992

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 25, NO. 4, APRIL 2010

Design and Analysis of a Grid-Connected Photovoltaic Power System

Bo Yang, Wuhua Li, Member, IEEE, Yi Zhao, and Xiangning He, Fellow, IEEE
AbstractA grid-connected photovoltaic (PV) power system with high voltage gain is proposed, and the steady-state model analysis and the control strategy of the system are presented in this paper. For a typical PV array, the output voltage is relatively low, and a high voltage gain is obligatory to realize the grid-connected function. The proposed PV system employs a ZVT-interleaved boost converter with winding-coupled inductors and active-clamp circuits as the rst power-processing stage, which can boost a low voltage of the PV array up to a high dc-bus voltage. Accordingly, an accurate steady-state model is obtained and veried by the simulation and experimental results, and a full-bridge inverter with bidirectional power ow is used as the second power-processing stage, which can stabilize the dc-bus voltage and shape the output current. Two compensation units are added to perform in the system control loops to achieve the low total harmonic distortion and fast dynamic response of the output current. Furthermore, a simple maximum-power-point-tracking method based on power balance is applied in the PV system to reduce the system complexity and cost with a high performance. At last, a 2-kW prototype has been built and tested to verify the theoretical analysis of the paper. Index TermsBidirectional power ow control, compensation units, direct current control, maximum-power-point-tracking (MPPT) method, photovoltaic (PV) system, steady-state model.

I. INTRODUCTION

ODAY photovoltaic (PV) power systems are becoming more and more popular, with the increase of energy demand and the concern of environmental pollution around the world. Four different system congurations are widely developed in grid-connected PV power applications: the centralized inverter system, the string inverter system, the multistring inverter system and the module-integrated inverter system [1][4]. Generally three types of inverter systems except the centralized inverter system can be employed as small-scale distributed generation (DG) systems, such as residential power applications. The most important design constraint of the PV DG system is to obtain a high voltage gain. For a typical PV module, the open-circuit voltage is about 21 V and the maximum power point (MPP) voltage is about 16 V. And the utility grid voltage is 220 or 110 Vac. Therefore, the high voltage amplication is obligatory to realize the grid-connected function and achieve the low total harmonic distortion (THD). The conventional system

Manuscript received February 28, 2009; revised July 27, 2009 and August 29, 2009. Current version published April 9, 2010. This work was supported by the National Nature Science Foundations of China (50737002 and 50777055) and by the China Postdoctoral Science Foundation (200902625). Recommended for publication by Associate Editor R. Burgos. The authors are with the College of Electrical Engineering, Zhejiang University, Hangzhou 310027, China (e-mail: yangbo@zju.edu.cn; woohualee@ zju.edu.cn; diabloturen@zju.edu.cn; hxn@zju.edu.cn). Digital Object Identier 10.1109/TPEL.2009.2036432

requires large numbers of PV modules in series, and the normal PV array voltage is between 150 and 450 V, and the system power is more than 500 W. This system is not applicable to the module-integrated inverters, because the typical power rating of the module-integrated inverter system is below 500 W [3], [4], and the modules with power ratings between 100 and 200 W are also quite common [5]. The other method is to use a line frequency step-up transformer, and the normal PV array voltage is between 30 and 150 V [3], [4]. But the line frequency transformer has the disadvantages of larger size and weight. In the grid-connected PV system, power electronic inverters are needed to realize the power conversion, grid interconnection, and control optimization [6], [7]. Generally, gird-connected pulsewidth modulation (PWM) voltage source inverters (VSIs) are widely applied in PV systems, which have two functions at least because of the unique features of PV modules. First, the dc-bus voltage of the inverter should be stabilized to a specic value because the output voltage of the PV modules varies with temperature, irradiance, and the effect of maximum power-point tracking (MPPT). Second, the energy should be fed from the PV modules into the utility grid by inverting the dc current into a sinusoidal waveform synchronized with utility grid. Therefore, it is clear that for the inverter-based PV system, the conversion power quality including the low THD, high power factor, and fast dynamic response, largely depends on the control strategy adopted by the grid-connected inverters. In this paper, a grid-connected PV power system with high voltage gain is proposed. The steady-state model analysis and the control strategy of the system are presented. The gridconnected PV system includes two power-processing stages: a high step-up ZVT-interleaved boost converter for boosting a low voltage of PV array up to the high dc-bus voltage, which is not less than grid voltage level; and a full-bridge inverter for inverting the dc current into a sinusoidal waveform synchronized with the utility grid. Furthermore, the dcdc converter is responsible for the MPPT and the dcac inverter has the capability of stabilizing the dc-bus voltage to a specic value. The grid-connected PV power system can offer a high voltage gain and guarantee the used PV array voltage is less than 50 V, while the power system interfaces the utility grid. On the one hand, the required quantity of PV modules in series is greatly reduced. And the system power can be controlled in a wide range from several hundred to thousand watts only by changing the quantity of PV module branches in parallel. Therefore, the proposed system can not only be applied to the string or multistring inverter system, but also to the module-integrated inverter system in low power applications. On the other hand, the nonisolation PV systems employing neutral-point-clamped (NPC)

0885-8993/$26.00 2010 IEEE

YANG et al.: DESIGN AND ANALYSIS OF A GRID-CONNECTED PHOTOVOLTAIC POWER SYSTEM

993

Fig. 1.

Proposed grid-connected PV power system.

Fig. 2. High step-up ZVT-interleaved boost converter and its equivalent circuit. (a) ZVT-interleaved boost converter. (b) Equivalent circuit.

topology, highly efcient reliable inverter concept (HERIC) topology, H5 topology, etc. [8][14], have been widely used especially in Europe. Although the transformerless system having a oating and nonearth-connected PV dc bus requires more protection [10], [15], [16], it has several advantages such as high efciency, lightweight, etc. Therefore, the nonisolation scheme in this paper is quite applicable by employing the high step-up ZVT-interleaved boost converter, because high voltage gain of the converter ensures that the PV array voltage is below 50 V and benets the personal safety even if in high-power application. Fig. 1 shows the proposed grid-connected PV power system. II. STEADY-STATE MODEL OF HIGH STEP-UP ZVT-INTERLEAVED BOOST CONVERTER Fig. 2 shows the ZVT-interleaved boost converter with winding-coupled inductors and active-clamp circuits, which is proposed by our research team [17]. The winding-coupled inductors offer the voltage-gain extension [18][20]. The activeclamp circuits realize the ZVT commutation of the main

switches and the auxiliary switches [17]. As shown in Fig. 2(a), S1 and S2 are the main switches; Sc1 and Sc2 are the activeclamp switches; Do1 and Do2 are the output diodes. The coupling method of the winding-coupled inductors is marked by open circles and asterisks. Each coupled inductor is modeled as the combination of a magnetizing inductor, an ideal transformer with corresponding turns ratio and a leakage inductor in series with the magnetizing inductor. The equivalent circuit model is demonstrated in Fig. 2(b), where Lm 1 and Lm 2 are the magnetizing inductors; Llk 1 and Llk 2 are the leakage inductors including the reected leakage inductors of the second and third windings of the coupled inductors; Cs1 and Cs2 are the parallel capacitors, including the parasitic capacitors of the switches; Cc1 and Cc2 are the clamp capacitors; N is the turns ratio n2 /n1 . The operation principle analysis and the steady-state waveforms of the high step-up ZVT-interleaved boost converter have been discussed in [17]. Compared with the proposed converter, the full-bridge dcdc converter is also employed commonly as a similar rst stage in the PV system. However, for the high stepup gain applications, the large current ripples of the primary-side switches increase the conduction losses, and the secondary-side diodes need to sustain a high voltage stress. Moreover, as a bucktype converter, a large turns ratio of the transformer is necessary to obtain a high step-up gain, which induces a large leakage inductance and large commutation energy on the primary-side switches. Therefore, the design of the transformer is difcult and the converters efciency is impacted. Furthermore, the resonant-mode converters such as LLC, LCC, and higher order element converters are studied and developed, which are attractive for potential higher efciency and higher power density than PWM counterparts. However, most of resonant converters include some inherent problems, such as electromagnetic interference (EMI) problems due to variable frequency operation and reduced conversion efciency due to circulating energy generation. Moreover, to make practical use of the resonant converters, the required precise control waveform and difcult overcurrent protection increase the design complexity of the whole system [21][24]. Correspondingly the ZVT-interleaved boost converter has the following three main advantages. 1) Voltage gain is extended greatly by using a proper turns ratio design. As the turns ratio increases, the voltage gain increases without the extreme duty ratio, which can reduces the input and output current ripples. Omitting the effect of the leakage inductance and applying the voltagesecond balance to the magnetizing inductor, the voltage gain is given by M= N +1 Vout . = Vin 1D (1)

2) Voltage stress of the main switches is reduced, as the turns ratio increases. Therefore, the low-voltage and highperformance devices can be used to reduce the switching and conduction losses. And the voltage spikes are clamped effectively and the leakage energy is recovered. If the clamp capacitance is assumed large enough and the voltage ripple on the switches can be ignored when they turn

994

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 25, NO. 4, APRIL 2010

off, the normalized voltage stress of the main switches is given by Vds = Vout . N +1 (2)

3) ZVT soft switching is achieved for both main switches and auxiliary switches during the whole switching transition, which means the switching losses are reduced greatly. Diode reverse-recovery loss is greatly reduced because the di/dt of the diode current is controlled by the inherent leakage inductor of a coupled boost inductor. Unfortunately, the leakage inductor of the winding-coupled inductors has great effect on the voltage gain expression and a big error is found in the steady-state model based on (1), especially when the leakage inductance increases to a certain level, which brings difculty to the design of circuit parameters. To derive a more accurate steady-state model of the converter, the leakage inductance of the winding-coupled inductors should be considered, since the leakage inductance strongly inuences the operation states of the circuit. To simplify the calculation, the following conditions are assumed in reason. 1) The clamp capacitance is large enough, so the voltage ripple on the main switches can be ignored and the voltage Vds is taken as a constant when they turn off. 2) The magnetizing inductance is much larger than the leakage inductance, so the magnetizing current IL m is taken as a constant in one switching period. 3) The dead times of the main switches and the corresponding auxiliary switches are ignored. 4) The two interleaved and intercoupled boost converter cells are provided with a strict symmetry. Based on the previous assumptions, the partial key waveforms of this converter are shown in Fig. 3, which have reached a steady state. And the following approximations are given: If Ir = I VL k 1 VL k 2 = VL k . (3) (4)

Fig. 3.

Partial key waveforms of the converter.

As shown in Fig. 2, the voltages on the winding-coupled inductors are decided by Vn1 = VL m 1 = Vds 1 VL k 1 Vn2 =N Vn1 Vn2 = N (Vin VL k 2 ) (8) (9) (10)

The equation of the output voltage is always true by the Kirchhoff voltage law Vout = Vds 1 + Vn2 + Vn2 (5)

where Vn1 represents the voltage of the rst winding L1 a . Therefore, substituting (6), (9), and (10) into (5), the equation of the output voltage in stage a is obtained Vout = (N + 1) Vds 1 2 N VL k = N +1 I Vin 2 N Llk 1 . (11) 1D (1 D)/fs

where Vn2 and Vn2 , respectively, represent the voltage of the second winding L1 b and the voltage of the third winding L2 c . A. Stage a (Main Switches S1 is OFF and S2 is ON) Based on the voltage-second balance to the magnetizing inductor, the switching voltage of S1 is given by Vds 1 = Vin . 1D (6)

B. Stage b (Main Switches S1 is ON and S2 is ON) Likewise, from the waveforms shown in Fig. 3, it can be found that Vds 1 = 0 VL k = VL k 1 = Llk 1 Ir I = Llk 1 . t2 t2 (12) (13)

From the waveform of iL k 1 shown in Fig. 3, it can be found that VL k = VL k 1 = Llk 1 If I = Llk 1 . t1 (1 D)/fs (7)

Considering the polarity of the voltages on the windingcoupled inductors in stage b, the voltage expressions for the

YANG et al.: DESIGN AND ANALYSIS OF A GRID-CONNECTED PHOTOVOLTAIC POWER SYSTEM

995

winding-coupled inductors are also obtained by Vn1 = VL m 1 = VL k 1 Vin Vn2 = N Vn1 Vn2 = N (VL k 2 + Vin ). (14) (15) (16)

TABLE I STEADY-STATE MODEL VERIFICATION

Therefore, substituting (12), (15), and (16) into (5), the equation of the output voltage in stage b is obtained Vout = 2 N VL k = 2 N Llk 1 I . t2 (17)

In addition, from the waveform of iD o1 shown in Fig. 3, the total charge through the two output diodes in one switching period is decided by I . (18) N Meanwhile, the charge through the load in one switching period is Q1 = 2QD o1 = (t1 + t2 ) Q2 = 1 Vout . R fs Vout 1 I = . N R fs (19)

Therefore, the charge-conservation equation can be found that (t1 + t2 ) (20)

Therefore, the (11), (17), and (20) can be solved to obtain the expression for the steady-state model of the converter. M= Vout Vin [(1 D)R]2 + 8N 2 fs Llk R (1 D)R 4N 2 fs Llk (21) where Llk is the equivalent leakage inductance of the windingcoupled inductors, and Llk = Llk 1 = Llk 2 , and R is the equivalent load of the converter. As shown in Table I, the results calculated by the two steadystate models and the simulation software PSIM, are compared to verify the proposed model. From the data in Table I, the maximum error of the output voltage between the accurate model based on (21) and the simulation results is only 0.6%, and the corresponding maximum error of the model based on (1) reaches up to 43.1%. It is clear that the proposed steady-state model approaches the simulation results closely. III. CONTROL STRATEGY OF FULL-BRIDGE INVERTER WITH BIDIRECTIONAL POWER FLOW The full-bridge inverter shown in Fig. 1 works as a voltage-source PWM (VS-PWM) converter in this paper, which implements the bidirectional power ow. By using a directcurrent-control strategy, the VS-PWM converter can force the instantaneous load current to accurately follow the sinusoidal reference, which synchronizes with the utility grid voltage. And the high power factor, the low THD and the fast dynamic response are achieved. Furthermore, the bidirectional ow of
Fig. 4. Control block of two-stage grid-connected PV system.

= (N + 1)

Fig. 5.

Control block of full-bridge inverter with bidirectional power ow.

power facilitates the compensation of the dc-bus and the ac-side voltage variations, which helps to stabilize the dc-bus voltage in startup and cloudy situations and improves the stability of overall system. Fig. 4 shows the control block of the two-stage grid-connected PV system. Fig. 5 shows the control block of the full-bridge inverter with bidirectional power ow.

996

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 25, NO. 4, APRIL 2010

A. Control of the Bidirectional Power Flow As shown in Fig. 5, the dc-bus voltage Udc is always controlled to keep a constant value Uref with zero error by the voltage-feedback control loop. Meanwhile, the direction and magnitude of VS-PWM converters output current and power are decided dynamically by the value of Ue , which is the output of negative PI regulator in the voltage loop. If Udc > Uref then Ue is increasing, and the VS-PWM converter works as an inverter which transfers the PV array power to the utility grid. The energy generation of PV power system is positively correlated with the magnitude of Ue . If Udc < Uref then Ue is decreasing, and when Ue < 0, the VS-PWM converter works as a PWM rectier, which draws the energy from the utility grid to the capacitor of dc bus, maintaining the stability of dc-bus voltage. The current in the negative direction nally approaches a quite small value, which is only used to compensate for the switch losses of VS-PWM converter. The earlier characteristic guarantees that the dc-bus voltage is stabilized to a required value by the back-end VS-PWM converter, whether the front-end dcdc converter works or not. Therefore, it is avoided that the ZVT-interleaved Boost converter works in an open-circuit state, on the condition that the back end is started up prior to the front end. B. Direct Current Control With Compensation Units As shown in Fig. 5, the load currents iout is detected and compared with the reference current iref , and the error signal is processed by a PI regulator in the current-feedback control loop. Main advantages of this direct current control are the low harmonics to reduce losses in steady state, the fast response to provide high dynamic performances, and the peakcurrent protection to reject overload [25], [26]. Generally, the current control loop is designed to have a bandwidth of 25 kHz, higher than the voltage loop bandwidth of 200500 Hz, to assure the stability of the proposed inverter control with two PI regulators. However, the instantaneous power and the dc-bus voltage include a ripple component with the frequency 2 in the case of a single-phase inverter. Furthermore, the grid voltage is not an ideal sinusoidal waveform in practice. Therefore, it is hard to achieve low THD of the output current by using the simple direct-current-control strategy in the real grid condition. Accordingly, two compensation units are added to the current control loop as the feedforward control units. The feedforward control has little impact on the systems zeros and poles conguration, but achieves to track the sinusoidal reference accurately and restrain the harmonics distortion of the load current, especially at the current peak. Compensation coefcient Kd directly processes the magnitude of reference currents iref , and can counteract the main inuences of the dc-bus voltage ripple because Kd represents a negative uctuating feature with the frequency 2 , compared with the dc-bus voltage ripple. Kd is dened as Kd = Uref . Udc (22)

TABLE II P&O ALGORITHM EMPLOYING U e

The PI regulator in the grid-voltage-feedforward control multiplies the real, defective grid voltage with a proportion gain Kf . Its output uc and the output ua of PI regulator in currentfeedback loop are together fed to the PWM modulator to produce the drive signals for the inverter switches. Therefore, the modulation wave uout includes the defective component of grid voltage to compensate grid voltage uctuation and obtain highly sinusoidal current waveform. The feedforward effect depends on the value of Kf . In addition, the frequency and phase of current reference iref are calculated to synchronize with utility grid voltage ugrid in real time by a phase-locked-loop (PLL) system, which is achieved via the digital algorithm of DSP chip in this paper. IV. SIMPLE MPPT SOLUTION BASED ON POWER BALANCE Many MPPT solutions are developed to ensure the optimal utilization of PV modules [27][29]. The implementations generally involve sensing the output current and voltage of PV modules, and the MPPT algorithms use the information to maximize power drawn from the PV modules. Unfortunately, such realizations are costly and complex [27], [28]. In this paper a simple MPPT solution is adopted in view of the power balance. If omitting the whole system losses, the generated power of the PV array is equal to the output power that is transferred to the utility grid. Therefore, the rms value of the load currents iout is in proportion to the PV array power, as the grid voltage is clamped to 220 or 110 V. And as discussed earlier, the magnitude of the load current iout directly depends on the output Ue of the negative PI controller in the voltage loop. Therefore, the majority of MPPT algorithms can be implemented by controlling the value Ue rather than the calculated power value by multiplying the inputs from voltage and current sensors, which holds universality. To verify the previous analysis, as shown in Table II, the perturb and observe (P&O) algorithm of employing the value Ue is applied in this paper. And perturbing the duty ratio D of the ZVT-interleaved boost converter perturbs the PV array current and voltage, and consequently, perturbs the PV array power. The operating point is then adjusted to maximize PV array power. The other advantage is the simple MPPT solution ensures the maximization of the power transferred to the utility grid, but not the output power from PV array. Although the two are equivalent in theory, the former is a real MPPT technique in practice that converts maximum energy to utility grid, because some

YANG et al.: DESIGN AND ANALYSIS OF A GRID-CONNECTED PHOTOVOLTAIC POWER SYSTEM

997

complications, like nonlinearity of the whole system losses and long-term uctuation of the utility grid voltage, are eliminated. V. EXPERIMENTAL RESULTS A. System Conguration To conrm the theoretical analysis in the previous sections, a 2-kW prototype of the proposed grid-connected PV power system was built. Two ZVT-interleaved boost converters of 1 kW are connected in parallel via a dc bus through a central inverter of 2 kW to the grid. The lower power dcdc converters are connected respectively to the individual PV arrays, and the central inverter can expand the power rate and reduce the system cost. The detailed components and parameters used are as follows. ZVT-interleaved boost converter: Vin : 3850 V; Vout : 380 V. fs : 50 kHz dead time: td 1 = td 2 = 250 ns. N = n2 :n1 = 36:18 = 2. S1 and S2 : FQA62N25C. Sc1 and Sc2 : FQA59N25C. Do1 and Do2 : RHRP15120. Cc1 and Cc2 : 2.2 F. Cs1 and Cs2 : 2.2 nF. Lm 1 and Lm 2 : 150 H. Llk 1 and Llk 2 : 4 H. Full-bridge inverter: fs : 20 kHz; dead time: 1 s. Insulated-gate bipolar transistor (IGBT): FGAF40N60. Cdc : 470 F 4. Lf 1 and Lf 2 : 1.2 mH. Digital controller: TMS320F2808. B. Experimental Results The following experimental results are given at 1 kW power generation of each PV array under a specic temperature and irradiance condition, and the total power of the grid-connected PV system can reaches 2 kW. Fig. 6 shows the experimental results of the ZVT-interleaved boost converter at 1 kW when the input voltage is 40 V. As shown in Fig. 6(a), the extreme duty ratio is avoided when the voltage gain is extended. Meanwhile, the voltage stress of the main switches is reduced to 170 V, far lower than the output voltage 380 V, and the low-voltage and high-performance devices can be used to reduce the conduction losses. The waveforms of ZVT soft switching of the main switch S1 and clamp switch Sc1 are, respectively, shown in Fig. 6(b) and (c). It is clear that the ZVT of the main switches and the auxiliary switches are achieved during the whole switching transition, which reduces the switching losses, improves the efciency and increases the power density. Fig. 6(d) shows the experimental waveform of the turn-OFF current of the output diode Do1 . It is clear that the reverse-recovery current is reduced to a small value, and the reverse-recovery problem is alleviated dramatically by the leakage inductor. The EMI noise is suppressed signicantly, and the losses caused by the reverse recovery are reduced greatly.

Fig. 6. Experimental results about ZVT-interleaved boost converter. (a) Switching voltages of main switch S 1 , S 2 . (b) ZVT operation for main switch S 1 . (c) ZVT operation for auxiliary switch S c 1 . (d) Turn-OFF current of output diode D o 1 .

Furthermore, to verify the proposed steady-state model of the ZVT-interleaved boost converter, the experimental results of the output voltage are compared in Table III. It is clear that the experimental error values are within 2.5% of the values calculated by the accurate model and the simulation software. The errors are induced by the assumptions in modeling.

998

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 25, NO. 4, APRIL 2010

TABLE III EXPERIMENTAL VERIFICATION OF STEADY-STATE MODEL

Fig. 8.

Harmonic spectrum of the output current at 2 kW.

Fig. 7. Experimental results about full-bridge inverter with bidirectional power ow. (a) Output current and dc-bus voltage waveforms in the rectifying condition. (b) Output current and dc-bus voltage waveforms in the inverting condition.

Fig. 7 shows the experimental results of the full-bridge inverter with bidirectional power ow. The output current and the dc-bus voltage waveforms in the rectifying condition are shown in Fig. 7(a). It is clear that the current is close to zero and the dc-bus voltage is stabilized to 380 V without a very small ripple, because no active power is consumed except compensating for the IGBTs losses. The output current and the dc-bus voltage waveforms at 2 kW in the inverting condition are shown in Fig. 7(b). It can be clearly seen that the output current is highly sinusoidal synchronized with the grid voltage, which is hardly deteriorated by the ripple component with the frequency 2 of the dc-bus voltage due to the two compensation units. Fig. 8 shows the harmonic spectrum of the output current at 2 kW, and the measured THD counted till to 50th is 3.384%. The experimental results of the dynamic response for the system under different step source changes are given in Fig. 9. As shown in Fig. 9(a), the power source is changed from 500

Fig. 9. Experimental results of the system dynamic response under different step source changes. (a) Dynamic response under power source change from 500 to 1600 W. (b) Dynamic response under power source change from 1600 to 500 W.

to 1600 W, and reversely, the power source is changed from 1600 to 500 W, as shown in Fig. 9(b). It can be clearly seen that the fast and effective responses to the abrupt source changes are achieved due to the direct-current-control strategy with two compensation units. Fig. 10 shows the MPPT effect of P&O algorithm by employing the value Ue , when the MPP is changed from 500 to 800 W. The output current waveform in the MPPT process is shown in Fig. 10(a), and the responses of output current rms and output power rms are shown in Fig. 10(b). It is clear that the output current and power gradually increase during the MPPT period and nally reach the steady stage. Therefore, this MPPT method based on power balance is practical with a good performance. Fig. 11 and Table IV show the detailed efciency data of the whole system including the maximum efciency and European

YANG et al.: DESIGN AND ANALYSIS OF A GRID-CONNECTED PHOTOVOLTAIC POWER SYSTEM

999

VI. CONCLUSION This paper presented a grid-connected PV power system with high voltage gain. The proposed PV system employs a high step-up ZVT-interleaved boost converter with winding-coupled inductors and active-clamp circuits as the rst power-processing stage, and high voltage gain is obtained by the turns ratio selection of winding-coupled inductors. An accurate steady-state model of the converter is obtained and veried by the simulation and experimental results. A full-bridge inverter with bidirectional power ow is used as the second power-processing stage, to stabilize the dc-bus voltage and shape the output current. Two compensation units are added to the system control loops, and the low current THD and the high dynamic performance are achieved. Furthermore, a simple MPPT method based on power balance is applied in the PV system and represents a good performance. A 2-kW prototype is built, and experimental results conrm the validity and applicability of the proposed PV system. REFERENCES
[1] S. B. Kjaer, J. K. Pedersen, and F. Blaabjerg, A review of single-phase grid-connected inverters for photovoltaic modules, IEEE Trans. Ind. Appl., vol. 41, no. 5, pp. 12921306, Sep./Oct. 2005. [2] Q. Li and P. Wolfs, A review of the single phase photovoltaic module integrated converter topologies with three different DC link congurations, IEEE Trans. Power Electron., vol. 23, no. 3, pp. 13201333, May 2008. [3] M. Calais, J. Myrzik, T. Spooner, and V. G. Agelidis, Inverters for singlephase grid connected photovoltaic systemsAn overview, in Proc. IEEE PESC, 2002, vol. 4, pp. 19952000. [4] S. B. Kjr, J. K. Pedersen, and F. Blaabjerg, Power inverter topologies for photovoltaic modulesA review, in Proc. IEEE IAS Conf., 2002, pp. 782788. [5] Q. Li and P. Wolfs, A current fed two-inductor boost converter with an integrated magnetic structure and passive lossless snubbers for photovoltaic module integrated converter applications, IEEE Trans. Power Electron., vol. 22, no. 1, pp. 309321, Jan. 2007. [6] M. P. Kazmierkowski and L. Malesani, Current control techniques for three-phase Voltage-Source PWM converters: A survey, IEEE Trans. Ind. Electron., vol. 45, no. 5, pp. 691703, Oct. 1998. [7] H. M. Kojabadi, B. Yu, I. A. Gadoura, L. Chang, and M. Ghribi, A novel DSP-based current-controlled PWM strategy for single phase grid connected inverters, IEEE Trans. Power Electron., vol. 21, no. 4, pp. 985 993, Jul. 2006. [8] R. Gonzalez, J. Lopez, P. Sanchis, and L. Marroyo, Transformerless inverter for single-phase photovoltaic systems, IEEE Trans. Power Electron., vol. 22, no. 2, pp. 693697, Mar. 2007. [9] T. Kerekes, R. Teodorescu, and U. Borup, Transformerless photovoltaic inverters connected to the grid, in Proc. IEEE APEC Conf., 2007, pp. 17331737. [10] O. Lopez, R. Teodorescu, F. Freijedo, and J. Doval-Gandoy, Eliminating ground current in a transformerless photovoltaic applictation, in Proc. IEEE PES Conf., 2007, pp. 15. [11] R. Gonzalez, E. Gubia, J. Lopez, and L. Marroyo, Transformerless singlephase multilevel-based photovoltaic inverter, IEEE Trans. Ind. Electron., vol. 55, no. 7, pp. 26942702, Jul. 2008. [12] German Patent, HERIC-topology, Patent DE 10 221 592 A1, Dec. 4, 2003. [13] German Patent, H5-topology, Patent DE 102 004 030 912 B3, Jan. 19, 2006. [14] International Patent, Circuit arrangement having a twin inductor for converting a DC voltage into an AC voltage or an alternating current, Patent WO/2007/077031, Jul. 12, 2007. [15] 2005 National Electrical Code, National Fire Protection Association, Quincy, MA, 2005. [16] H. Haeberlin, Photovoltaik. Berlin, Germany: VDE Publishing House, 606 pp. (in German). ISBN: 978-3-8007-3003-2. [17] W. Li and X. He, ZVT interleaved boost converters for high-efciency, high step-up DC/DC conversion, IET Electr. Power Appl., vol. 1, no. 2, pp. 284290, Mar. 2007.

Fig. 10. MPPT effect of P&O algorithm by employing value U e . (a) Output current waveform about MPPT effect. (b) Responses of output current and power about MPPT effect.

Fig. 11.

Measured efciency of the whole system.

TABLE IV MEASURED EFFICIENCY OF THE WHOLE SYSTEM

efciency, which were measured by the power analyzer PZ4000 from Yokogawa, respectively, when the input voltages were selected as 40 and 48 V. The maximum efciency of 93.9% and European efciency of 92.7% are achieved when the input voltage is 48 V.

1000

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 25, NO. 4, APRIL 2010

[18] Q. Zhao and F. C. Lee, High-efciency, high step-up DCDC converters, IEEE Trans. Power Electron., vol. 18, no. 1, pp. 6573, Jan. 2003. [19] R. J. Wai and R. Y. Duan, High step-up converter with coupled-inductor, IEEE Trans. Power Electron., vol. 20, no. 5, pp. 10251035, Sep. 2005. [20] K. C. Tseng and T. J. Liang, Novel high-efciency step-up converter, IEE Electr. Power Appl., vol. 151, no. 2, pp. 182190, Mar. 2004. [21] F. C. Lee, S. Wang, P. Kong, C. Wang, and D. Fu, Power architecture design with improved system efciency, EMI and power density, in Proc. IEEE PESC Conf., 2008, pp. 41314137. [22] A. Bellini, S. Bifaretti, and V. Iacovone, Resonant DCDC converters for photovoltaic energy generation systems, in Proc. IEEE SPEEDAM Conf., 2008, pp. 815820. [23] D. Fu, F. C. Lee, Y. Liu, and M. Xu, Novel multi-element resonant converters for front-end dc/dc converters, in Proc. IEEE PESC Conf., 2008, pp. 250256. [24] D. Fu, F. C. Lee, Y. Qiu, and F. Wang, A novel high-power-density three level LCC resonant converter with constant-power-factor-control for charging applications, IEEE Trans. Power Electron., vol. 23, no. 5, pp. 24112420, Sep. 2008. [25] M. P. Kazmierkowski and L. Malesani, Current control techniques for three-phase Voltage-Source PWM Converters: A survey, IEEE Trans. Ind. Electron., vol. 45, no. 5, pp. 691703, Oct. 1998. [26] M. Cichowias and M. P. Kamierkowski, Comparison of current control techniques for PWM rectiers, in Proc. IEEE Int. Symp., 2002, pp. 1259 1263. [27] T. Esram and P. L. Chapman, Comparison of photovoltaic array maximum power point tracking techniques, IEEE Trans. Energy Convers., vol. 22, no. 2, pp. 439449, Jun. 2007. [28] A. Pandey, N. Dasgupta, and A. K. Mukeriee, A simple single-sensor MPPT solution, IEEE Trans. Power Electron., vol. 22, no. 2, pp. 698 700, Mar. 2007. [29] S. Jain and V. Agarwal, A single-stage grid connected inverter topology for solar PV systems with maximum power point tracking, IEEE Trans. Power Electron., vol. 22, no. 5, pp. 19281940, Sep. 2007.

Yi Zhao was born in Liaoning, China, in 1983. He received the B.Sc. degree in the College of Electrical and Electronic Engineering, Huazhong University of Science and Technology, Wuhan, China, in 2006. He is currently working toward the Ph.D. degree with the College of Electrical Engineering, Zhejiang University, Hangzhou, China. His research interests include dc/dc converters and photovoltaic power system.

Bo Yang was born in Sichuan, China, in 1981. He received the B.Sc. degree in applied power electronics in 2004 from Zhejiang University, Hangzhou, China, where he is currently working the Ph.D. degree in electrical engineering. His current research interests include photovoltaic power system, power converter modeling, and digital control techniques.

Xiangning He (M95SM96F10) received the B.Sc. and M.Sc. degrees from Nanjing University of Aeronautical and Astronautical, Nanjing, China, in 1982 and 1985, respectively, and the Ph.D. degree from Zhejiang University, Hangzhou, China, in 1989. From 1985 to 1986, he was an Assistant Engineer at the 608 Institute of Aeronautical Industrial General Company, Zhuzhou, China. From 1989 to 1991, he was a Lecturer at Zhejiang University. In 1991, he obtained a Fellowship from the Royal Society, U.K., and conducted research in the Department of Computing and Electrical Engineering, Heriot-Watt University, Edinburgh, U.K., as a Post-Doctoral Research Fellow for two years. In 1994, he joined Zhejiang University as an Associate Professor, where he has been a Full Professor with the College of Electrical Engineering, Zhejiang University, since 1996. He was the Director of the Power Electronics Research Institute and the Head of the Department of Applied Electronics. He is currently the Vice Dean of the College of Electrical Engineering, Zhejiang University. His research interests are power electronics and their industrial applications. He is the author or coauthor of more than 200 papers and one book Theory and Applications of Multi-level Converters. He holds 12 patents. Dr. He received the 1989 Excellent Ph.D. Graduate Award, the 1995 Elite Prize Excellence Award, the 1996 Outstanding Young Staff Member Award, and the 2006 Excellent Staff Award from Zhejiang University for his teaching and research contributions. He received ve Scientic and Technological Progress Awards from Zhejiang Provincial Government and the State Educational Ministry of China in 1998, 2002, and 2009, respectively, and ve Excellent Paper Awards. He is a Fellow of the Institution of Engineering and Technology (formerly IEE), U.K.

Wuhua Li (M09) received the B.Sc. and Ph.D. degrees in applied power electronics and electrical engineering from Zhejiang University, Hangzhou, China, in 2002 and 2008, respectively. From September 2004 to March 2005, he was an Intern, and from January 2007 to June 2008, a Research Assistant in GE Global Research Center, Shanghai, China. In July 2008, he joined the College of Electrical Engineering, Zhejiang University, where he is currently a Postdoctoral Fellow. His research interests include high frequency dc/dc converters, soft-switching techniques, RF power ampliers, converter modeling, and digital control.