. Moreover, DRT during the weekend is generated from coal is shut down to reduce the supply. Moreover, DRT during the weekend is relatively relatively low due to low demand during the weekend. These tariff fluctuations are adopted to low due to low demand during the weekend. These tariff fluctuations are adopted to analyze the analyze the charging and discharging costs of EVs for one month of the analysis period. charging and discharging costs of EVs for one month of the analysis period. 1475 Tariff (IDR) 1470 1465 Natural (Weekday) Natural (Weekend) Regular DRT (Weekday) DRT (Weekend) 1460 1455 00:00 03:00 06:00 09:00 12:00 15:00 18:00 21:00 24:00 Time Figure Figure7.7. Comparison Comparisonof ofthe theregular, regular,natural, natural,and anddemand demandresponse responsetariffs. tariffs. Energies 2020, 13, 1162 11 of 16 The total charging cost (CC) is influenced by several parameters, including charging capacity Ccap , charger efficiency Ce f f , and electricity tariffs for each corresponding fixed regular tariff, natural tariff, and DRT. CC = Ccap × Ce f f × tari f f 5. Results and Discussion The utilization of EVs has significant advantages, owing to its significantly faster response and higher accuracy in the stabilization of the power system as compared to conventional power generation, such as gas turbines and steam turbines (coal-fired) [47]. However, among the above three frequency regulations, secondary reserves occur more frequently than others, although their capacity is relatively small. The results are discussed and analyzed from two different points of view: technical and economic, as mentioned in Section 3. 5.1. Technical Analysis 5.1.1. V2G for Load Leveling As energy storage, EVs can cover the high ramp rate in the peak period via V2G and can be dispatched to supply electricity to the grid when the demand is high, both during noon and evening. In addition, they can also absorb the electricity when the demand is low, such as from midnight to morning. Table 3 shows the possible contribution to the peak reduction via V2G in all evaluated cases. As can be observed, there is a clear potential reduction of energy demand during the peak period (MWh) in Indonesia via V2G, especially in the JAMALI power system. Case 5, where the EVs are available for both charging (electricity absorption) and discharging (electricity delivery) during weekdays, it can provide the largest peak load reduction as compared to the other cases (2.8% for coal and 8.8% for gas). Table 3. Potential fossil fuel reduction on the peak period. Cases Energy Supply from EVs Monthly (GWh) 1 2 3 4 5 Total Peak Energy Reduction (%) by Fuel Replacement Day Night Coal Gas 0 0 68.4 34.2 68.4 18.8 18.8 18.1 35.5 34.2 0.5 0.5 2.4 1.9 2.8 1.6 1.6 7.5 6.0 8.8 5.1.2. Emission Reduction V2G implementation in the power system also has the potential to reduce the consumption of fossil fuels in specific periods [48]. Therefore, it significantly impacts the decrease in greenhouse gas (GHG) emissions. The reduction of GHG can be calculated by using an emission factor, generally explained in an energy outlook (coal: 92.8 t-CO2 /TJ, and gas: 54.3 t-CO2 /TJ). Figure 8 shows a comparison of the potential amount of CO2 reduction for each case, with coal and gas replacement by V2G during the peak load hours. Case 5 shows the highest potential for both coal and gas because of its significantly longer availability, especially during weekdays. Furthermore, Case 3 follows in second for both coal and gas replacement. The comparison of fossil fuel consumption between peak and off-peak hours was not proportional, depending on the demand response issued by the TSO. In contrast, the grid supply during the off-peak period (night time) is relatively stable and proportional and is able to meet a predictable demand, as it is mostly not fluctuating. Energies 2020, 13, Energies2020, 2020,13, 13,x1162 xFOR FORPEER PEERREVIEW REVIEW Energies 12 of 16 12of of17 17 12 Figure 8.CO2 CO2 potentialreduction reduction fromV2G V2G substitutingboth both coal-and and gas-fueledpower power generators Figure Figure 8. 8. CO2 potential potential reduction from from V2G substituting substituting both coalcoal- and gas-fueled gas-fueled power generators generators during peak hours. during during peak peak hours. hours. 5.2. Economic 5.2. Economic Analysis Analysis 5.2. The calculation calculationofof of cost and potential revenue is conducted conducted using the tariff assumptions assumptions The cost andand potential revenue is conducted using the tariffthe assumptions explained The calculation cost potential revenue is using tariff explained in Section 3. These tariffs also include some incentives from the TSO or the aggregator to in Section 3. These tariffs also include some incentives from the TSO or the aggregator to increase explained in Section 3. These tariffs also include some incentives from the TSO or the aggregator to increase participation in V2G and to avoid additional demand from EVs charging during the peak participation in V2G and to avoid additional demand from EVs charging during the peak period. As increase participation in V2G and to avoid additional demand from EVs charging during the peaka period. As result of of the the calculations, calculations, the charging charging cost and and potential revenue of V2G V2Gfor areeach different result ofAs theaacalculations, the charging the cost and potential revenue of V2G are different case, period. result cost potential revenue of are different for each case, depending on the driving pattern and usage tariffs from Section 4.3. Each tariff has depending on the driving pattern and usage tariffs from Section 4.3. Each tariff has been compared, for each case, depending on the driving pattern and usage tariffs from Section 4.3. Each tariff has been compared, including cost and revenue for every respected case. The EVs that participated in including cost and revenuecost for every respected EVs thatcase. participated V2G have similar been compared, including and revenue forcase. everyThe respected The EVs in that participated in V2G have have similar similar parameters and numbers.isThis This comparation is considered aseffort the first first evaluation parameters and numbers. Thisand comparation considered as theis first evaluation for evaluation V2G study V2G parameters numbers. comparation considered as the effort for V2G study in Indonesia. in Indonesia. effort for V2G study in Indonesia. Figure99shows shows the comparison ofof both cost and revenue during theimplementation implementation ofV2G. V2G. The Figure showsthe thecomparison comparison both cost and revenue during the implementation of V2G. Figure of both cost and revenue during the of The charging costcost decreases owing to aatolow low tariff when charging isisconducted conducted during the night, because because The charging decreases owing a low tariff when charging conductedduring duringthe thenight, because charging cost decreases owing to tariff when charging is of the lower electricity price. study also concluded reduce charging cost when of the lower electricity price. This concluded that V2G could reduce the charging cost when of the lower electricity price. This study also concluded that V2G could reduce the charging cost when EVsare are parked parked and and connected connected to to charging chargingstations. stations. EVs are parked and connected to charging EVs stations. Figure 9. Cost and revenue during V2G participation. Figure9. 9.Cost Costand andrevenue revenueduring duringV2G V2Gparticipation. participation. Figure The economic analysis shows that Cases 3 and 5 have a high potential benefit from the cost The economic economic analysis analysis shows shows that that Cases 33 and and 55 have have aa high high potential benefit benefit from the the cost cost The reduction, as the EVs are mostly parked Cases during the weekdays. From potential the comparison offrom the different reduction, as the EVs are mostly parked during the weekdays. From the comparison of the different reduction, the EVs areismostly parked the weekdays. Fromtothe comparison of thecost different tariffs, the as natural tariff considered to during have significant potential reduce the charging in all tariffs, the the natural natural tariff tariff is is considered considered to to have have significant significant potential potential to to reduce reduce the the charging charging cost cost in in all all tariffs, cases, as as well well as as to to maximize maximize revenue revenue from from V2G V2G participation. participation. All All of of these these cases cases are are optional optional for for the the cases, Energies 2020, 13, 1162 13 of 16 cases, as well as to maximize revenue from V2G participation. All of these cases are optional for the EV owners, depending on their conditions for utilizing EVs. Table 4 shows the potential reduction of charging costs for all cases with different tariffs. However, the percentage of cost reduction is similar among all the tariffs, the natural tariff results in the highest reduction. Table 4. Potential reduction of charging cost. Cases Tariffs Regular Natural DRT Case 1 Case 2 Case 3 Case 4 Case 5 22.74% 22.77% 22.73% 30.51% 30.53% 30.46% 47.12% 47.17% 47.08% 46.24% 46.21% 46.14% 59.87% 59.93% 59.83% The results from the economic analysis clearly show that the number of participants and the electricity market has significant impacts on the economic performance of EVs. They also potentially reduce the total cost of ownership (TCO), which is strongly related to the concerns of EV owners in the future. Table 5 shows the potential revenue improvement for the power company (operator) for different tariff options. Case 5, with a natural tariff and gas replacement, leads to the largest annual improvement of approximately 3.65% for the electric company as compared to other cases. It is caused by the longest V2G participation time as compared to other cases, as the cost of gas fuel, which was reduced, is more expensive than the cost of V2G. Massive EV adoption and participation in the V2G program can be considered as a potential way to achieve a high quality of electricity during weekdays with a stable regulatory framework. Coal replacement is not profitable for an electric company, because the generation cost is still cheaper for producing electricity. Table 5. Net profit improvement for the power company (operator) in different tariff options. Fuel Replacement Tariffs Cases Case 1 Case 2 Case 3 Case 4 Case 5 Gas Regular Natural DRT 2.10% 1.57% 2.10% 2.19% 0.63% 2.20% 2.16% 1.99% 2.16% 2.45% 2.40% 2.46% 1.31% 3.65% 1.31% Coal Regular Natural DRT 0.87% 0.35% 0.87% 0.71% −0.83% 0.71% −0.01% −0.18% −0.01% 0.16% 0.10% 0.17% −0.84% 1.45% −0.84% However, there are several challenges and barriers that must be clarified. The number of EVs in the future is predicted to increase gradually, and the potential of EVs for ancillary services is also increasing accordingly. The assumed general barriers in V2G are: (1) an EV aggregator should be available on a large scale, and should have advanced technology to cover the requested functions during V2G; (2) the lifetime of an EV battery potentially degrades because of continuous EV operation; (3) the ancillary services market using EVs has not yet been well-established; (4) the knowledge and motivation to participate in the V2G program by early adopters remain low, (5) the policies, incentives, and tax systems must be improved and applied to facilitate and increase participation, (6) further developments in technologies and supporting infrastructures, including chargers and batteries, are urgently required; and (7) several environmental impacts, such as battery recycling caused by high battery degradation, must be clarified and developed. The economic value of V2G can be improved if there are optimization and incentives on charging and discharging following V2G participation in the ancillary services. As a future study, an optimization with a genetic algorithm method is considered to be capable of providing a clear analysis. The utilization of a non-traditional optimization technique (genetic algorithms) which suits a large-scale problem to optimally decide and perform charging and discharging to achieve high power quality. It is expected that the proposed framework can help the EV owners and power Energies 2020, 13, 1162 14 of 16 company to make appropriate decisions that ensure a sustainable operation, achieve a minimum battery degradation, and optimize the revenue of the standby EVs, with the least cost and optimum operational condition. In the real application, the value and impacts of V2G depend strongly on the charging and driving behaviors of EV drivers and owners. The policy and regulation, as well as incentives, set by the government in order to drive EV drivers and owners to participate in V2G services or fulfill their commitment in providing the services, are strongly expected. In addition, an objective and accurate measurement for both charging and discharging are urgently required, increasing the trust of all involved players, including owners, drivers, operators, and utilities. A clear and single tax system in V2G is also demanded, therefore, the economic performance of V2G may be improved. 6. Conclusions A revolution in the transportation sector will take place in the next decade. This change impacts the overall energy cycle directly and indirectly, including the fossil fuels, which are currently mainly used for internal combustion engine vehicles. The use of EVs is not only limited to the primary use in transportation but also includes the electricity sector. This study focuses on the potential use of EVs in the Indonesian power grid, especially for load leveling and frequency regulation. Initially, both charging and discharging behaviors of EVs owners were analyzed. The gap between the highest and lowest loads is very high, leading to inefficient generator operation, and large frequency fluctuation. Massive EV adoption and participation in V2G programs can be considered as a potential method for achieving a high quality of electricity in the country. The results show that the peak supply from fossil fuels is effectively reduced by up to 2.8% (for coal) and 8.8% (for gas) in Case 5, i.e., EVs that stay home the entire day except on the weekends, as representative of commuter behavior in a metropolitan city. The driving pattern of Case 4 is one option, as commuting workers with EVs reduce the fossil generation supply by approximately 2.4% (for coal) and 7.5% (for gas). From the economic assessment, there are conclusions from two perspectives, showing that both the power company and EV owner can receive a benefit from V2G implementation. For the EV owners, the charging cost of EVs can be reduced by up to 60.15%, in Case 4 with a natural tariff. Meanwhile, the power company gets the benefit of increasing the power consumption, owing to EVs charging in the night (off-peak period), and of reducing the cost of fossil fuels, which only peaks in certain periods. The calculation shows that when the V2G is used to replace the supply from fossil fuel partly, the power company (PLN) can improve its revenue by at least 3.65% from the reduction in gas consumption in Case 5 with a natural tariff. The recurrent and large fluctuation in frequency urgently demands a very responsive regulator (service provider), such as EVs or a battery. Several challenges and barriers related to the use of EVs for grid ancillary services need to be analyzed and solved urgently, as the number of EVs is increasing sharply. Author Contributions: Conceptualization, M.H.; T.K.; M.A.; Methodology, M.H. and M.A.; Data collection, M.H.; Draft preparation, M.H.; Draft revision and check, M.A. and T.K. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Conflicts of Interest: The authors declare no conflict of interest in the submission of this paper. References 1. Tokimatsu, K.; Höök, M.; McLellan, B.; Wachtmeister, H.; Murakami, S.; Yasuoka, R.; Nishio, M. Energy modeling approach to the global energy-mineral nexus: Exploring metal requirements and the well-below 2 β¦ C target with 100 percent renewable energy. Appl. Energy 2018, 225, 1158–1175. [CrossRef] Energies 2020, 13, 1162 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 15 of 16 Mitani, T.; Aziz, M.; Oda, T.; Uetsuji, A.; Watanabe, Y.; Kashiwagi, T. Annual assessment of large-scale introduction of renewable energy: Modeling of unit commitment schedule for thermal power generators and pumped storages. Energies 2017, 10, 738. [CrossRef] IEA. Energy and Climate Change; World Energy Outlook Special Report; IEA: Paris, France, 2015; pp. 1–200. [CrossRef] Aziz, M.; Oda, T.; Kashiwagi, T. Extended utilization of electric vehicles and their re-used batteries to support the building energy management system. Energy Procedia 2015, 75, 1938–1943. [CrossRef] Oda, T.; Aziz, M.; Mitani, T.; Watanabe, Y.; Kashiwagi, T. Actual congestion and effect of charger addition in the quick charger station: Case study based on the records of expressway. Electr. Eng. Jpn. (Engl. Transl. Denki Gakkai Ronbunshi) 2017, 198, 11–18. [CrossRef] Hodge, B.-M.S.; Huang, S.; Shukla, A.; Pekny, J.F.; Reklaitis, G.V. The effects of vehicle-to-grid systems on wind power integration in California. Comput. Aided Chem. Eng. 2010, 28, 1039–1044. [CrossRef] Aziz, M.; Oda, T.; Mitani, T.; Watanabe, Y.; Kashiwagi, T. Utilization of electric vehicles and their used batteries for peak-load shifting. Energies 2015, 8, 3720–3738. [CrossRef] Teixeira, A.C.R.; Sodré, J.R. Impacts of replacement of engine powered vehicles by electric vehicles on energy consumption and CO2 emissions. Transp. Res. Part D Transp. Environ. 2018, 59, 375–384. [CrossRef] Juangsa, F.B.; Budiman, B.A.; Aziz, M.; Soelaiman, T.A.F. Design of an airborne vertical axis wind turbine for low electrical power demands. Int. J. Energy Environ. Eng. 2017, 8, 1–9. [CrossRef] Huda, M.; Aziz, M.; Tokimatsu, K. Potential ancillary services of electric vehicles (vehicle-to-grid) in Indonesia. Energy Procedia 2018, 152, 1218–1223. [CrossRef] Prananto, L.A.; Juangsa, F.B.; Iqbal, R.M.; Aziz, M.; Soelaiman, T.A.F. Dry steam cycle application for excess steam utilization: Kamojang geothermal power plant case study. Renew. Energy 2018, 117, 157–165. [CrossRef] Jian, L.; Zheng, Y.; Shao, Z. High efficient valley-filling strategy for centralized coordinated charging of large-scale electric vehicles. Appl. Energy 2017, 186, 46–55. [CrossRef] Chung, Y. Overpaid and Underutilized: How Capacity Payment to Coal-Fired Power PLants Could Lock Indonesia into a High-Cost Electricity Future; Institute for Energy Economics and Financial Analysis: Singapore, 2017. Aziz, M.; Huda, M.; Nandiyanto, A.; Indonesia, U.P.; Abdullah, A.G.; Indonesia, U.P. Opportunity of frequency regulation using electric vehicles in Denmark. J. Eng. Sci. Technol. 2018, 13, 1700–1712. May, G.J.; Davidson, A.; Monahov, B. Lead batteries for utility energy storage: A review. J. Energy Storage 2018, 15, 145–157. [CrossRef] Saxena, S.; Le, C.; Macdonald, J.; Moura, S. Quantifying EV battery end-of-life through analysis of travel needs with vehicle powertrain models. J. Power Sources 2015, 282, 265–276. [CrossRef] Gaines, L. The future of automotive lithium-ion battery recycling: Charting a sustainable course. Sustain. Mater. Technol. 2014, 1–2, 2–7. [CrossRef] Bishop, J.D.K.; Axon, C.J.; Bonilla, D.; Tran, M.; Banister, D.; Mcculloch, M.D. Evaluating the impact of V2G services on the degradation of batteries in PHEV and EV. Appl. Energy 2013, 111, 206–218. [CrossRef] Uddin, K.; Jackson, T.; Widanage, W.D.; Chouchelamane, G.; Jennings, P.A.; Marco, J. On the possibility of extending the lifetime of lithium-ion batteries through optimal V2G facilitated by an integrated vehicle and smart-grid system. Energy 2017, 133, 710–722. [CrossRef] Dubarry, M.; Devie, A.; McKenzie, K. Durability and reliability of electric vehicle batteries under electric utility grid operations: Bidirectional charging impact analysis. J. Power Sources 2017, 358, 39–49. [CrossRef] Al-hadhrami, L.M.; Alam, M. Pumped hydro energy storage system: A technological review. Renew. Sustain. Energy Rev. 2015, 44, 586–598. [CrossRef] Ma, T.; Yang, H.; Lu, L. Feasibility study and economic analysis of pumped hydro storage and battery storage for a renewable energy powered island. Energy Convers. Manag. 2014, 79, 387–397. [CrossRef] Bolund, Ã.; Bernhoff, H.; Leijon, M. Flywheel energy and power storage systems. Renew. Sustain. Energy Rev. 2007, 11, 235–258. [CrossRef] Lund, H.; Salgi, G. The role of compressed air energy storage (CAES) in future sustainable energy systems. Energy Convers. Manag. 2009, 50, 1172–1179. [CrossRef] Alipour, M.; Mohammadi-Ivatloo, B.; Moradi-Dalvand, M.; Zare, K. Stochastic scheduling of aggregators of plug-in electric vehicles for participation in energy and ancillary service markets. Energy 2017, 118, 1168–1179. [CrossRef] Energies 2020, 13, 1162 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 16 of 16 Nunes, P.; Brito, M.C. Displacing natural gas with electric vehicles for grid stabilization. Energy 2017, 141, 87–96. [CrossRef] Sarabi, S.; Davigny, A.; Courtecuisse, V.; Riffonneau, Y.; Robyns, B. Potential of vehicle-to-grid ancillary services considering the uncertainties in plug-in electric vehicle availability and service/localization limitations in distribution grids. Appl. Energy 2016, 171, 523–540. [CrossRef] Noel, L.; Zarazua de Rubens, G.; Sovacool, B.K. Optimizing innovation, carbon and health in transport: Assessing socially optimal electric mobility and vehicle-to-grid pathways in Denmark. Energy 2018, 153, 628–637. [CrossRef] Taljegard, M.; Göransson, L.; Odenberger, M.; Johnsson, F. Impacts of electric vehicles on the electricity generation portfolio—A Scandinavian-German case study. Appl. Energy 2019, 235, 1637–1650. [CrossRef] Dallinger, D.; Gerda, S.; Wietschel, M. Integration of intermittent renewable power supply using grid-connected vehicles—A 2030 case study for California and Germany. Appl. Energy 2013, 104, 666–682. [CrossRef] Dhar, S.; Pathak, M.; Shukla, P.R. Electric vehicles and India’s low carbon passenger transport: A long-term co-benefits assessment. J. Clean Prod. 2017, 146, 139–148. [CrossRef] Li, Y.; Davis, C.; Lukszo, Z.; Weijnen, M. Electric vehicle charging in China’s power system: Energy, economic and environmental trade-offs and policy implications. Appl. Energy 2016, 173, 535–554. [CrossRef] PT PLN (Persero). Electricity Supply Business Plan (RUPTL) PLN 2017-2026; PT PLN (Persero): Jakarta, Indonesia, 2017. PT PLN (Persero). PLN Statsistics; PT PLN (Persero): Jakarta, Indonesia, 2017. PT PLN (Persero). Daily Load Curve Java Bali Grid; PT PLN (Persero): Jakarta, Indonesia, 2017. PT PLN (Persero). Energy Balance in Jawa Bali; PT PLN (Persero): Jakarta, Indonesia, 2017. (in Indonesian) Mathur, H.D.; Bhateshvar, Y.K. Frequency regulation with vehicle-to-grid (V2G) option in multi-generation power network. Energetika 2016, 62, 68–77. [CrossRef] PT PLN (Persero). Electricity Supply Business Plan (2018–2019); PT PLN (Persero): Jakarta, Indonesia, 2019. Habib, S.; Kamran, M.; Rashid, U. Impact analysis of vehicle-to-grid technology and charging strategies of electric vehicles on distribution networks—A review. J. Power Sources 2015, 277, 205–214. [CrossRef] Ozer, B.; Arikan, O.; Moral, G.; Altintas, A. Electrical Power and Energy Systems Extraction of primary and secondary frequency control from active power generation data of power plants. Int. J. Electr. Power Energy Syst. 2015, 73, 16–22. [CrossRef] Kempton, W.; Tomi, J. Vehicle-to-grid power fundamentals: Calculating capacity and net revenue. J. Power Sources 2005, 144, 268–279. [CrossRef] Weldon, P.; Morrissey, P.; Brady, J.; Mahony, M.O. An investigation into usage patterns of electric vehicles in Ireland. Transp. Res. Part D Transp. Environ. 2016, 43, 207–225. [CrossRef] Electric Vehicle Database. Available online: ev-database.org (accessed on 10 March 2019). Nissan Leaf Charging Guide. Available online: https://www.zap-map.com/charge-points/nissan-leafcharging-guide/ (accessed on 10 March 2019). Peng, C.; Zou, J.; Lian, L. Dispatching strategies of electric vehicles participating in frequency regulation on power grid: A review. Renew. Sustain. Energy Rev. 2017, 68, 147–152. [CrossRef] Papadogiannis, K.A.; Hatziargyriou, N.D.; Member, S. Optimal Allocation of Primary Reserve Services in Energy Markets. IEEE Trans. Power Syst. 2004, 19, 652–659. [CrossRef] Luo, X.; Xia, S.; Chan, K.W. A decentralized charging control strategy for plug-in electric vehicles to mitigate wind farm intermittency and enhance frequency regulation. J. Power Sources 2014, 248, 604–614. [CrossRef] Hao, H.; Cheng, X.; Liu, Z.; Zhao, F. Electric vehicles for greenhouse gas reduction in China: A cost- e ff ectiveness analysis. Transp. Res. Part D Transp. Environ. 2017, 56, 68–84. [CrossRef] © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).