EV/PHEV Bidirectional Charger Assessment for V2G ... - IEEE Xplore

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 28, NO. 12, DECEMBER 2013

5717

EV/PHEV Bidirectional Charger Assessment for V2G Reactive Power Operation Mithat C. Kisacikoglu, Student Member, IEEE, Burak Ozpineci, Senior Member, IEEE, and Leon M. Tolbert, Fellow, IEEE

Abstract—This paper presents a summary of the available singlephase ac–dc topologies used for EV/PHEV, level-1 and -2 on-board charging and for providing reactive power support to the utility grid. It presents the design motives of single-phase on-board chargers in detail and makes a classification of the chargers based on their future vehicle-to-grid usage. The pros and cons of each different ac–dc topology are discussed to shed light on their suitability for reactive power support. This paper also presents and analyzes the differences between charging-only operation and capacitive reactive power operation that results in increased demand from the dc-link capacitor (more charge/discharge cycles and increased second harmonic ripple current). Moreover, battery state of charge is spared from losses during reactive power operation, but converter output power must be limited below its rated power rating to have the same stress on the dc-link capacitor. Index Terms—Battery charger, electric vehicle, reactive power, vehicle-to-grid (V2G).

I. INTRODUCTION CCORDING to the international energy outlook report, the world transportation energy usage is going to increase up to 44% until 2035 (compared to 2008) [1]. Therefore, technologies related to reducing oil consumption have one of the utmost challenges in today’s vehicle research. Alternative vehicle technologies that can be connected to the grid include plug-in hybrid electric vehicles (PHEVs) and electric vehicles (EVs)—also known as battery electric vehicles (BEVs) which will be termed collectively as plug-in electric vehicles (PEVs) in this study. The dichotomy between hybrid electric vehicles (HEVs) and PEVs is the presence of a charger in the latter group. Chargers convert the ac voltage to a dc magnitude for the specific battery

A

Manuscript received November 6, 2012; revised January 21, 2013; accepted February 11, 2013. Date of current version June 6, 2013. This work was supported by the Engineering Research Center Program of the National Science Foundation and the Department of Energy under NSF Award EEC-1041877 and the CURENT Industry Partnership Program. This paper was prepared by the Oak Ridge National Laboratory, Oak Ridge, TN, USA, and managed by UT-Battelle for the U.S. Department of Energy under Contract DE-AC05-00OR22725. Recommended for publication by Associate Editor A. Khaligh. M. C. Kisacikoglu is with the Department of Electrical Engineering and Computer Science, University of Tennessee, Knoxville, TN 37996 USA (e-mail: [email protected]). B. Ozpineci and L. M. Tolbert are with the Oak Ridge National Laboratory, Oak Ridge, TN 37831 USA, and also with the University of Tennessee Knoxville, TN 37996 USA (e-mail: [email protected]; [email protected]). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TPEL.2013.2251007

needs of PEVs. Chargers have an important role in the grid integration of PEVs that can help reduce the negative impact of a large number of PEV connections. It is possible to incorporate more than one operation mode in a charger by allowing the power to flow bidirectionally. Usually, the bidirectional power transfer stands for two-way transfer of active power between the charger and the grid. The general term of sending active power from the vehicle to the grid is called vehicle-to-grid (V2G). The economic benefits of this operation has been a research subject for more than a decade because of the large energy reserve of an electric vehicle battery and the potential of thousands of these connected to the grid [2], [3]. While PEVs potentially have the capability to fulfill the energy storage needs of the electric grid, the degradation on the battery during this operation makes it less preferable by the auto manufacturers and consumers unless a properly structured battery warranty and compensation model is implemented [4]–[7]. On the other hand, the on-board chargers can also supply energy storage system applications such as reactive power compensation and voltage regulation without the need of engaging the battery with the grid and thereby preserving their lifetime. Reactive power consumed at the load side is transmitted from the energy source to the load through the transmission and distribution system. This causes increased energy losses and decreases the system efficiency. For long distances, line reactance for line “k” (Xk ) becomes much larger than the line resistance Rk . Because reactive power losses are proportional to line susceptance Bk = −Xk /(Rk2 + Xk2) and real power losses are proportional to line conductance Gk = Rk /(Rk2 + Xk2 ) , the relative losses of reactive power become much greater than the relative losses of active power on the transmission lines [8]. Therefore, reactive power is best utilized when it is generated close to where it is needed. Moreover, residential appliances such as microwaves, washing machines, air conditioners, dishwashers, and refrigerators consume reactive power for which the residential customers do not pay, but the utility is responsible to deliver. PEVs can readily supply this reactive power need locally without the need of remote VAR transmission. Fig. 1 shows the proposed application of PEVs. Customers with a PEV that carries an on-board charger can negotiate with the utility grid to allow the usage of the charger for grid support. The charger compensates for the reactive current ic that either the customer with the PEV or other customers without a PEV demand from the utility grid. Generating reactive current at the point of common coupling (PCC) provides increased efficiency of power transfer through transmission lines and decreases transformer overloading.

0885-8993/$31.00 © 2013 IEEE

5718


TABLE I MAXIMUM HARMONIC CURRENT DISTORTION FOR SINGLE-PHASE ON-BOARD BIDIRECTIONAL CHARGERS [16], [17]

II. BATTERY CHARGERS FOR PHEVS AND EVS Fig. 1.

Proposed reactive power support diagram using PEVs.

A vehicle can provide reactive power irrespective of the battery state of charge (SOC). The charger can supply reactive power at any time even during charging. However, the selected topology and the effect of the reactive power on the operation of the charger and the battery should be well analyzed. On-board single-phase charging systems have been researched in terms of different power factor corrected (PFC) rectifier topologies that can be used for unidirectional charging operation [9], [10]. Other studies have surveyed bidirectional single-phase ac–dc converter topologies that are suitable for V2G applications [11]–[13]. Single-phase battery-powered renewable energy systems have also been well researched in terms of ac–dc power transfer calculations, second harmonic current ripple elimination, and reduction of electrolytic dc-link capacitors [14], [15]. However, there is a need in the literature for technical analysis and survey of topologies suitable for V2G reactive power operation for single-phase on-board PEV charging systems and its effect on both the charger design and battery charging operation. The analysis in this paper shows the effect of reactive power operation on the design and operation of single-phase on-board chargers that are suitable for reactive power support. It further introduces a classification of single-phase ac–dc converters that can be used in on-board PEV chargers based on their power transfer capabilities in addition to the currently available surveys. Section II of this paper discusses battery charging and charger design requirements. Section III is concerned with the classification and introduction of different ac–dc topologies. An onboard charger is composed of two power conversion stages: a singlephase ac–dc conversion stage and a dc–dc conversion stage. This section technically focuses on single-phase ac–dc converters suitable for on-board charging and V2G applications. The design of the charger changes considerably between the different options and applications. Section IV focuses on reactive power analysis of single-phase ac–dc converters and their adverse effect on the dc-link capacitance and battery. Section V compares all the topologies based on the component count and size, and their suitability of V2G reactive power operation.

The superiority of lithium-ion (Li-ion) batteries has been demonstrated over other types of batteries in supplying greater discharge power for faster acceleration and higher energy density for increased all-electric range. Higher efficiency operation and lower weight make them preferable for vehicular applications. The common charging profiles used in the industry for Li-ion batteries are constant current (CC) and constant voltage (CV) charging. During CC charging, the current is regulated at a constant value until the battery cell voltage reaches a certain voltage level. Then, the charging is switched to CV charging, and the battery is charged with a trickle current applied by a CV. One of the important requirements of an EV/PHEV charger is the amount of current distortion that it draws from the grid. The harmonic currents need to be well regulated not to cause excess heat which decreases the distribution transformer lifetime. There are two definitions to measure the harmonic content of the battery charger current. The first parameter is total harmonic distortion (THD) defined as follows: THD =

Ic,h Ic,1

(1)

where Ic,h is the root mean square (rms) sum of the harmonics (usually up to n = 39) of the charger current (A), i.e., 39 2 Ic,h = n =2 Ic,n , and Ic,1 is the rms fundamental (60 Hz) component of the charger current (A). However, this definition is not enough to account for all charging current rates of a charger. When there is a need to control the charger input current to help reduce the demand from the grid, the rms charger current may need to be reduced to less than 50% of the rated current. As loading on the grid decreases, the harmonic content of the charger current is not as disturbing to the grid as when the loading is high. In such cases, THD does not reflect the real impact of the harmonic content of the charger on the grid. Therefore, total demand distortion (TDD) can be used to accurately evaluate the harmonic content of the charger between 0% and 100% loading range. The definition of the TDD is shown in TDD =

Ic,h Ic,1,rated

(2)

where Ic,1,rated is the rated fundamental current of the charger (A). The only difference between TDD and THD is the change in the denominator. TDD is equal to THD when charging occurs at the rated current, i.e., Ic,1 = Ic,1,rated . Table I lists the limits for the harmonic content of the single-phase chargers

KISACIKOGLU et al.: EV/PHEV BIDIRECTIONAL CHARGER ASSESSMENT FOR V2G REACTIVE POWER OPERATION

5719

TABLE II DIFFERENT TYPES OF CHARGERS BASED ON POWER TRANSFER OPERATION

operating either as a load or as a distributed generator based on the limits shown in [16] and [17]. It is important to note that the charger should meet the individual harmonic limits as well as the TDD limit which are calculated separately, i.e., Ic,3 /Ic,1,rated < 4.0%, Ic,13 /Ic,1,rated < 2.0%, etc., and TDD < 5% using (2). The charger’s dc output voltage and current must also be well regulated. In other words, the harmonic components present at the output voltage and current must be less than the maximum levels allowed to protect the health, and thereby lifetime of the battery. Increased losses in the cell due to ripple current could cause temperature increase, which results in degradation on the battery cell. Therefore, battery manufacturers implement battery charging current and voltage standards in the industry. General limits for Li-ion and/or lead-acid batteries are charging rms current ripple of 5–10% of the rated charging current and rms ripple voltage of 1.5% of the rated battery voltage [18], [19].

Fig. 2.

Available power operation region (P–Q plane) for chargers.

Fig. 3.

Conventional ac–dc boost converter.

Fig. 4.

Interleaved ac–dc boost converter.

III. EV/PHEV CHARGER POWER ELECTRONICS FOR SMART-GRID INTEGRATION The charger topologies investigated in this section are singlephase level-1 and level-2 compatible bidirectional chargers. The common nominal battery voltage levels in PHEVs and EVs, that are in the market, are in between 300 and 400 V [20]. Due to high battery voltage and a 120 V/240 V grid connection, a boost rectification stage is preferred over a buck rectification stage to prevent an unnecessarily high conversion ratio between the dc-link and the battery terminals. A charger can be configured in two different ways in terms of its active Ps and reactive power Qs transfer capability with the utility grid. Table II lists the possible charger configurations. The P–Q plane shown in Fig. 2 describes the available operation regions for the charger (positive Ps or Qs means that the power is sent to the charger by the grid). Pm ax and Qm ax are equal to the apparent power rating of the charger. First, the PFC unidirectional chargers that are in use in today’s PHEVs and EVs operate at the positive x-axis of the P–Q plane. It operates close to unity power factor (pf) and allows control of the active power used to charge the battery. This type of charger can also be used for limited reactive power support, but it is not intended to be used for low-pf applications and has several disadvantages that are discussed in Section IV. The second option is the four-quadrant bidirectional charger that operates in all four regions as shown in Fig. 2.

A. PFC Unidirectional Chargers PFC unidirectional chargers only transmit power from the utility to the vehicle battery and operate with almost unity input pf. In other words, they are not designed to exchange reactive power with the grid. Today, most of the PHEVs and EVs in the market use this type of charger. Some of the unidirectional ac–dc rectification stages are highlighted in the next sections. 1) Conventional AC–DC Boost Converter: In this topology, a front-end diode bridge is used to rectify the input voltage, and it is followed by a boost section as shown in Fig. 3. This topology is widespread for low-power applications. Due to conduction losses of the diode bridge, it is not well suited for power levels higher than 1 kW [9], [21]. Another problem is the design of the dc inductor at high power levels. As a solution to this problem, interleaving techniques are proposed as shown in the next section. 2) Interleaved AC–DC Boost Converter: Interleaving the boost section of the conventional PFC is first introduced in [22] and shown in Fig. 4. The main advantage of this topology is

5720


Fig. 5.

Symmetrical bridgeless boost rectifier.

Fig. 6.

Asymmetrical bridgeless boost rectifier.

Fig. 7.

Dual-buck ac–dc half-bridge converter.

decreased high-frequency pulse width modulation (PWM) rectifier input current ripple caused by the switching action. Reducing input ripple decreases the required switching frequency to meet a current TDD limit imposed by the utility. Reducing PWM input ripple current also decreases the ac ripple current supplied by the dc-link capacitor, thereby reducing its stress. Another advantage is the reduced current rating of the active switches as the interleaving converter halves the input current. One disadvantage of the topology is the high conduction losses of the input bridge rectifier as well as increased number of semiconductor devices and associated gate control circuitry. This topology is preferred by the industry for on-board charging applications and is used for 3.3-kW level-2 chargers [23], [24]. 3) Symmetrical and Asymmetrical AC–DC Boost Converters: The topology proposed in [25] is called symmetrical bridgeless boost rectifier and is shown in Fig. 5. Another topology called asymmetrical bridgeless boost rectifier is proposed in [26] and is shown in Fig. 6. They eliminate the input diode bridge to attain higher efficiencies at increased power levels. B. Four-Quadrant Bidirectional Chargers 1) Dual-Buck AC–DC Half-Bridge Converter: A dual-buck ac–dc half-bridge converter shown in Fig. 7 was first introduced in [27] and was also employed for a battery storage

Fig. 8.

AC–DC half-bridge converter diagram.

Fig. 9.

AC–DC full-bridge converter diagram.

system to demonstrate four-quadrant operation capability with increased efficiency [28]. By placing the two active semiconductor switches in a diagonal structure rather than symmetrical/asymmetrical structure, four-quadrant operation is achieved. The circuit does not need shoot through protection as there are no active switches connected in series. The circuit requires two split dc-link capacitors and two coupling inductors. 2) Conventional AC–DC Half-Bridge Converter: This type of converter diagram is illustrated in Fig. 8. It includes two dc-link capacitors, two switches, two diodes, and a coupling inductor for grid interconnection. Two sufficiently large capacitors share the dc-link voltage equally. The switches Q1 and Q2 cannot be ON at the same time to prevent any short circuit or shoot through. This requires a dead time when the switches are operated sequentially. When the switch Q1 is ON, either Q1 or D1 conducts depending on the direction of the charger current. Similarly, when the switch Q2 is ON, either Q2 or D2 conducts depending on the current direction. The topology is suitable to transfer power in four quadrants. A half-bridge converter requires bipolar switching because there are only two possible output voltage levels, +Vdc and −Vdc . 3) AC–DC Full-Bridge Converter: The full-bridge converter, shown in Fig. 9, is comprised of a dc-link capacitor, four transistors (either MOSFETs or IGBTs), four diodes, and a coupling inductor. Voltage of the capacitor is doubled in this configuration. The topology is suitable for four-quadrant operation. The full-bridge converter can operate in unipolar modulation and has three output voltage levels: +Vdc , −Vdc , and zero. Since there are three output voltage levels for the full bridge inverter, the number of switchings required for the same current THD level is effectively reduced with the full-bridge converter compared to half-bridge converter.


5721

Using the phasor analysis, one can find that √ ic (t) = 2 Ic sin(ωt − θ) where θ = tan−1

Fig. 10.

Vs − Vc cos(δ) Vc sin(δ)

ps (t) = vs (t) ic (t). IV. REACTIVE POWER OPERATION ANALYSIS Although PFC unidirectional chargers are mostly suited for high pf applications, they can still be used for reactive power compensation with certain limits. However, there are two main disadvantages of this operation. First, reactive power operation can only be achieved by natural commutation of current through the diodes. This poses a strict limit on the amount of phase difference that can be introduced between the grid voltage and grid current depending on the inductance value of the boost inductor. Otherwise, the current THD exceeds the allowed limit by the utility. A second disadvantage is that the charger must always be charging the battery in order to supply reactive power to the grid. In other words, if the battery has full SOC, reactive power operation is not possible. Considering these two limitations, PFC unidirectional chargers are not promising compared to other type of topologies for reactive power operation. All of the four-quadrant topologies are suitable for reactive power compensation. However, reactive power analysis should be performed to identify the limitations of the topologies (if any) during different operation modes. Four-quadrant topologies can be simplified to a circuit as shown in Fig. 10 with a couple assumptions. First, the grid voltage vs (t) is assumed to be purely sinusoidal and the total impedance between the grid and the charger is all inductive and lumped together as Lc . Second, for analysis purposes, the equivalent system has only one frequency component (60 Hz) and the charger input voltage can be simplified to a sinusoidal voltage source, neglecting PWM ripple components. The current flow is designated positive for current flowing from the grid to the inverter, since the primary function of the circuit is to charge the battery from the grid. Therefore, positive power sign of active power Ps , reactive power Qs , and apparent power S correspond to the power sent by the grid to the charger. The grid voltage is √ (3) vs (t) = 2 Vs sin(ωt).

(4)

where Vs and Vc are the rms of the grid voltage and charger output voltage, respectively (V), f is the system frequency (Hz), and ω is the angular frequency (rad/s).

(7)

After plugging (3) and (5) into (7), the instantaneous input power can be written as ps (t) = Vs Ic cos(θ) − Vs Ic cos(2ωt − θ).

(8)

Using the definition of the voltage of an inductor (vL (t)), the instantaneous power of the coupling inductor can be calculated pL (t) = vL (t) ic (t) = Lc

dic (t) ic (t) dt

= w Lc Ic 2 sin(2ωt − 2θ).

(9)

The instantaneous power that the charger receives is equal to pc (t) = ps (t) − pL (t) = Vs Ic cos(θ) − Vs Ic cos(2ωt − θ) − w Lc Ic 2 sin(2ωt − 2θ).

(10)

As shown in (10), the instantaneous charger input power contains two components: the average power and the ripple power component at twice the grid frequency Pave = Vs Ic cos(θ)

(11)

pripple (t) = −Vs Ic cos(2ωt − θ) − ωLc Ic 2 sin(2ωt − 2θ).

(12)

The ripple component can be summed into a single sinusoidal function using phasor addition since both of the sinusoidal components of (12) are at the same frequency. Therefore pripple (t) = Pripple cos(2ωt + β)

(13)

where ripple power is found using the fundamental power equations (S = Vs Ic and Qs = Vs Ic sin(θ)) as follows: 2 S2 S2 2 Pripple = S + ωLc 2 − 2ωLc 2 Qs (14) Vs Vs and

Because the default active power flow is from the grid to the charger, vc (t) is lagging vs (t) by δ ◦ : √ 2 Vc sin(ωt − δ)

(6)

and Ic is the rms value of the charger current (A). The definition of the instantaneous power drawn from the grid is as follows:

Equivalent circuit of the charger-grid connection.

vc (t) =

(5)

β = tan−1

Vs Ic sin(θ) + ωLc Ic 2 cos(2θ) . −Vs Ic cos(θ) + ωLc Ic 2 sin(2θ)

(15)

The ripple power component of the instantaneous charger power (pripple (t)), shown in Fig. 11, is a result of the singlephase conversion of ac power into dc power, and it is an oscillating power between the grid and the charger. The average power is used to charge/discharge the battery. The oscillating

5722


dc-link voltage is equal to Vdc = (Vdc,m ax + Vdc,m in )/2. Therefore Eripple = Cdc ΔVdc Vdc .

(19)

Plugging (17) into (19) results in the following equation:

2 2 2 S 2 + ωLc VS 2 − 2ωLc VS 2 Qs s s . (20) Cdc = ω ΔVdc Vdc Equation (20) suggests that for a fixed Vdc and ΔVdc , capacitance requirement increases with increasing reactive power support to the grid. The second harmonic rms current ripple of the dc-link capacitor can be found using the definition of the dc-link voltage. Neglecting the PWM ripple components, instantaneous dc-link voltage can be expressed as Fig. 11.

Instantaneous charger input ripple power.

ripple power is temporarily stored in the dc-link capacitor of the charger. It is used to balance the power transfer between the grid and the charger. The energy that needs to be stored in the dc-link capacitor is calculated based on the integration of the ripple power between its minimum and maximum values. Therefore tm a x

pripple (t) dt Eripple = tm in

tm a x

=2

Pripple cos(2ωt + β)dt.

(16)

tz e r o

Therefore Eripple

1 = ω

2 S2 S2 S + ωLc 2 − 2ωLc 2 Qs . Vs Vs 2

(17)

For a predefined average dc-link voltage Vdc and a dc-link peak-peak voltage ripple ΔVdc , the required capacitance and the required ripple current capability of the capacitor are directly proportional to the ripple energy, Eripple calculated in (17) . Equation (17) is an important conclusion of the reactive power analysis of the aforementioned four-quadrant single-phase ac– dc power conversion systems used for chargers. Equation (17) suggests that if the charger is used to source reactive power (Qs < 0) in addition to the charging active power, the required ripple energy increases even if the rated operation power S stays the same. This, in turn, causes more second harmonic ripple current and a higher peak-to-peak ripple voltage at the dc link. Once the values for Vdc and ΔVdc are selected for the battery charger, the selection of the dc-link capacitor is done using two parameters: capacitance and second harmonic ripple current rms value. In order to find the required capacitance, the maximum energy that can be stored in a capacitor is made equal to the ripple energy of the charger, Eripple , such that 1 2 2 Cdc {Vdc,m (18) ax − Vdc,m in } 2 where Vdc,m ax and Vdc,m in are the maximum and minimum dc-link voltages, respectively. The difference between the maximum and minimum voltages is ΔVdc = Vdc,m ax − Vdc,m in and Eripple =

ΔVdc sin(2ωt). 2 Therefore, the second harmonic current can be found as vdc (t) = Vdc +

dvdc dt = ω Cdc ΔVdc cos(2ωt).

(21)

icap (t) = Cdc

(22)

Hence, the rms value of the current is 1 Icap = √ ω Cdc ΔVdc . (23) 2 By using (20), it can also be written in terms of circuit parameters

2 2 2 − 2ωLc VS 2 Qs S 2 + ωLc VS 2 s s √ . (24) Icap = 2 Vdc Again, the required ripple current increases with increasing capacitive reactive power support to the grid. Both (20) and (24) suggest that the reactive power operation of the converter increases the ripple current and capacitance requirement of the converter. The selection of the dc-link voltage is also dependent on the operation mode of the charger. The ac–dc converter has to operate in linear modulation to satisfy the sinusoidal current assumption. Therefore √ (25) Vdc ≥ 2Vc . In order to find the limit on the dc-link voltage and how this limit changes with the power transfer, the charger voltage Vc will be derived in terms of reactive power. Applying Kirchhoff’s voltage law KVL on Fig. 10 and using voltage definitions and their phasor representations yields Vc = Vs − VL √ √ = −j 2 Vs − jω Lc { 2 Ic (−sin(θ) + jcos(θ))} √ = 2 ω Lc Ic cos(θ) √ √ − j{ 2 Vs + 2 ω Lc Ic sin(θ)} √ √ Ps Qs − j 2 Vs − ω Lc = 2 ω Lc . (26) Vs Vs


5723

TABLE III SYSTEM PARAMETERS USED FOR CASE 1

Calculating the rms value of (26), it can be written that 2 S 2 2 Vc = Vs + (ω Lc ) − 2 ωLc Qs . Vs

Fig. 12. Safe reactive power support area of a 3.3-kW charger (shown as the dark-red shaded area).

(27)

Using (25), the limit on the dc-link voltage can be written as 2 S 2 2 Vdc ≥ 2 Vs + (ω Lc ) − 2 ωLc Qs . (28) Vs The next section investigates two different charger cases available in the market [20] to analyze and quantify the limitations on reactive power operation to preserve charger operating limits. A. Case 1: 3.3-kVA Charger Design The first design case has the design parameters shown in Table III. Based on the battery terminal voltage survey in the market, the full SOC of the battery can be as high as 390 V [20]. Moreover, using the inequality of (28), the dc-link voltage must be in excess of 360 V depending on the operation mode of the converter and the selected coupling inductor. In addition, considering the voltage drops and second harmonic peak-peak voltage ripple at the dc link, the dc-link voltage is selected to be 450 V. The dc–dc converter operates in buck operation mode and steps down the voltage from 450 V to the required battery terminal voltage. Therefore, the gain is not constant and less than 1. The capacitance and ripple current rms values are calculated based on charging only operation of the converter (S = Ps and Qs = 0) using (20) and (24). Using the parameters given in Table III, the resulting values are Cdc = 432 μF and Icap = 5.2 A. Again, using (20) and (24), the reactive power operation limits of the charger can be calculated. If the charger dc-link capacitor is designed for unidirectional charging only operation, the righthand side of (20) and (24) should not exceed 432 μF and 5.2 A, respectively. The active and reactive power operation points that satisfy Cdc = 432 μF and Icap = 5.2 A have been calculated using (20) and (24). The result is that the charger input reactive power must be limited. In other words, the charger cannot operate symmetrically among the power plane and at the same time require the same dc-link capacitor parameters with charging only operation. Depending on the coupling inductor selected at the front end (i.e., Lc = 1.0 – 2.0 mH), the reactive power output is limited to 95–97.5% of the total input power. For instance, Fig. 12 shows the limit on reactive power for Lc =2 mH coupling inductor case to warrant safe operation of the dc-link capacitor. It shows

Fig. 13. Change of C d c and Ic a p with varying Q s and for different L c values for the case of S = 3.3 kVA.

that the charger should have a limitation on its input power if it sends reactive power back to the grid. The maximum reduction of power occurs when it supplies full capacitive reactive power (Ps =0 and Qs = −S). The limitation gets worse if the total coupling inductor between the charger and the grid increases further. On the other hand, using (20), Fig. 13 shows the change in the required dc-link capacitance and ripple current to further increase the reactive power output to the full 100% rated power. There are three different cases analyzed: 1, 1.5, and 2 mH coupling inductor values. As shown, the converter requires higher dc-link capacitance and higher ripple current for increasing inductor values and for increasing reactive power output. Investigating Fig. 13, the net increase from 0% reactive power to full 100% reactive power in both capacitance and ripple current requirements of the dc-link capacitor is 2.1–4.3% depending on the selected coupling inductor. In conclusion, the higher the coupling inductance gets, the larger the dc-link capacitor becomes to supply 100% reactive power. B. Case 2: 6.6-kVA Charger Design The second design case has also the same reactive power versus capacitance/ripple current relationship, but the effect is more pronounced. The charger input apparent power S is 6.6 kVA, Vdc =450 V, and ΔVdc = 45 V. For the charging only operation, the converter requires twice the capacitance and ripple current capability [using (20) and (24)] resulting in 865 μF and 10.4 A, respectively. Again, using (20) and (24), the reactive

5724


TABLE IV COMPARISON OF THE AC–DC CONVERTERS BASED ON COMPONENT COUNT AND SIZE

TABLE V COMPARISON OF THE AC–DC CONVERTERS BASED ON V2G OPERATION

power output limitation of the charger can be calculated. Depending on the coupling inductor selected at the front end (i.e., Lc = 1.0–2.0 mH), the reactive power output is limited to 92– 96% of the total input power. The limitation gets worse if the total coupling inductor between the charger and the grid increases further. On the other hand, if the charger is required to supply full reactive power, based on (20) and (24), the charger requires 4.3–7.6% more capacitance and ripple current requirement for 100% capacitive operation compared to unidirectional charging operation. In short, the 6.6-kVA charger requires a more profound increase at the dc-link capacitor (compared to 3.3-kVA charger) to satisfy full reactive power operation. V. COMPARISON OF AC–DC CONVERTERS CAPABLE OF V2G REACTIVE POWER SUPPORT The topologies that can be used for reactive power support are compared in this section. The first comparison lists the active component count and rating comparison for the different topologies shown in Table IV. The full-bridge converter requires four active switches whereas all the other converters require only two active switches. All the current and voltage ratings of the switches among all the converters are the same. The dual-buck and half-bridge converters require two dc-link capacitors each with the same size of what the full-bridge converter has. Also, the dual-buck converter requires two separate ac coupling inductors. Table V shows the comparison of the converters in their ability of V2G operation in all four quadrants for two different design cases: 3.3 and 6.6 kVA to summarize the results shown in the previous section. The benchmark dc-link capacitance values used for Table V are 432 and 865 μF for 3.3 and 6.6 kVA chargers, respectively. They are calculated based on rated active power charging. All of the converters can supply and consume reactive power in addition to active power operation. However, each of them have limitations in the amount of capacitive reactive power (Qs < 0) that they supply to the grid. If the charger is designed to operate at 3.3 kVA, then the capacitive reactive power-only

TABLE VI COMPARISON OF THE AC–DC CONVERTERS BASED ON CAPACITOR INCREASE FOR 100% REACTIVE POWER OPERATION

operation must be limited to a maximum of 95% of the rated power (for Lc =2 mH) to keep the stress on the dc-link capacitor the same. The limitation decreases to 92% of the rated power if the converter is designed at 6.6 kVA. On the other hand, Table VI shows the required increase in the dc-link capacitor in order to supply 100% reactive power back to the grid (symmetrical operation). Full-bridge converter requires a maximum increase of 4.3% and 7.6% in both dc-link capacitance and ripple current capability for 3.3 and 6.6 kVA charger designs, respectively. Half-bridge and dualbuck converters require the greatest increase in the dc link due to having twice the dc-link capacitor compared to the full-bridge converter. They require a maximum of 8.6% and 15.2% increase for both dc-link capacitance and ripple current capability for 3.3and 6.6-kVA charger designs, respectively. In order to validate the reactive power operation analysis, an example simulation case is performed using ac–dc full-bridge converter [29]. There are five different operation modes simulated in the following order: 1) full inductive power; 2) full capacitive power; 3) full charging only operation; 4) 65% charging and 75% inductive power operation; and 5) 65% charging and 75% capacitive power operation. The charger is rated at 3.3 kVA, the coupling inductor Lc is 2.0 mH, and the dc-link capacitor Cdc is 432 μF. Fig. 14 shows the output power of the charger (again, Ps > 0 for charging and Qs > 0 for inductive reactive power). The converter operates below its power rating (3.3 kVA) at all operation modes. It satisfies the grid and battery current harmonic requirements at all operation modes. To further demonstrate the grid and charger interaction, Fig. 16 shows the zoomed versions (six-grid cycle) of grid voltage vs (t) and charger current ic (t) during each operation mode. Fig. 15 shows the corresponding battery current and dc-link capacitor second harmonic rms ripple current Icap . Compared to charging only operation (mode 3), Icap is higher during mode 5 but lower during mode 1 and mode 4. It is highest in mode 2 as expected. Since reactive power is not limited, Icap increases above its value measured during charging only operation. The


5725

Fig. 14. Simulation result for active and reactive power outputs of the 3.3-kVA charger operation.

Fig. 15. Simulation result for the dc-link capacitor second harmonic ripple current and battery charging current for different operation modes.

increase in the ripple current requirement is similar to the calculation shown in Table VI. Fig. 15 also shows that battery charging current is not affected whether the charger supplies inductive or capacitive reactive power during the charging operation.

Fig. 16. Grid voltage v s (t) (1/10 scaled) and charger current ic (t), waveforms for each operation mode. (a) Mode#1-inductive operation. (b) Mode# 2-capacitive operation. (c) Mode#3-charging operation. (d) Mode#4-charging and inductive operation. (e) Mode#5-charging and capacitive operation.

VI. CONCLUSION PEVs are already in the market and their number is increasing each year. Each grid-connected vehicle is equipped with two important assets: a larger energy storage unit (battery) and an ac–dc conversion device. They can be controlled to support grid energy storage services. This paper focuses on the integration of the grid connected vehicles to the utility grid by explaining PEV chargers, charging terms and definitions, and charger topology selections based on different operation modes. It lists and compares available ac–dc topologies to be used for V2G reactive power support using on-board chargers. Furthermore,

it analyzes the effect of reactive power operation on the charger dc-link capacitor and defines the safe reactive power operation area for the dc-link capacitor. The study concludes that a single-phase charger needs to be limited in its reactive power output and cannot be used for symmetrical power operation for V2G applications unless the dc-link capacitor size is increased. The increase in the dc-link capacitor size depends on the selected topology, the input coupling inductor, and the power rating of the charger. Dual-buck and half-bridge converters require the most increase in dc-link

5726


capacitor for reactive power operation making them less suitable for V2G reactive power support. However, full-bridge converter requires the least increase in dc-link capacitor compared to dualbuck and half-bridge. Therefore, they are more suitable to be used for reactive power support to the grid. ACKNOWLEDGMENT This paper was authored by a contractor of the U.S. Government under Contract DE-AC05-00OR22725. Accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish from the contribution, or allow others to do so, for U.S. Government purposes. REFERENCES [1] (2011, Sep.). International energy outlook 2011. U.S. Dept. Energy, Tech. Rep. OE/EIA-0484 (2011), [Online]. Available: http://www.eia.gov/ forecasts/ieo/pdf/0484(2011).pdf [2] S. E. Letendre and W. Kempton, “The V2G concept: A new model for power?” Public Utilit. Fortnightly, pp. 16–26, Feb. 2002. [3] W. Kempton and S. E. Letendre, “Electric vehicles as a new power source for electric utilities,” Transpn. Res., vol. 2, no. 3, pp. 157–175, 1997. [4] M. Yilmaz and P. T. Krein, “Review of the impact of vehicle-to-grid technologies on distribution systems and utility interfaces,” IEEE Trans. Power Electron., to be published, DOI: 10.1109/TPEL.2012.2227500. [5] J. Kessen. (2012, Sep.). Vehicle-to-grid: An intriguing, but challenging, opportunity. [Online]. Available: http://info.a123systems.com/blog/bid/ 158802/Vehicle-to-Grid-An-Intrigui ng-but-Challenging-Opportunity [6] C. Zhou, K. Qian, and W. Zhou, “Modeling of the cost of EV battery wear due to V2G application in power systems,” IEEE Trans. Energy Convers., vol. 26, no. 4, pp. 1041–1050, Dec. 2011. [7] E. Sortomme and M. A. El-Sharkawi, “Optimal charging strategies for unidirectional vehicle-to-grid,” IEEE Trans. Smart Grid, vol. 2, no. 1, pp. 131–138, Mar. 2011. [8] (2005, Feb.). Principles for efficient and reliable reactive power supply and consumption. Federal Energy Regulatory Commission, Tech. Rep., [Online]. Available: http://www.ferc.gov/eventcalendar/files/ 20050310144430-02-04-05-reactiv e-power.pdf [9] F. Musavi, M. Edington, W. Eberle, and W. G. Dunford, “Energy efficiency in plug-in hybrid electric vehicle chargers: evaluation and comparison of front end ac-dc topologies,” in Proc. IEEE Energy Convers. Congr. Expo., Phoenix, AZ, USA, Sep. 2011, pp. 273–280. [10] A. Khaligh and S. Dusmez, “Comprehensive topological analysis of conductive and inductive charging solutions for plug-in electric vehicles,” IEEE Trans. Veh. Technol., vol. 61, no. 8, pp. 3475–3489, Oct. 2012. [11] D. C. Erb, O. C. Onar, and A. Khaligh, “Bi-directional charging topologies for plug-in hybrid electric vehicles,” in Proc. IEEE Appl. Power Electron. Conf. Expo., Palm Springs, CA, USA, Feb. 21–25, 2010, pp. 2066–2072. [12] M. Yilmaz and P. T. Krein, “Review of battery charger topologies, charging power levels and infrastructure for plug-in electric and hybrid vehicles,” IEEE Trans. Power Electron., vol. 28, no. 5, pp. 2151–2169, May 2013. [13] B. Singh, B. N. Singh, A. Chandra, K. Al-Haddad, A. Pandey, and D. P. Kothari, “A review of single-phase improved power quality ac-dc converters,” IEEE Trans. Ind. Electron., vol. 50, no. 5, pp. 962–981, Oct. 2003. [14] R. Wang, F. Wang, D. Boroyevich, R. Burgos, R. Lai, P. Ning, and K. Rajashekara, “A high power density single-phase PWM rectifier with active ripple energy storage,” IEEE Trans. Power Electron., vol. 26, no. 5, pp. 1430–1443, May 2011. [15] D. Dong, D. Boroyevich, R. Wang, and I. Cvetkovic, “A two-stage high power density single-phase ac-dc bi-directional PWM converter for renewable energy systems,” in Proc. IEEE Energy Convers. Congr. Expo., Atlanta, GA, USA, Sep. 12–16, 2010, pp. 3862–3869. [16] IEEE Recommended Practices and Requirements for Harmonic Control in Electrical Power Systems, IEEE Std. 519-1992, issued in 1993, updated in 2002 [17] IEEE Standard for Interconnecting Distributed Resources with Electric Power Systems, IEEE Std. 1547-2003, issued in 2003, reaffirmed in 2008.

[18] R. F. Nelson and M. A. Kepros, “Ac ripple effects on VRLA batteries in float applications,” in Proc. Battery Conf. Appl. Adv., Long Beach, CA, Jan. 1999, pp. 281–289. [19] (2012, Dec.). Featherweight lithium-ion battery. [Online]. Available: http: //www.cdtechno.com/pdf/lit/12 1093 0512.pdf [20] M. C. Kisacikoglu, A. Bedir, B. Ozpineci, and L. M. Tolbert, “PHEV-EV charger technology assessment with an emphasis on V2G operation,” Oak Ridge Nat. Lab., Oak Ridge, TN, USA, Tech. Rep. ORNL/TM-2010/221, Mar. 2012 [21] J. P. M. Figueiredo, F. L. Tofoli, and B. L. A. Silva, “A review of singlephase PFC topologies based on the boost converter,” in Proc. IEEE Int. Conf. Ind. Appl., Sao Paulo, Brazil, Nov. 2010, pp. 1–6. [22] B. A. Miwa, D. M. Otten, and M. F. Schlecht, “High efficiency power factor correction using interleaving techniques,” in Proc. IEEE Appl. Power Electron. Conf. Expo., Boston, MA, USA, Feb. 1992, pp. 557–568. [23] D. Gautam, F. Musavi, M. Edington, W. Eberle, and W. G. Dunford, “An automotive on-board 3.3 kW battery charger for PHEV application,” in Proc. IEEE Veh. Power Propul. Conf., Chicago, IL, USA, Sep. 2011, pp. 1–6. [24] D. S. Gautam, F. Musavi, M. Eddington, W. Eberle, and W. G. Dunford, “An automotive on-board 3.3-kW battery charger for PHEV application,” IEEE Trans. Veh. Technol., vol. 61, no. 8, pp. 3466–3474, Oct. 2012. [25] R. Martinez and P. N. Enjeti, “A high-performance single-phase rectifier with input power factor correction,” IEEE Trans. Power Electron., vol. 11, no. 2, pp. 311–317, Mar. 1996. [26] J.-W. Lim and B.-H. Kwon, “A power factor controller for single-phase PWM rectifiers,” IEEE Trans. Ind. Electron., vol. 46, no. 5, pp. 1035– 1037, Oct. 1999. [27] G. R. Stanley and K. M. Bradshaw, “Precision dc-to-ac power conversion by optimization of the output current waveform-the half bridge revisited,” IEEE Trans. Power Electron., vol. 14, no. 2, pp. 372–380, Mar. 1999. [28] H. Qian, J. Zhang, J.-S. Lai, and W. Yu, “A high-efficiency grid-tie battery energy storage system,” IEEE Trans. Power Electron., vol. 26, no. 3, pp. 886–896, Mar. 2011. [29] M. C. Kisacikoglu, B. Ozpineci, and L. M. Tolbert, “Examination of a PHEV bidirectional charger for V2G reactive power compensation,” in Proc. IEEE Appl. Power Electron. Conf. Expo., Palm Springs, CA, USA, Feb. 21–25, 2010, pp. 458–465.

Mithat C. Kisacikoglu (S’04) received the B.S. degree from Istanbul Technical University, Istanbul, Turkey, in 2005, and the M.S. degree from the University of South Alabama, Mobile, AL, USA, in 2007, both in electrical engineering. He is currently working toward the Ph.D. degree in power electronics major from the University of Tennessee, Knoxville, TN, USA. He has been a student researcher in the Power and Energy Systems Group and in the the Power Electronics and Electric Machinery Group at the Oak Ridge National Laboratory since 2008. His research interests include plug-in electric vehicles and renewable energy and energy storage system grid-interface circuit design and control.


Burak Ozpineci (S’92–M’02–SM’05) received the B.S. degree in electrical engineering from Orta Dogu Technical University, Ankara, Turkey, in 1994, and the M.S. and Ph.D. degrees in electrical engineering from the University of Tennessee, Knoxville, TN, USA, in 1998 and 2002, respectively. He joined the Post-Masters Program with the Power Electronics and Electric Machinery Research Center, Oak Ridge National Laboratory (ORNL), Knoxville, TN, USA, in 2001 and became a FullTime Research and Development Staff Member in 2002 and Group Leader of the Power and Energy Systems Group in 2008. He is currently leading the Power Electronics and Electric Machinery Group at ORNL and is also a Joint Faculty Associate Professor with The University of Tennessee, Knoxville. His research interests include system-level impact of SiC power devices, multilevel inverters, power converters for distributed energy resources and hybrid electric vehicles, and intelligent control applications for power converters. Dr. Ozpineci is the Vice Chair of the IAS Transportation Systems Committee, was the Chair of the IEEE PELS Rectifiers and Inverters Technical Committee, and was Transactions Review Chairman of the IEEE Industry Applications Society Industrial Power Converter Committee. He is also an Associate Editor of the IEEE TRANSACTIONS ON POWER ELECTRONICS and is on the Editorial Board of the IEEE PELS Digital Media Committee. He received the 2006 IEEE Industry Applications Society Outstanding Young Member Award, 2001 IEEE International Conference on Systems, Man, and Cybernetics Best Student Paper Award, and 2005 UT-Battelle (ORNL) Early Career Award for Engineering Accomplishment.

5727

Leon M. Tolbert (S’88–M’91–SM’98–F’13) received the B.E.E., M.S., and Ph.D. degrees in electrical engineering from the Georgia Institute of Technology, Atlanta, GA, USA, in 1989, 1991, and 1999, respectively. He was with Oak Ridge National Laboratory (ORNL) in 1991 and worked on several electrical distribution projects at the three U.S. Department of Energy plants in Oak Ridge, TN, USA. He joined the University of Tennessee, Knoxville, TN, USA, in 1999, where he is currently the Min H. Kao Professor and Head of the Department of Electrical Engineering and Computer Science. He is also a part time Senior Research Engineer at ORNL and conducts joint research at the National Transportation Research Center. His research interests include electric power conversion for distributed energy sources, motor drives, multilevel converters, hybrid electric vehicles, and application of SiC power electronics. Dr. Tolbert is a registered Professional Engineer in the state of Tennessee. He is a member of the following societies: Industry Applications, Industrial Electronics, Power and Energy, and Power Electronics. He was elected as a member-at-large to the IEEE Power Electronics Society Advisory Committee during 2010–2012, and he served as the Chair of the PELS Membership Committee during 2011–2012. He was an Associate Editor of the IEEE TRANSACTIONS ON POWER ELECTRONICS from 2007 to 2012 and of the IEEE POWER ELECTRONICS LETTERS from 2003 to 2006. He was the Chair of the Education Activities Committee of the IEEE Power Electronics Society from 2003 to 2007. He received the 2001 IEEE Industry Applications Society Outstanding Young Member Award, and he has three prize paper awards from the IEEE.