Scholarly article on topic 'Analysis method and utilization mechanism of the overall value of EV charging'

Analysis method and utilization mechanism of the overall value of EV charging Academic research paper on "Economics and business"

CC BY-NC-SA
0
0
Share paper
Academic journal
Energy Conversion and Management
OECD Field of science
Keywords
{"Electric vehicles" / "Charging infrastructure" / "Load shifting" / "Emission reduction" / "Overall value"}

Abstract of research paper on Economics and business, author of scientific article — Chunlin Guo, Ching Chuen Chan

Abstract Electric Vehicle (EV) can save energy while reducing emissions and has thus attracted the attention of both academics and industry. The cost and benefit of charging are one of the key issues in relation to EV development that has been researched extensively. But many studies are carried out from a viewpoint of some local entities rather than a global system, focus on specific types or aspects of EV charging, or use mixed models that can only be computed by computer simulation and lack physical transparency. This paper illuminated that it is necessary to consider the value of EV charging on a system scale. In order to achieve this, it presents an analytical model for analyzing the overall value of EVs, an analysis model to evaluate the reduction of pollutions relevant to photovoltaic power, and a model to transfer the intrinsic savings of wind power to the off-peak charging loads. It is estimated that EV charging has a significant positive value, providing the basis for enhanced EV subsidies. Accordingly, a utilization mechanism apt to optimize globally is proposed, upon which sustainable business models can be formed by providing adequate support, including the implementation of a peak–valley tariff, charging subsidies and one-time battery subsidies. This utilization mechanism, by taking full advantage of the operation system of power utilities to provide basic support and service, may provide new approaches to the development of EVs. The method proposed here is of important value for the systematic considerations about EV development and maybe can help broaden the possibility of EV development.

Academic research paper on topic "Analysis method and utilization mechanism of the overall value of EV charging"

Contents lists available at ScienceDirect

Energy Conversion and Management

journal homepage: www.elsevier.com/locate/enconman

Energy

Conversion IManagement

Analysis method and utilization mechanism of the overall value ■. cnwMai

of EV charging ^

Chunlin Guoa *, Ching Chuen Chan

a State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources, North China Electric Power University, Beijing 102206, China b Department ofEEE, University of Hong Kong, Pokfulam Road, Hong Kong

ARTICLE INFO

ABSTRACT

Article history: Received 28 May 2014 Accepted 6 October 2014 Available online 24 October 2014

Keywords: Electric vehicles Charging infrastructure Load shifting Emission reduction Overall value

Electric Vehicle (EV) can save energy while reducing emissions and has thus attracted the attention of both academics and industry. The cost and benefit of charging are one of the key issues in relation to EV development that has been researched extensively. But many studies are carried out from a viewpoint of some local entities rather than a global system, focus on specific types or aspects of EV charging, or use mixed models that can only be computed by computer simulation and lack physical transparency. This paper illuminated that it is necessary to consider the value of EV charging on a system scale. In order to achieve this, it presents an analytical model for analyzing the overall value of EVs, an analysis model to evaluate the reduction of pollutions relevant to photovoltaic power, and a model to transfer the intrinsic savings of wind power to the off-peak charging loads. It is estimated that EV charging has a significant positive value, providing the basis for enhanced EV subsidies. Accordingly, a utilization mechanism apt to optimize globally is proposed, upon which sustainable business models can be formed by providing adequate support, including the implementation of a peak-valley tariff, charging subsidies and one-time battery subsidies. This utilization mechanism, by taking full advantage of the operation system of power utilities to provide basic support and service, may provide new approaches to the development of EVs. The method proposed here is of important value for the systematic considerations about EV development and maybe can help broaden the possibility of EV development.

© 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-SA license

(http://creativecommons.org/licenses/by-nc-sa/3XI/).

1. Introduction

The escalating issues of oil resource depletion and environmental pollution have drawn worldwide attention to the Electric Vehicle (EV). The EV and charging infrastructure have been extensively studied, and a significant number of demonstrations have been built in various countries [1]. However, in spite of the huge investment and resource, the development of EV remains very slow. The primary reasons include the fact that, the development of the power battery was slower than expected, such that EVs are not competitive with fuel vehicles. Meanwhile, the operational mode of the charging infrastructure remains uncertain, and its construction is slow; both of these have seriously stymied EV development. Additionally, different aspects of the related industry function with near autonomy, lacking coordination and optimization, with the

q This work was supported in part by National High Technology R&D Program of China (863 Program) (2011AA05A109).

* Corresponding author. E-mail addresses: gcl@ncepu.edu.cn (C. Guo), ccchan@eee.hku.hk (C.C. Chan).

consequence that they are unable to form a resultant, coherent force.

The cost and benefit of charging are the key issues for the development of EV. And they must be considered taking EV, charging infrastructure, and power grid as a whole; and this is the base to determine reasonable charging price scheme and subsidy policy, and maximum the whole social welfare. Though many elaborated researches in relation to the impact, cost and benefits of EV charging have been carried out, there are still some drawbacks as following:

Firstly, it is only from a viewpoint of some local entities rather than a global system that many literatures studied the economy and benefits of EV charging. Jonathan and David [2] using Western Australia, the smallest wholesale electricity market in the world, as a test case, discussed the economic and commercial viability of vehicle-to-grid (V2G). It calculated the benefits of V2G for arbitrage in short term energy market, providing spinning reserve, load following and participating demand side management. And the cost of battery wear, communication systems and alternative options of V2G were analyzed in the other hand. The report concludes that most variants of V2G are currently too costly to

http://dx.doi.org/10.1016/j.enconman.2014.10.016 0196-8904/© 2014 The Authors. Published by Elsevier Ltd.

This is an open access article under the CC BY-NC-SA license (http://creativecommons.org/licenses/by-nc-sa/3.0/).

implement in the light of alternative options. However, EV charging can be added to demand side management without substantial investment. Based on the Well-to-Wheel (WTW) methodology, Ricardo et al. [3] presents a study of the economic and environmental balances for EVs versus fuel vehicle. It takes into account different primary energy supply includes fossil fuels, nuclear and Renewable Energy Source (RES), and different vehicle technologies include Battery Electric Vehicle (BEV), Hybrid Electric Vehicle (HEV) and Plug-in Hybrid Electric Vehicle (PHEV). BEVs have less than a half of the emissions than an fuel vehicle. However, the ownership costs during its life cycle (about 10 year) are similar to an equivalent fuel vehicle, despite the lower operational costs for BEVs. A comprehensive approach with two large-scale distribution areas and three penetrations is proposed to evaluate the impact of PEV on distribution network and shows that, the smart charging can avoid up to 60-70% of the required incremental investment [4]. The comparison in [5] concludes that battery and hybrid EV are offering clearly short-term and medium-term solutions with a large number of common components, and maybe fuel cell vehicles can contribute to long-term solutions. A stochastic model based on the driving behavior of Western European drivers is studied to determine the electrical power required for PHEVs in Belgium and shows that, coordinated charging is essential to minimize the impact, while the double tariffs scheme will create a large, unwanted peak [6]. A value model is presented to assessing the integration of EVs into Australia's electricity industry, which is based on the relationship of supply and demand [7]. In these studies, only the cost and benefits of EV user or power grid are analyzed and optimized, and being under special electricity price scheme or in certain power market; this is distinct from the analysis and optimization in a systematic manners. From the viewpoint of system, it is the electricity price scheme that needs to be optimized at first; otherwise, with an unreasonable electricity price scheme, it is unlikely to maximum system benefits, and will reduce the significance of partial optimization.

Several articles present studies on the cost and benefits just about specific types or aspects of EV charging, while without involving general methods and results about the value analysis of EV charging. The economics of fast charging is studied in a case of Germany [8], where potential users and organization structures are investigated as well as different tariff types. The key drivers of revenue are the tariff, the capacity and the utilization rates. Cost components include initial capital expenditure and operational expenditures. The results show that, a market-driven roll-out of Level III fast charging infrastructure is unlikely to be profitable in Germany at 2011 EV penetration rates. And the general EV adoption rate is detected as being a main risk factor for investment in public charging infrastructure. The economic costs and benefits of ''smart charging'' policy is explored using a number of variables for creating alternative scenarios: level of PEV penetration, 120 V versus 240 V charging, and whether there is a carbon price or not [9]. There are significant absolute dollar savings associated with load shifting, although they represent a fairly low percentage of total system costs. The value of smart charging policy varies significantly across electric grids. A simple 12-8 am time-of-use tariff coupled with a circuit timer appears to be the most economical way to maximize the net benefits. But the economic benefits of optimal charging cannot justify investing in the smart grid infrastructure. A new broader method is proposed to compare different vehicle technologies, and giving insight on electric grid impact and CO2 emissions [10]. The method was applied to series and parallel PHEVs with different driving cycles, driving distances and user behaviors. The result shows that, total driven kilometers greatly affects total CO2 emissions and user cost. A stochastic optimization algorithm is presented and justified to maximize the use of renewable energy, where the Monte Carlo simulation of transportation

patterns and Hong's estimation method to mimic renewable energy sources are used [11]. A decentralized valley-filling charging strategy is presented, in which a day-ahead pricing scheme is optimized and broadcasted to EV owners [12]. An intelligent charging method in response to time of use (TOU) price is presented, and the results have validated its effectiveness [13].

Some papers use mixed models with multiple objectives and constraints that can only be computed by computer simulation. Although this method can generate more accurate numerical results, the conclusions tend to be vague in physical transparency, and common analytical models and general rules are difficult to obtain, because a variety of interactions and influence factors are intertwined. Liu et al. [14] estimates the costs and benefits of PHEV by elaborated models. Typical daily driving patterns are derived by the data from Finland, taking into account differences between weekdays and holidays. The model treats the vehicles connected as a storage pool, which participates in the day-ahead and intraday markets, considering reserves and a minimum level for the leaving battery. The impact of different options are compared on the power system of Finland, including using mixed integer programming model or linear programming model, providing spinning/non-spinning reserves or not, and using smart charging or dumb charging. It turns out to be that the system cost of EV charge was around 36 €/ vehicle/year by smart charging while around 263 €/vehicle/year by dumb EVs. An assessment of impact of PHEV charging patterns on power system is presented by using of detailed stochastic models, where the problem is formulated by objective function, start-up and shut-down costs constraints, power balance and reserve constraints, coupling constraints of individual unit status, start-up and shut-down indicators, individual unit balance and reserve constraints, individual unit ramping constraints and minimum on/off time, wind power constraints, PHEV load charging balance, PHEV load hourly charging limit, reserve provided by V2G [14]. Rotering and Ilic [15] uses two dynamic programming to find the economically optimal solution for the vehicle owner, and the analysis of the California independent system operator indicates that, smart charge timing reduces driving costs from $0.43 to $0.2. Provision of regulating power substantially improves the daily profits to $1.71. A stochastic optimization algorithm with detail model is presented to coordinate charging of EVs to maximize the use of renewable energy [16].

In addition, for utilization of EV charging value, despite different types of business models are discussed, the overall value of EV charging gets little attentions. A holistic approach to developing business models for EV is presented based on morphological methods, which can capture the complex interrelations, show potential design options, and reveals technical and organizational limitations [1]. Andersen et al. [17] introduces the business model for integration of EVs into private transport systems established by the firm better place, where an intelligent rechargeable network is provided by an Electric Recharge Grid Operator (ERGO).

As mentioned above, because of the comprehensiveness, complexity and uncertainty of EV development, the universal and simple models and methods for the value analysis of EV charging remain a challenge to be resolved. Further, this will lead to various difficult in many aspects, such as policy analysis and decisions, business model selection, and develop route choosing. Therefore, this paper aims to develop an analytical model and method for determination of the overall value of EV charging, so as to provide a useful tool for analysis of charging price, subsidy policy, and other system issues. Section 2 proposes the analysis model of the overall value of EV charging which include marginal cost of electricity for charging, the value of pollution reduction, and the value corresponding to the increase in new energy utilization, presents the evaluation method of each part value, and estimates the change range of the overall value. In Section 3, a utilization

mechanism of the overall value of EVs is proposed. Section 4 discusses the results of this paper. And the conclusions, the advantages and defects of the method and possible future research direction are given in Section 5. These studies present new ideas and methods for systematically analyzing EV, which can facilitate the further research and implement of EVs.

2. Analysis model and method for the overall value of EV charging

2.1. Basic model for the overall value of EV charging

From the system perspective, EV can participate in the peak-valley regulation of power systems, increasing the utilization of new energy and reducing environmental pollution. Therefore, when replacing fuel-powered vehicles with EVs, a value model of charging that considers the overall benefits can be calculated with the following Eq. (1):

Vev = Cc + Dev - D,v + Vn

where VEV is the overall value of EV charging, CC is the marginal cost of electricity for charging, DEV is the value corresponding to the charging-generated pollution, DiV is the value corresponding to the pollution generated by the fuel vehicle that were replaced, and VNE is the value corresponding to the increase in new energy utilization. Here, gain is defined as a positive value, and cost is defined as a negative value.

Because both new energy and electricity loads have peak-valley characteristics obviously, the cost and value of on-peak and off-peak charging instances must be calculated separately. And the method to quantify these values is analyzed in the following section.

2.2. The marginal electricity cost for EV charging

Because electricity cannot be stored in significant quantities, the power system must build power facilities that can meet the peak load requirements. And increasing the load during off-peak hours does not require the installation of additional power facilities; instead, only the corresponding marginal cost of electricity must be considered.

Based on the operating characteristics of China's electric power system, the electricity demand of off-peak EV charging is met mainly by increasing the output of thermal power plants. Thus, the only additional cost is the marginal cost of the coal consumed, while other operating costs essentially do not increase. According to data of 2012, the standard coal consumption of power supply of thermal power is 326 g/kW h, which is equivalent to an average coal cost of approximately 0.23 Yuan/kW h [18]. Because the marginal cost of coal is generally lower than the average cost of coal, the marginal cost of off-peak EV charging is less than 0.23 Yuan/ kWh.

With the development of new energy power generation, the phenomenon of abandoning wind power during off-peak hours has grown to a considerable quantity in China, and this problem will worsen when the installed capacity of wind power increases. Studies have shown that shifting EV charging to off-peak hours can increase the utilization of new energy and reduce the abandonment of wind power. This method does not alter the cost of fuels, nor will it increase the operating costs; therefore, the marginal cost of electricity for this part of EV charging during off-peak hours can approach zero.

Therefore, the marginal electricity cost for the load of off-peak EV charging is in the range of 0 to 0.23 Yuan/kW h. The cost of electricity during peak hours is the same as that of general loads and is

calculated using the commercial and industrial electricity price of approximately 1.2 Yuan/kW h.

2.3. The value of pollution reduction

EVs can reduce pollution and environmental damage, the value of which is expressed as (DEV - DIV) in Eq. (1). The remaining problem is how the associated environmental benefits can be quantified in terms of economic value.

At present, photovoltaic power is commonly subsidized in China and other countries because the generation of photovoltaic power is believed to reduce the associated damage from pollution and, thus, is considered to be worth the economic cost. Therefore, the equation used to calculate the value of emission reduction of EV charging by comparison with photovoltaic power is given as follows:

S -n n - Hev - H,v S

Sev — DEV — D,V — 77-¡j— Spv

HPV — HTE

where SPV and SEV are the subsidies for photovoltaic power and EV charging, respectively, and the values should be proportional to their reduction in pollution damage. According to the Chinese National Development and Reform Commission, the current subsidy for distributed photovoltaic power generation is 0.42 Yuan/kW h, while the remaining electricity feeding into grid must be purchased at the benchmark tariff of coal-burning units [19]. The latter reflects the cost of the coal-burning electricity that is replaced, and the former is just the subsidy for reducing the damage caused by pollution. HEV is the pollution damage per unit of electricity used by EVs, HIV is the pollution damage of fuel vehicles that is replaced by EVs at one unit of equivalent electricity, HPV is the pollution damage caused per unit of photovoltaic power, and HTE is the pollution damage caused per unit thermal power.

The pollution associated with photovoltaic power is generally believed to be very small and can be ignored compared with that from thermal power. Meanwhile, most of the electricity used in EV charging is thermal power; thus, the pollution damage of EV is less than, but approximately equal to, the pollution damage caused by the thermal power consumed, i.e., HEV HTE. Then, the following equation can be derived by using the above relationship in Eq. (2):

Sev — Dev — D,v >~ (H,v/Hev — 1)Spv — (k — 1)Spv

where k is the ratio of pollution damage between fuel vehicles and EVs with same drive distance. Eq. (3) shows that the subsidies for EVs should be (k - 1) times greater as the subsidies for photovoltaic power.

According to the standard of China, the smoke emission per unit of thermal power in 2012 was 0.39 g/kWh, with the emission of sulfur oxides (SOx) reduced to 2.26 g/kW h. Thus, these values can be regarded as the approximate pollution emission per unit of electricity consumed by EV charging [18]. Ricardo et al. [3] presents the pollution emission level of different scenarios for the primary energy supply in Portugal. The emission of SOx is 9.3 g/kW h while NOx is 3 g/kW h. However, it did not calculate the cost corresponding to pollution or governance cost. And compared with China, the pollution emission per unit generation in Portugal is not necessarily larger due to low ratio of thermal power and big penetration of renewable energy.

According to the stage IV emission standards for light-duty vehicles [20], the emission limits for gasoline passenger cars are 1.00 g/km for carbon monoxide (CO), 0.10 g/km for hydrocarbons (HC), and 0.08 g/km for nitrogen oxides (NOx); additionally, the limit of particulate matter emission for diesel vehicles is 0.025 g/ km. The emission limits per unit of distance traveled are converted

to emission limits per unit of equivalent electricity used by EV with the following equation:

h'v = h'v/le

where H[V and HjV are the emission limit per unit of the equivalent electricity used by EV and the emission limit per unit of mileage for fuel vehicles. LE is the mean power consumption per kilometer, assuming the common used value of 0.20 kW h/km [21].

Using the above equation, the emission limits per unit of equivalent electricity used by EVs are calculated to be 5.00 g/kW h (CO), 0.50 g/kW h (HC) and 0.40 g/kW h (NOx), and the particulate matter emission limit for diesel vehicles is 0.125 g/kWh.

The above comparison shows that there are more pollutant species in the emissions of fuel vehicles, where the total emission limit is 6.025 g/kW h. So the mean total emission is set to 5.00 g/kW h in our study. As for thermal power, the NOx emission is set to be half of the SOx emission, resulting in a total emission of 3.78 g/kW h. Then, the ratio (k1) between the former and the latter is the ratio of total pollution, and is approximately 1.32.

Additionally, vehicle emissions are concentrated with small areas in urban regions, which also tend to involve poor circulation. Thus, secondary pollutants, such as "photochemical" smog, are prone to generation through chemical reactions between the aggregated pollutants. Therefore, the same amount of emissions can cause more pollutants in the cities than in remote areas, and the degree of pollution should be multiplied by an aggregation coefficient of k2, which is supposed to be in the range from 1.5 to 2.0.

Further, the number and sensitivity of pollution victims in urban areas are significantly greater than those in remote areas, where the power plants are located. Taking the population as an example, according to the sixth national census of China (2010 data), the population density was 8562 persons/km2 in the urban area of Beijing and 1195 persons/km2 in the entire Beijing metropolitan region, while the population density of China's eastern region was 506 persons/km2. Beijing's mean population density was 2.4-16.9 times that of east China. I t is evident that the damage caused by pollution of the same degree in urban areas is significantly greater than that in remote areas. Therefore, the pollution damage caused by vehicle emissions should be multiplied by a sensitivity coefficient, k3, whose value is set to be in the range from 2.0 to 3.0.

The equation used to calculate the overall multiplier (k) of pollution damage can be derived by integrating the ratio of total pollution (k1), aggregation coefficient (k2) and sensitivity coefficient (k3) as follows:

k = k1k2k3

Therefore, the range of k (fuel vehicle-to-EV pollution damage ratio) is 3.96-7.92. The subsidy for EVs, as calculated by using k in Eq. (2), is in the range from 1.24 to 2.91 Yuan/kWh, reflecting the value of pollution reduction as a result of EVs.

t should be noted that the mechanism of environmental pollution and damage is very complex, influenced by many factors. However, no generally recognized calculation method is available at present. Therefore, a simplified model for analysis, together with a conservative aggregation coefficient and sensitivity coefficient, is used in our study for estimation. The serious situation of air pollution in the metropolises all over the world shows, their actual values may be greater; therefore, the values of k and the value of pollution reduction as a result of EVs may be even greater.

2.4. The value of improving new energy utilization

Among the new energy sources, wind power, which also exhibits properties of an unfavorable peak-valley feature, rapid develop-

ment and large capacity, has come to significantly impact the power grid. However, EV can help consume the power generated by wind. The value of EVs in this aspect is analyzed in the following section.

The unit peak-valley difference rate of the power curve shown in Fig. 1 is defined as follows, to help analyze the role that EVs play in consuming wind power:

: = (Ppeak - Pvalley)/ (24 ^ Pdt^j

I n which Ppeak and Pvalley are the powers at peak and off-peak hours, respectively.

Clearly, power supply with a positive unit peak-valley difference rate and loads with a negative one can reduce the peak-valley difference of the entire system. Greater absolute values of these suggest that they will contribute more significantly. The unit peak-valley difference rate of wind power is usually negative, which will increase the peak-valley difference of the entire system. If the EV charging load is adjusted so that its peak-valley difference rate is also negative, then the peak-valley difference caused by the wind power can be offset, helping the grid to consume more wind power. The total energy consumption can be approximated using the following equation:

E aEV E

EWP =-Eev

where EWP is the integrated wind power, Eev is the charging energy of EVs, and aWP and aEV are their peak-valley difference rates, respectively. The wind power integration rate increased by EVs, b = £wp/£ev = aEV/aWP, can be derived from the above equation, which can be used to approximate the amount of wind power that could be integrated per unit of EV charging electricity.

Neither wind power nor photovoltaic power technology consumes physical resources or produces environmental pollution. Thus, the total intrinsic value of each technology is the same, which includes the value of the electricity and the environmental value corresponding to the emission reduction. When the cost of wind power is high, the maximum subsidy can be the difference between the intrinsic value and the feed-in tariff. As the cost drops, it will require less support by subsidization. The savings can be used for off-peak EVs charging loads to promote wind power integration and, thus, to avoid the abandonment of it while improving the overall efficiency of the system. Therefore, the subsidy for the loads of off-peak EV charging can be calculated using the following equation:

(5) VNE = ß(VWP - Pwp)/c

where VNE is the value of the increase in new energy utilization in Eq. (1), VWP is the intrinsic value of the power from new energy resources such as wind power, approximated with the current

Time (h)

Fig. 1. Power supply or demand curve. This curve is the supply or demand power (kW) versus the time (h).

benchmark tariff of photovoltaic power plants (i.e., 1.0 Yuan/kWh in China). PWP is the benchmark tariff of wind power, which currently ranges from 0.51 to 0.61 Yuan/kWh. y is the proportion of the off-peak EV charging energy in the entire charging load; its value is less than 1 and generally increases with an increasing peak-valley price difference. In this study, it is set to 0.4-0.7.

Clearly, b (the wind power integration rate) and y (the proportion of off-peak charging energy) are related to the peak-valley feature of EV loads, which is affected by the scheme of EV load scheduling and the response level of users. All of them are uncertain factors and, therefore, are difficult to evaluate accurately. However, appropriate tariff incentives may increase the consumption of wind power largely because the charging loads can be significantly adjusted with a suitable scheme.

For a conservative calculation, the wind power integration rate b increased by EVs is set to 0.5-1.5, the wind power feed-in tariff is 0.61 Yuan/kW h, and the proportion of off-peak charging electricity y is set to 0.7 in our study. Then, the electricity price for off-peak charging is calculated to range from 0.28 to 0.84 Yuan/kW h with Eq. (8), due to improving the wind power integration. On-peak charging cannot increase the utilization of wind power; hence, the corresponding value is zero, i.e., the consumption should not be subsidized.

2.5. Overall value of EV charging

The integrated overall value of EV charging derived from the marginal electricity cost, pollution reduction value and the value of improving new energy utilization is calculated with Eq. (1) and shown in Table 1.

The above table clearly shows that EV charging has a positive overall value and, therefore, should be subsidized and rewarded. If the life-time mileage of an EV is 200,000 km and the electricity consumed is 0.2 kW h/km, then the total amount of subsidies is 51,600 to 150,000 Yuan if only off-peak charging is used, and is 1600-68,400 Yuan if only peak-hour charging is used (here the results listed are the values with no additional charge for charging electricity). Therefore, the nature of the EV charging load should be clearly stated: the EV charging load is different from ordinary electricity sales, but a load bearing the emission reduction function, in which the off-peak loads also provide the service of load shifting and the promotion of new energy utilization. If the cost of EVs is high, then the society should pay for such services to promote its development, as a source of value for providing subsidies for EVs.

3. Utilization mechanism for sufficient exertion of the overall value of EV charging

3.1. Drawback of EVs support at present

The existing support for EVs focuses on battery subsidies; however, this subsidy is instituted with insufficient scale. According to ''Advice on continuing to promote the application of new energy vehicles'' in China, which grants a subsidy of 2000 Yuan/kW h for batteries, equivalent to a subsidy of 1 Yuan/kW h for electricity if used 2000 times. And at the same time, the charging must pay 1.2 Yuan/kW h at the price of commercial electricity, so that the total operating cost is approximately 0.2 Yuan/kW h. Demonstration

cases show that such support is not sufficient to compensate for the cost of charging facilities and other aspects; thus, it is difficult to form an attractive business model.

Meanwhile, this kind of subsidy mechanism is suitable for establishing a self-contained EV industry system, although is not conducive to coordination and optimization between EVs and the power grid. Thus, developments in EV depend entirely on improving the performance-price ratio of batteries, which poses significant challenges in terms of surpassing fuel vehicles. If battery technology cannot make significant progress within a reasonable amount of time, while hybrid vehicles and other alternative products continue to develop rapidly, then the gap between them will widen increasingly and the EV industry may die prematurely.

3.2. Experience of new energy development

Currently, countries around the world are supporting new energy technologies, such as wind power and photovoltaic power with a preferential feed-in policy and tariffs. For example, China's regulations grant a current subsidy of 0.42 Yuan/kW h for distributed photovoltaic power, paid directly by the grid companies, as well as other services such as grid access and metering [19]. Simultaneously, a tariff of 1.5 Yuan/kW h was added to the price of electricity, except for that used by residents and in agricultural production. Thus, approximately 70-80 billion Yuan can be raised to support renewable energy each year.

Based on the policy to promote wind power and photovoltaic power, several key considerations are worth contemplating. First, to promote EV development, there should be stable and adequate support, so that all parties can have clear expectations in terms of benefits. Once the EV industry develops widely, the construction and operating costs will be significantly reduced; hence, the subsidies required will also be significantly reduced. Two, subsidizing according to the amount of electricity is more reasonable than a one-time subsidy because the benefits of wind power, photovoltaic power, and EVs are generated through long-term operation. Moreover, only electricity-based subsidies can motivate manufacturers and operators to be responsible for the quality of their products. Three, EVs must be effectively combined with the power system, to take full advantage of its operating system to support EVs and reduce the operating costs, while improving the efficiency of the system through coordination. Finally, an appropriate competition mechanism should also be introduced to promote industrial development and technological advancement.

3.3. Utilization mechanism to exert the overall value sufficiently

Thus, Fig. 2 presents an EV operation and support system that is integrated with the power system, which is based on the above analysis of the overall EV values and the experience of new energy development, and can utilize the overall value more sufficiently.

In this system, three approaches can be adopted to support EVs with adequate scale. The first is peak-valley tariff for EV charging. It should include three prices at least, specifically, for on-peak charging, off-peak charging and regulated charging. This approach will reduce the cost of charging for users, and is also likely to attract EVs to participate in load shifting and promotion of new energies utilization, to optimize the system. The second is the

Table 1

The value of EV charging (Yuan/kW h).

Charging time Marginal cost Pollution reduction New energy promotion Overall value

Off-peak hour -0.23-0.0 1.24-2.91 0.28-0.84 1.29-3.75

Peak hour -1.2 1.24-2.91 0 0.04-1.71

for EV

Fig. 2. An EV operation system integrated with power system, and a third-party operator could be present in this system.

charging subsidy paid to EV manufacturers. These subsidies can be calculated according to the amount of electricity that is actually used by EVs or based on off-peak charging. The subsidies provided should increase with the amount of electricity used. The last is one-time battery subsidy. It should be collected by the power companies in responsibility of the government to raise funds for providing a one-time battery subsidy to manufacturers.

Reasonable business models can be formed by incorporating these mechanisms. In particular, the one-time subsidies for batteries should be reduced while increasing the charging subsidies. Accordingly, EV manufacturers that provide excellent products can withstand higher initial costs to lower EV prices, with the costs covered based on charging subsidies, thereby forming a sustainable business model. Here, all the subsidies should be given to the same stakeholder so as to promote a reasonable business model surround it vie market mechanism. This method has been validated in the field of distributed photovoltaic power: manufacturers can provide users with the free installation of photovoltaic equipment. Then, by charging for the electricity and collecting the charging subsidies during the contract lifetime, the manufacturer can hand over the equipment to the user free of charge.

In this system, the government is responsible for the development of rational policy mechanisms, whereas the power companies implement these policies through their operating systems, while providing supporting services to all parties. Funds for demonstration projects can also be raised through flexible approaches, such as setting up a specialized fund for technology and development, or making periodic support policies, as needed. The contents in the above just present an operation mechanism, which can utilize the overall value sufficiently, and will lead to specific operation methods and commercial models in different scenarios. But we cannot discuss these issues due to the length limit of this paper. Additionally, a third-party operator could be present in this system to undertake infrastructure construction and operations management and to provide scheduling, maintenance, accounting, financing and other services to achieve smooth interactions between parties.

The above operating system presents many advantages. The overall benefits of EVs are taken into full account to provide stronger and more adequate financial support. Optimized coordination between power grids can be achieved, and the advantages of EVs are fully used to promote new energy utilization and load shifting. Moreover, the technological progress and survival fitness of the EV industry are better promoted. EV manufacturers must continue to improve the performance of EVs and batteries, and the manufacturers with outstanding products will receive additional subsidies, leaving the unqualified manufacturers subject to elimination. This approach should help to solve the problem that the long-term performance of EVs is difficult to assess. In addition, the tariff policy and power grid operating system are fully considered in this

approach, to provide solid support to the development of EV in terms of funding and operation, while the implementation costs of policy is low.

Of course, the concept presented above is a preliminary idea. Many details remain to be studied in-depth to form detailed policies and methods, which should then be validated and improved through demonstrations.

4. Discussions

This paper mainly focuses on the study of the overall value of EV charging, including marginal cost of electricity for charging, the value of pollution reduction, and the value corresponding to the increase in new energy utilization. Currently, a great deal of researches pay close attention to the economy and environmental benefits of EV charging and have provided many research results.

In the California independent system, smart charge timing can reduce daily electricity costs for driving from $0.46 to $0.2, while provision of regulating power substantially can improve the daily profits amount to $1.71 [15]. Kiviluoma and Meibom [21] estimates the costs and benefits of PHEV considering wind power production and electricity demand by an elaborated model on the power system of Finland. It turns out to be that the system cost to charge an EV was around 36 €/vehicle/year by smart charging while around 263 €/vehicle/year by dumb EVs. The benefits of smart charging come from spinning reserves (17%), intraday flexibility (47%) and day-ahead planning (36%). The smart charging would make the power system emit less CO2 by 211 kg/vehicle/ year, but dumb charging will increase CO2 emission by 169 kg/ vehicle/year. Based on the real data of Western Australia, Jonathan and David [2] gives results that, a vehicle with a 13 kW h battery could attract approximately AU$36/year in capacity payments in national electricity market, and AU$412/year for supply of spinning reserve only, or AU$36/year for supply of load balancing as well as spinning reserve.

These results are very similar to ours', and indicate that the marginal cost of EV charging can be very low, and the provision of ancillary service by EV can achieve a lot of economic profits and environment benefits. An argument just like the suggestion pointed out by our paper is given that, the provision of ancillary services can reduce the net present value of EV greatly and can be even an alternative to expensive tax subsidies [4].

This paper has many contributions and advantages. It illuminated that, when carrying out system issues such as system analysis and design, it is necessary to consider the value of EV charging on a system scale to obtain the overall value, rather than be limited in eyes of single entity such as customers and power grid, or be limited in single aspect like economy, emission reduction, pollution reduction, promotion of new energy, etc. It also presented a model and method for analyzing the overall value of EV charging, which convert various benefits into equivalent economic values and can be used to estimate the overall value of EV charging. The principle of this method is clear and is simple for calculation. So it can be used widely and can provide a useful tool for system analysis and policy design. From the perspective of practice, the method proposed here is of important value for the systematic considerations about EV's development and potential help to EV's industrialization. Facing the shortage of impetus and uncertainty in prospects of EV development, the main idea of this paper is timing, and maybe can help to broaden the possibility of EV development.

Of course, there are still some aspects that need to be improved. The understanding on EV pollution damage and the interaction with renewable energy remain deficient. Due to the lack of relevant statistical data and research experience, the accuracy of this

model needs to be improved, too. It is needed to analyze other benefits of EV charging; maybe some other issues should be included into consideration such as the ancillary services to power system.

By comparing the research methods and results of this paper and the references, it is easy to see that EV charging can participate in various interactions and produce many profits. However, the existing literatures, including this paper, only consider part of these contents without providing more comprehensive study. Also, the analysis results from different countries and different power grids are obviously diverse. So, the value of EV charging is influenced by many factors which we haven't known well. Moreover, facing to these complex interactions and influences, how to grasp the main aspects while ignore secondary aspects needs in-depth study. Finally, we also need to study how to use the analysis results about EV charging to design reasonable charging tariff and support policy, so as to promote the development of EV. Therefore, more comprehensive and systematic research method about EV charging is still a difficulty and a key issue need to be solved for the development of EV. These are also the main issues requiring in-depth study in the future.

5. Conclusions and recommendations

In this research, a model to analyze the overall value of EV that includes peak regulation, pollution reduction and the increasing of new energy utilization was presented, indicating a clear positive value of EV charging and providing the sources for increasing EV subsidies. By comparing EV with photovoltaic power, a model was proposed to calculate the value of emission reduction as a result of EV charging. It was presented that if the amount of damage caused by fuel vehicles is k times that caused by EVs, then the subsidy for EVs should be (k - 1) times greater than the subsidy for photovoltaic power. An analysis model was also proposed for evaluating the improvement in new energy utilization as a result of EVs. If the cost of wind power declines, the savings in the intrinsic value can be transferred to off-peak charging loads to attract EVs to promote wind power consumption and improve the overall efficiency of the system. Meanwhile, a utilization mechanism apt to overall optimization was proposed, where adequate support for EVs can be provided through on-peak and off-peak tariffs, charging subsidies, and one-time subsidies for batteries, and offers many advantages. Finally, some policy hints can be obtained that, when determining whether or not to develop EVs, we should consider this issue from the perspective of the overall value. Although some references pointed out that EV is not as good as other alternatives in some aspect, but its overall value as a system to the society is larger than else options. In addition, it would be a hard work to build an entire operation mechanism to utilize the overall value of EV charging sufficiently.

Acknowledgments

The authors gratefully acknowledge the contributions of Xiang-ning Xiao, Yansong Li, Wenxia Liu, and Hongjun Li for their helps on this document.

References

[1] Fabian K, Christian L. New business models for electric cars - a holistic approach. Energy Policy 2011;39(6). 3392-03.

[2] Jonathan M, David H. The technical, economic and commercial viability of the vehicle-to-grid concept. Energy Policy 2012;48. 394-06.

[3] Ricardo F, Pedro M, Joaquim D. A sustainability assessment of electric vehicles as a personal mobility system. Energy Convers Manage 2012;61:19-30.

[4] Fernández LP, Román tGs, et al. Assessment of the impact of plug-in electric vehicles on distribution networks. IEEE Trans Power Syst 2011;1(26):206-13.

[5] Van MJ, Maggetto G, Lataire PH. Which energy source for road transport in the future? A comparison of battery, hybrid and fuel cell vehicles. Energy Convers Manage 2006;47(17):2748-60.

[6] Geth F, Willekens K, et al. Impact-analysis of the charging of plug-in hybrid vehicles on the production park in Belgium. In: 15th IEEE Mediterranean Electrotechnical Conference (MELECON 2010), Univ Malta; 2010. p. 425-30.

[7] Mills G, MacGill IF. A framework for maximising the economic value of electric vehicle integration into the Australian NEM. In: Universities Power Engineering Conference (AUPEC), Australasian; 2012. p. 1-6.

[8] Andreas S, Thure T. The economics of fast charging infrastructure for electric vehicles. Energy Policy 2012;43(6):136-44.

[9] Lyon TP, Michelin M. Is, ''smart charging'' policy for electric vehicles worthwhile? Energy Policy 2012;41:259-68.

[10] Carla S, Marc S, Tiago F. Evaluation of energy consumption, emissions and cost of plug-in hybrid vehicles. Energy Convers Manage 2009;50(7):1635-43.

[11] Milos P. Stochastic optimal charging of electric-drive vehicles with renewable energy. Energy 2012;36(11):6567-76.

[12] Zhang KK, Xu L, Ouyang M, Wang H, Lu L Optimal decentralized valley-filling charging strategy for electric vehicles. Energy Convers Manage 2014;78:537-50.

[13] Cao Y, Tang S, Li C, Zhang P, Tan Y, Zhang Z, et al. An optimized EV charging model considering TOU price and SOC curve. IEEE Trans Smart Grid 2012;3(1):388-93.

[14] Liu C, Wang J, et al. Assessment of impacts of PHEV charging patterns on wind-thermal scheduling by stochastic unit commitment. IEEE Trans Smart Grid 2012;2(3):675-83.

[15] Rotering N, Ilic M. Optimal charge control of plug-in hybrid electric vehicles in deregulated electricity markets. IEEE Trans Power Syst 2011;26(3):1021-9.

[16] Hamid QR. Distributed recharging rate control for energy demand management of electric vehicles. IEEE Trans Power Syst 2013;28(3):2688-99.

[17] Andersen PH, Mathews JA, Rask Morten. Integrating private transport into renewable energy policy: the strategy of creating intelligent recharging grids for electric vehicles. Energy policy 2009;37(7):2481-6.

[18] CEC, EDF-USA. Study of emission reduction in China electric power. Beijing: China Electricity Council, Environmental Defense Fund (USA); 2013.

[19] CNDRC. Notifications about promoting the healthy development of the PV industry using price leverage. Beijing: Chinese National Development and Reform Commission; 2013.

[20] CSEPA. Emission limits and measurement methods for light vehicle (China III, IV stage). Beijing: Chinese State Environmental Protection Administration Standard of China; 2012.

[21] Kiviluoma J, Meibom P. Methodology for modeling plug-in electric vehicles in the power system and cost estimates for a system with either smart or dumb electric vehicles. Energy 2011;36(3):1758-67.