Inductive Power Transfer for Electric Vehicles: Potential Benefits for the Distribution Grid PDF

Preview

Summary

Inductive Power Transfer for Electric Vehicles: Potential Benefits for the Distribution Grid Salman Mohagheghi, Member, IEEE, Babak Parkhideh, Student Member, IEEE, and Subhashish Bhattacharya, Member, IEEE Abstract—It is believed that the latest advances in battery and external source of electricit...

Description

Inductive Power Transfer for Electric Vehicles: Potential Benefits for the Distribution Grid Salman Mohagheghi, Member, IEEE, Babak Parkhideh, Student Member, IEEE, and Subhashish Bhattacharya, Member, IEEE

Abstract—It is believed that the latest advances in battery and converter technology, along with government mandates on energy independence and resilience, will pave the way for higher deployment of electric vehicles in the transportation fleet. These vehicles, when equipped with bidirectional energy transfer capabilities, can function as mobile energy resources and be utilized in a vehicle-to-grid (V2G) scheme to temporarily inject energy back into the power grid. The forecasted increase in the number of these vehicles can turn them into a considerable energy resource to be used by the utilities as ancillary services or even for long-term integration with the grid. The energy injection into the power system by electric vehicles has been investigated in the literature for charging stations or single residential charging devices. The need for the vehicle to be stationary during the transfer, and the possible drive and/or change in the driving route in order to go the station are some of the hurdles that may lead to inconvenience and hence lower V2G participation by the vehicle drivers. Moreover, the need for an electrical connection between the vehicle and the station makes implementing remote supervisory control schemes difficult, if not impractical. However, with the advent of inductive charging systems for contactless transfer of energy, new horizons have been opened for seamless integration of these resources of energy into the distribution grid. This paper focuses on the applications of inductive power transfer systems for V2G purposes in the modern distribution grid. It will be shown here that such a scheme could potentially allow for supervisory control and management of the mobile energy resources with the ultimate goal of improving the reliability and security of the power grid without the need for capacity expansion. Index Terms— Inductive power transfer, electric vehicles, hybrid electric vehicle, distribution system, demand response, service restoration.

I. INTRODUCTION

T

HE High dependence of traditional vehicles on the finite sources of fossil fuels, the environmental concerns on vehicular pollution and the need for higher energy efficiency, fuel economy and fuel flexibility have all paved the way for the introduction of Plug-in hybrid electric vehicles (PHEV) and electric vehicles (EV). The former is defined [1] as any hybrid electric vehicle which contains at least a battery storage system of 4 kWh or more to power the motion of the vehicle, a means of recharging the battery system from an

S. Mohagheghi is with the Electrical Engineering and Computer Science Department, Colorado School of Mines, Golden, CO, 80401, USA (email: [email protected]). B. Parkhideh and S. Bhattacharya are with the NSF Future Renewable Electric Energy Delivery and Management (FREEDM) System Center of North Carolina State University, Raleigh, NC 27606, USA (email: [email protected])

external source of electricity, and has an ability to drive at least ten miles in all-electric mode, and consume no gasoline. The latter on the other hand denotes any type of vehicle that is solely propelled by electric motors, and has no internal combustion engine (ICE). Such a fleet of electric vehicles, generally referred to here as plug-in electric vehicles (PEV) can be effectively powered by the underutilized electric power grid during the off-peak hours with little need to increase its energy delivery capacity [2]. Traditionally, PEVs have been considered as nonlinear loads for the grid, whose impacts on stability and quality of supply have been studied in detail [2]-[4]. However, new technological advances in power electronics and machine design, together with the government mandates and subsidies for energy independence and resilience of the transportation system, are expected to further accelerate the penetration rate of PEVs into the transportation fleet. At high penetration rates, it is conceptually possible to utilize these vehicles not just as loads but also as energy resources. The energy stored in the batteries of the PEVs can be potentially extracted by discharging the battery for a relatively short duration of time and injecting its energy back into the grid. This service, often referred to as vehicle-to-grid (V2G), can in principle provide peak load shaving, smoothing generation from nondispatchable renewable energy resources and act as a reserve against unexpected outages [5]. As the size of the PEV fleet increases, the bulk energy storage available becomes considerable in size. For instance, Germany is targeting to deploy 1 million EVs by the year 2020, which corresponds to 2.5% of all the passenger cars in the country [6]. Considering the 45 kW continuous electrical power available in Chevy Volt and 80 kW in Nissan Leaf, this could be translated as 45-80 GW of distributed power sources. This power range is comparable to the Germany’s generation capacity which further emphasizes the importance of EVs. The penetration level of EVs in the western states of US almost follows the same pattern [7]. Clearly, utilization of renewable energy resources such as wind and solar is another comparable source of energy; although, their availability is limited due to their intermittent nature. A generation availability index is defined as in (1) and determined based on the data available from US Energy Information Administration (EIA) [8]. A selected set of results for the year 2009 is depicted in Fig. 1. Generation Availability Index (GAI) = Generation Energy (1) (%) Generation Capacity × 8760(h / yr )

30%

30.0 24.7

23.9

25.0

25%

20.0

16.5 20%

15.0

12.2 15%

10.0

10%

5.3 3.3

5% 0.5

3.2 0.003

1.9 0.2 0.1

0% US

Germany

Spain

Denmark

5.0

Generation Capacity (GW)

Generation Availability Index(%)

35%

China

0.0

Australia

Wind Generation Availability Index

Solar Generation Availability Index

Wind Installed Capacity (GW)

Solar Installed Capacity (GW)

Fig. 1. Wind and solar generation availability index for selected countries.

It can be seen in Fig. 1 that despite the diverse regional generation capacities and characteristics, the GAI is less than 35% for all the countries investigated. On the other hand, it has been shown in the past that energy storage can significantly improve the availability factor of renewable resources, [9] . In fact, the amount of centralized energy storage to achieve an hourly dispatch has been found to be around 20% of the power rating of the wind/solar farm, [10][12]. Currently, the size, complexity and economics of utilityscale centralized energy storage control and management may not have been widely accepted and justified. However, with moderate assumptions, the required power and energy available in the EVs –even with today’s battery technologies– can become available in the very near future. Assuming the targeted numbers for Germany in 2020, with only 10% participation of EV owners, a 4.5-8 GW of power, or –as an example– a 2.25-4 GWh of continuous energy can be made available to the utilities. This is considerable enough to maintain all solar and most of wind generation. However, what differentiates the energy stored in electric vehicles from other conventional sources of energy storage is their mobile nature. Like other storage systems, accessibility of these resources depends on the availability of the primary source of energy (charge on the batteries in this case). Although unlike other storage systems, accessibility of PEV energy depends also on the locations of the vehicles. This complicates matters further by adding another constraint to the problem. Traditionally, people have been looking at extracting the stored energy either from locations where a relatively large number of vehicles exist (e.g., charging stations, parking lots) or from individual charging devices at residential units. The common factor in both lies in the fact that the vehicles need to be stationary (i.e., parked). While this is a reasonable assumption in many cases, it undermines a great potential source of energy in the system: vehicles in motion. Furthermore, in order to inject energy back into the grid, it is necessary for the vehicle to drive to the charging station/device, and park the vehicle for the duration of the

power transfer. On top of those, the driver may have to change its normal route to get to the charging station. All this could inconvenience the driver(s) and hence reduce the tendency (or participation level) for taking part in V2G applications. Moreover, when the source of energy is needed fast (e.g., fast reserves) the distances of the vehicles to the charging stations would serve as a limiting factor. Lastly, for the vehicle to be able to transfer power to the grid, the driver (or an operator) needs to be physically present to establish an electrical connection between the vehicle and the station. Connectionless charging stations (based on inductive charging) have been proposed [13]-[16] as a means for contactless energy transfer. Without the need for establishing an actual electrical connection, these systems enable transfer of energy from the power grid to the vehicle and possibly in the reverse direction through the usage of magnetic circuits. Given the right technical specifications, the vehicle can exchange energy with the power grid while driving even close to normal speed. Utilizing this technology can turn the potential energy stored in the batteries of electric vehicles into an accessible source of energy dispersed across the power grid. In addition, a supervisory control scheme can be designed and implemented that remotely controls these individual sources of energy in order to provide an additional source of energy storage for the power grid. This paper focuses on the applications of Inductive Power Transfer (IPT) systems in the distribution grid, and the added values that such installations can provide for the utility, the customers and the power grid itself. The paper provides a brief introduction to V2G benefits and the concept of IPT in sections II and III. In section IV some of the specifications of inductive charging stations as pertained to distribution grid management are discussed. Section V provides a technical insight into the topic of supervisory control of the IPT systems. Some typical applications of the IPT systems in the distribution grid are presented in section VI. Some practical considerations are provided in section VII, and finally concluding remarks appear in section VIII of the paper. II. BENEFITS OF VEHICLE-TO-GRID (V2G) OPERATION A. Additional Resource for Grid Management Energy stored in the battery of PEVs can be utilized as ancillary services for grid management purposes. With very short response times, especially if fast chargers are used, the vehicles’ batteries can quickly inject power into the grid when necessary, whereas a typical peaking power plant could take up to 30 minutes to ramp up to full capacity. A small amount of discharge provided by a large number of vehicles may very well lead to the same behavior as a traditional energy resource. This way, the bulk energy stored in the fleets of PEVs can be theoretically employed to achieve the following: • Frequency regulation – for fast load changes in the order of a minute, assuming the vehicles are equipped with fast chargers • Load following – for slower load changes in the order of 5-30 minutes

•

• •

Spinning reserves – the available energy can be exploited as spinning reserve for immediate response that can reach full output in a matter of a few minutes, supplemental reserve or replacement reserve with a response time of a few minutes to an hour. Voltage control – especially that the converters allow injection of reactive power into the grid Peak shaving – to provide power during the peak load hours

B. Local Integration with Non-Dispatchable Renewable Energy Resources As discussed in the previous section, the energy stored in the batteries of PEVs can also be used in conjunction with renewable energy resources that are inherently nondispatchable, e.g. wind and solar photovoltaics. If and when the vehicles are available for dispatch they can partly alleviate the volatility of these stochastic generation resources by injecting power into the grid when the renewable resource is

not available, or use the excess energy to charge up their batteries when generation exceeds the load. This can potentially help turn the non-dispatchable renewable resource into a dispatchable source of power. Theoretically, the objective can be achieved by using either the vehicles parked at the parking lots and charging stations, the individual vehicles parked at the residential units, or alternatively by utilizing the energy available through mobile vehicles. Regardless of the approach adopted, the energy stored in the vehicles’ batteries can be used to respond to the fluctuations in the nature of the non-dispatchable energy resources, as well as uncertainties in the network demand. A coordinated control structure as shown in Fig. 2 would ideally monitor the capacity of the main energy source, the forecasted load of the network, and the market data associated with energy prices. In return, it would coordinate the charge/discharge process of the individual chargers and/or charging stations.

Fig. 2. Integration of electric vehicles with non-dispatchable energy resources.

C. Environmental Impact By injecting power into the network, PEVs can help reduce the environmental burdens placed on the air, land and water by reducing or delaying construction and development of new power plants and by allowing usage of the current generation and transmission capacity more effectively. Furthermore, this could help reduce the environmental emissions of the power plants during peak hours. Similar to the financial benefits gained from demand response, the utility can take advantage of the PEV storage to defer building new plants and hence save money. As an example, the overall US capacity is close to 900 GW. Assuming the cost of building a power plant is 1 M$/MW with an average lifespan of 50 years, a hypothetical reduction of

1% in the demand would lead to savings of 1 M$/MW × 9000MW / 50 yr = 180 M$/yr. III. PRINCIPLES OF INDUCTIVE POWER TRANSFER Generally, an IPT system consists of two main parts. One part takes the power from the grid and energizes the track. The other part is a so-called pickup coil. The power is induced in the pickup and transferred to the load. A typical IPT system is illustrated in Fig. 3. Compensation circuits in the IPT system are often resonant circuits to increase the induced current (power) which is normally limited by the self-inductance of the track, misalignment, etc. The output of the IPT system is a regulated DC voltage using a combination of a diode bridge and a switched-mode DC/DC converter. The details of the IPT

system shown in Fig. 3 can be found in [14], [15].

Fig. 3. Schematic diagram of a typical IPT system.

Thus far, unidirectional IPT systems have been mainly investigated because of several improvement potentials in the key market of the technology. However, the IPT system can be bidirectional. With minor changes in the resonant circuits (to be symmetrical), and bidirectional converters, two way IPT system can be realized. A general schematic of a bidirectional IPT system is presented in Fig. 4. In this scheme, a switched mode bidirectional converter would replace the DC/DC converter in the conventional IPT. This way it allows for power transfer in both directions. One of the very few examples of this circuit has been presented in [16].

present. Furthermore, in theory, the inductive charging station can allows for charge/discharge of the batteries while the vehicle is driving on the track, which leads to increased convenience for the drivers. Clearly, this requires the usage of fast chargers that can accomplish the energy transfer within the available time-frame. Such convenient realization of energy transfer would increase the participation level of drivers in V2G schemes initiated by the utility, such as demand response. Although, the slower speed of the charger can be compensated for by building longer tracks that increase the contact time with the vehicle. For instance, the track can be built as a side lane on the street, where the vehicles intending to discharge (or charge) their batteries would take a detour and drive over the inductive charging track. The other extreme alternative is of course the park-andcharge or park-and-discharge type of structure where the vehicles would park on the track for the duration of the energy transfer. In order to justify the higher expenses of installing IPT versus the more conventional connection-type chargers, this scheme could be best beneficial in large parking areas such as terminal parking lots where a remotely enabled supervisory controller would charge/discharge the batteries as needed. More explanations are provided in the following section. V. SUPERVISORY CONTROL OF IPT STATIONS

Fig. 4. General schematic of bidirectional IPT system.

The bidirectional flow of power in the IPT if employed for electric vehicles enables the energy transfer from the vehicle’s battery back to the grid. This way, the vehicle can be viewed as a mobile energy resource that can be integrated into the distribution grid for various V2G applications. IV. INDUCTIVE CHARGING STATION SPECIFICATIONS Inductive charging stations provide benefits over the connection-oriented stations. If energy transfer to the grid is desired, the latter requires the driver to change its route and drive to the station, park the vehicle for the duration of the transaction, and connect to and subsequently unplug the electrical connection from the charger. However, the former scheme is connectionless, which means the energy transfer can be more easily enabled through a remote control process, even when the human operator (i.e., driver) is not physically

The traditional power system control where all the generation units were owned and operated by the utility has now transformed into an open system where various sources of energy (utility owned or autonomous) coexist and interact in the same environment with the overall objective of providing a secure reliable network with high quality of supply. Any portion of such a system is typically equipped with thousands of sensors and measurement units that capture the local state of the system and send it to actuators and controllers. The sheer size of the network necessitates the transport of this data and command signals over long distances across the wide geographical area of the grid, essentially forming a networked control system [17]. Depending on the structure of the grid, the number of components to control and the size of the system, the control system can take on the form of a fully centralized, decentralized or hierarchical system [18]. Regardless of the control structure adopted, for distribution systems, due to the high and direct impact that the system has on the end use customers, having a centralized structure that accurately supervises the performance of the system components deems necessary. Such a system would be the heart of the distribution management system (DMS) and would be in charge of ensuring that all the devices are operating within acceptable limits and the system meets the overall operation criteria set forth by the operator. Among other things, such a system which resides in the utility control center would supervise and manage the ...