pv inverter

1764 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 28, NO. 4, APRIL 2013

Design, Analysis, and Implementation of Solar PowerOptimizer for DC Distribution System

Shih-Ming (Orion) Chen, Student Member, IEEE, Tsorng-Juu (Peter) Liang, Member, IEEE, and Ke-Ren Hu

Abstract—This paper proposes a high step-up solar power op-timizer (SPO) that efficiently harvests maximum energy from aphotovoltaic (PV) panel then outputs energy to a dc-microgrid.Its structure integrates coupled inductor and switched capacitortechnologies to realize high step-up voltage gain. The leakage in-ductance energy of the coupled inductor can be recycled to re-duce voltage stress and power losses. A low voltage rating andlow-conduction resistance switch improves system efficiency byemploying the incremental conductance method for the maximumpower point tracking (MPPT) algorithm. Because of its high track-ing accuracy, the method is widely used in the energy harvesting ofPV systems. laboratory prototypes of the proposed SPO that havean input voltage range of 20 to 40 V and a maximum PV outputpower of 400 V/300 W are applied. The highest PV power conver-sion efficiency is 96.7%. The maximum MPPT accuracy is 99.9%,and the full load average MPPT accuracy is 97.8%.

Index Terms—High step-up voltage gain, maximum power pointtracking (MPPT), solar power optimizer (SPO).

I. INTRODUCTION

FOSSIL fuels continue to be depleted, and their use has beeninstrumental to climate change, a problem that grows more

severe each year. A photovoltaic (PV) power generation system,which uses a renewable resource, has been extensively used inemergency facilities and in generating electricity for mass use.A conventional PV generation system is either a single- or amultistring PV array that is connected to one or several cen-tral PV inverters. Numerous series-connected PV modules areconnected in the PV array to achieve the DC link voltage thatis high enough to be connected to electricity through the DC-AC inverter. However, the power reduction that is caused bythe shadow effect is an inevitable problem in a centralized PVsystem. The use of a microinverter or ac module has recentlybeen proposed for individual PV panels [1], [2]. Although thisdiscrete PV power generation solution may partially eliminatethe shadow problem, a microinverter structure constrains the

Manuscript received March 31, 2012; revised June 27, 2012; acceptedAugust 6, 2012. Date of current version October 26, 2012. This work wassupported in part by the financial support from the Bureau of Energy, the Min-istry of Economic Affairs, Taiwan, under Project 101-D0204-2, and in part bythe National Science Council of Taiwan under Project NSC 101-3313-E-006-017 and Project NSC 101-2218-E-006-014. Recommended for publication byAssociate Editor T. Shimizu.

S.-M. (Orion) Chen and T.-J. (Peter) Liang are with the Department ofElectrical Engineering, Advanced Optoelectronic Technology Center (AOTC),Green Energy Electronics Research Center (GREERC), National Cheng-Kung University, Tainan 701, Taiwan (e-mail: [email protected];[email protected]).

K.-R. Hu is with the Richtek Technology Company, Hsinchu 302, Taiwan(e-mail: [email protected]).

Digital Object Identifier 10.1109/TPEL.2012.2213270

Fig. 1. Configuration of multiple parallel SPO for a dc-microgrid system.

system energy’s harvesting efficiency and entails high costs. Asolar power optimizer (SPO) was developed as an alternative tomaximize energy harvest from each individual PV module. AnSPO is used as a dc–dc converter with maximum power pointtracking (MPPT), which increases PV panel voltage to opti-mum voltage levels for a dc microgrid connection or through adc–ac inverter for electricity [3]–[6]. Fig. 1 shows a single PVpanel’s energy, which passes through an SPO to a dc micro-grid system. A 400 V dc-microgrid system was proposed as anenergy-efficient distribution option for data center systems andtelecommunication facilities [7]. The SPO attempts to improvethe use of distributed renewable resources and lower systemcost. It may also potentially improve the efficiency of PV sys-tems, has an antishadow effect, and can monitor the status ofPV modules [8]. Moreover, the dc-grid voltage is regulated bybidirectional inverter and battery tank. In case of low-loadingcondition, the redundant energy will store into battery or throughbidirectional inverter to ac grid.

The maximum power point (MPP) voltage range of a singlePV panel ranges from 15 to 40 V and has a power capac-ity of about 100 to 300 W [9]. An SPO has a high step-upconverter that increases low-input voltage to a sufficient volt-age level. Various step-up dc–dc converter topologies include aconventional boost and flyback converters [10], [11], switched-inductor converter, and switched capacitor converter [12]–[16],as well as a transformerless switched capacitor types [17], [18],voltage-lift types [19]–[21], capacitor–diode voltage multipli-ers [22]–[25], and boost types that are integrated with coupledinductors [26]–[29]. With increasing voltage gain, recycling theleakage inductance energy of a coupled inductor will reduce thevoltage stress on the active switch, which enables the coupled

0885-8993/$31.00 © 2012 IEEE

CHEN et al.: DESIGN, ANALYSIS, AND IMPLEMENTATION OF SOLAR POWER OPTIMIZER FOR DC DISTRIBUTION SYSTEM 1765

Fig. 2. Configuration of the proposed SPO.

inductor and voltage multiplier or voltage-lift technique to real-ize high-voltage gain [9]–[29].

The proposed SPO is shown in Fig. 2; its configuration isbased on a high step-up dc–dc converter with an MPPT con-trol circuit. The converter includes a floating active switch Sand a coupled inductor T1 with primary winding N1 , which issimilar to the input inductor of a conventional boost convertercapacitor C1 , and diode D1 recycle leakage inductance energyfrom N1 . Secondary winding N2 is connected to another pairof capacitors, C2 and C3 , and to diodes D2 and D3 . Rectifierdiode D4 connects to output capacitor Co and load R. The dutyratio is modulated by the MPPT algorithm, which uses the incre-mental conductance method [30]–[35] that is employed in theproposed SPO. It detects PV module voltage Vpv and current Ipvto determine the increase and decrease in the duty cycle of thedc converter. Therefore, the MPP can be obtained by compar-ing instantaneous conductance I/V and incremental conductancedI/dV. The algorithm is programmed into TMS320LF2407A, adigital signal microprocessor.

The proposed converter has the following features: 1) itsvoltage conversion ratio is efficiently increased by using theswitched capacitor and coupled inductor techniques; 2) the leak-age inductance energy of the coupled inductor can be recycledto increase efficiency, and the voltage spike on the active switchis restrained; 3) the floating active switch isolates the PV panel’senergy during nonoperating conditions, thereby preventing anypotential electric hazard to humans or facilities. The MPPTcontrol algorithm exhibits high-tracking efficiency; hence, it iswidely used in the energy harvesting of PV systems.

The rest of the paper is organized as follows. Sections IIand III discuss the operating principle and steady-state analysisof the proposed converter, respectively. Section IV addressesthe practical implementation and component selection of theproposed converter. Section V presents the experimental results,and VI concludes the paper.

II. OPERATING PRINCIPLES

The operating principles for continuous conduction mode(CCM) and discontinuous conduction mode (DCM) are pre-sented in detail. Fig. 3 illustrates a typical waveform of severalmajor components in CCM operation during one switching pe-riod. To simplify the circuit analysis of the proposed converter,the following assumptions are made:

1) all components are ideal, except for the leakage induc-tance of coupled inductor T1 , which is taken into account.On-state resistance RDS(ON) and all the parasitic capaci-

Fig. 3. Typical waveforms of the proposed converter in CCM operation.

tances of main switch S are disregarded, as are the forwardvoltage drops of diodes D1 to D4 ;

2) capacitors C1 to C3 and Co are sufficiently large that thevoltages across them are considered constant;

3) the equivalent series resistance (ESR) of capacitors C1 toC3 and Co , as well as the parasitic resistance of coupledinductor T1 , is neglected;

4) turns ratio n of coupled inductor T1 windings is equal toN2 /N1 .

The CCM operating modes are described as follows.

A. CCM Operation

Mode I [t0 , t1]: During this interval, switch S and diodesD2 and D3 are conducted; diodes D1 and D4 are turned OFF.The current flow path is shown in Fig. 4(a). Magnetizing in-ductor Lm continues to release energy to capacitors C2 and C3through secondary winding N2 of coupled inductor T1 . Leakageinductance Lk 1 denotes the stored energy from source energyVin . The energy that is stored in capacitor Co is constantly dis-charged to load R. This mode ends when increasing iLk1 is


Fig. 4. Current flow path in five operating modes during one switching periodin CCM operation: (a) Mode I, (b) Mode II, (c) Mode III, (d) Mode IV, and(e) Mode V.

equal to decreasing iLm at t = t1

vLm = Vin (1)

ΔiLm =Vin

Lm(t1 − t0). (2)

Mode II [t1 , t2]: During this interval, switch S and diodeD4 are conducted. Source energy Vin is serially connected toC1 , C2 , and C3 , and secondary winding N2 ; Lk2 discharges theenergy that is stored in charge output capacitor Co and loadsR. Meanwhile, magnetizing inductor Lm also receives energyfrom Vin . The current flow path is shown in Fig. 4(b). This modeends when switch S is turned OFF at t = t2

vLm =Vo − Vin − Vc1 − Vc2 − Vc3

n(3)

n =N2

N1(4)

ΔiLm =Vo − Vin − Vc1 − Vc2 − Vc3

n · Lm· (t2 − t1). (5)

Mode III [t2 , t3]: During this transition interval, switch Sand diodes D2 and D3 are turned OFF, and diodes D1 and D4are conducted. The current flow path is shown in Fig. 4(c). Theenergy stored in leakage inductance Lk 1 instantly flows throughthe diode D1 to charge capacitor C1 . The energy is released tomagnetizing inductor Lm through coupled inductor T1 , whichis serially connected to C1 , C2 , and C3 , and secondary windingN2 ; Lk2 discharges the energy that is stored in charge outputcapacitor Co and loads R. This mode ends when decreasingiLk1 is equal to increasing iLm at t = t3

vLm = −Vc1 (6)

ΔiLm =−Vc1

Lm· (t3 − t2). (7)

Mode IV [t3 , t4]: During this interval, switch S and diodeD4 are turned OFF, and diodes D1 ,D2 , and D3 are conducted.The current flow path is shown in Fig. 4(d). Leakage inductanceLk 1 continues to release energy to charge capacitor C1 throughdiode D1 . Magnetizing inductor Lm through coupled inductorT1 transfers energy to capacitors C2 and C3 . The energy that isstored in capacitor CO is constantly discharged to load R. Thismode ends when decreasing iLk1 is zero at t = t4

vLm = −Vc1 (8)

ΔiLm =−Vc1

Lm· (t4 − t3). (9)

Mode V [t4 , t5]: During this interval, diodes D2 and D3 areconducted. The current flow path is shown in Fig. 4(e). Mag-netizing inductor Lm constantly transfers energy to secondarywinding N2 , and charges capacitors C2 and C3 . The energy thatis stored in capacitor CO is constantly discharged to load R.This mode ends when switch S is turned ON at the beginning ofthe next switching period

vLm =−V c2

n=

−Vc3

n(10)

ΔiLm =−Vc2

n · Lm· (t5 − t4) =

−Vc3

n · Lm· (t5 − t4). (11)

B. DCM Operation

Fig. 5 illustrates a typical waveform of several major compo-nents in DCM operation during one switching period.

Mode I [t0 , t1]: During this interval, switch S and D4 areconducted, and diodes D1 ,D2 , and D3 are turned OFF. Thecurrent flow path is shown in Fig. 6(a). Magnetizing inductor Lm

with leakage inductance Lk 1 stores energy from source energyVin . Meanwhile, source energy Vin is also serially connected tocapacitors C1 , C2 , and C3 , and secondary winding N2 to chargecapacitor Co and load R. This mode ends when switch S isturned OFF at t = t1

vLm = Vin =Vo − Vin − Vc1 − Vc2 − Vc3

n(12)


Fig. 5. Typical waveforms of the proposed converter during DCM operation.

ΔiLm =Vin

Lm· (t1 − t0) =

Vo − Vin − Vc1 − Vc2 − Vc3

n · Lm

· (t1 − t0). (13)

Mode II [t1 , t2]: During this transition interval, switch Sand diodes D2 and D3 are turned OFF, and diodes D1 and D4are conducted. The current flow path is shown in Fig. 6(b). Theenergy stored in leakage inductance Lk 1 instantly flows throughthe diode D1 to charge capacitor C1 ; this energy is also releasedto magnetizing inductor Lm through the coupled inductor T1series that is connected to C1 , C2 , and C3 , secondary windingN2 , and Lk2 to charge output capacitor Co and load R. Thismode ends when decreasing iD4 is zero at t = t2

vLm = −Vc1 (14)

ΔiLm =−Vc1

n · Lm· (t2 − t1). (15)

Fig. 6. Current flow path in five operating modes during one switching periodin DCM operation: (a) Mode I, (b) Mode II, (c) Mode III, (d) Mode IV, and(e) Mode V.

Mode III [t2 , t3]: During this transition interval, switch Sand diode D4 are turned OFF, and diodes D1 ,D2 , and D3 areconducted. The current flow path is shown in Fig. 6(c). Leakageinductance Lk 1 continues to release energy to charge capacitorC1 through diode D1 . Magnetizing inductor Lm transfers en-ergy to capacitors C2 and C3 through coupled inductor T1 . Theenergy stored in capacitor Co is constantly discharged to loadR. This mode ends when decreasing iLk1 is zero at t = t3

vLm = −Vc1 =−Vc2

n=

−Vc3

n(16)

ΔiLm =−Vc1

n · Lm· (t3 − t2) =

−Vc2

n · Lm· (t3 − t2)

=−Vc3

n · Lm· (t3 − t2). (17)


Mode IV [t3 , t4]: During this interval, switch S, diodes D1and D4 are turned OFF, and diodes D2 and D3 are conducted.The current flow path is shown in Fig. 6(d). Magnetizing induc-tor Lm constantly transfers energy to secondary winding N2and charges capacitors C2 and C3 . The energy that is stored incapacitor Co is constantly discharged to load R. This mode endswhen decreasing iLm is zero at t = t4

vLm =−Vc2

n=

−Vc3

n(18)

ΔiLm =−Vc2

n · Lm· (t5 − t4) =

−Vc3

n · Lm· (t4 − t3). (19)

Mode V [t4 , t5]: During this interval, the switch and allthe diodes are turned OFF. The current flow path is shown inFig. 6(e). The energy that is stored in capacitor CO is constantlydischarged to load R. This mode ends when switch S is turnedON at the beginning of the next switching period

vLm = 0 (20)

ΔiLm = 0. (21)

III. STEADY-STATE ANALYSIS

A. CCM Operation

Only steady-state analysis is considered during CCM oper-ation, and the leakage inductances at primary and secondarysides are disregarded. Applying a volt-second balance on themagnetizing inductance Lm yields

∫ DTS

0Vindt +

∫ TS

DTS

(−VC 1)dt = 0 (22)

and ∫ DTS

0(nVin)dt +

∫ TS

DTS

(−VC 2)dt = 0 (23)

from which the voltage across capacitors C1 and C2 are obtainedas follows:

VC 1 =D

1 − D· Vin (24)

and

VC 2 = VC 3 = n · VC 1 =n · D1 − D

· Vin . (25)

At Mode I, the output voltage is Vo = Vin + VC 1 + VN 2 +VC 2+ VC 3 . Voltage gain VO /Vin can be obtained as follows:∫ DTS

0(Vo − Vin − VC 1 − 2 · VC 2)dt +

∫ TS

DTS

(−VC 2)dt = 0

(26)

Vo = Vin + VC 1 +1 + D

D· VC 2 (27)

Vo

Vin=

n · D + n + 11 − D

(28)

B. BCM Operating Conditions

The leakage inductances at primary and secondary sides aredisregarded in this approach. We refer to (12) and (16) in

applying volt-second balance on magnetizing inductance Lm ,deriving

∫ DTS

0Vindt +

∫ DL TS

DTS

(−VC 1)dt = 0 (29)

where DLTS is the time during which current iLm declines frompeak current to zero

DL ≤ (1 − D). (30)

From (16), (29), and (30), the voltage across capacitors C2and C3 are obtained

VC 1 =D

DL· Vin (31)

VC 2 = VC 3 = n · Vc1 =n · DDL

· Vin . (32)

On the basis of (12) and (18), the voltage across Lm can bedetermined using the volt-second balance principle

∫ DTS

0(Vo −Vin −VC 1 −2 ·VC 2)dt+

∫ (DL +D )TS

DTS

(−VC 2)dt = 0

(33)

Vo = Vin + Vc1 +2 · D + DL

D· Vc2 . (34)

Substituting (31) and (32) into (34) yields the voltage gain inDCM operation; DL is defined as follows:

Vo

Vin=

(2 · n + 1) · D + (n + 1) · DL

DL(35)

DL =(2 · n + 1) · DVo

V in− (n + 1)

. (36)

In terms of IN 1 = nIo , input current peak iin,peak and mag-netizing current peak iLm,peak are determined by

iin,peak = iLm,peak + (n + 1) · io,peak (37)

iLm,peak =Vin

Lm· DTS . (38)

On the basis of (35), (37), and (38), the input current isdetermined as follows:

Iin =Vin · D2 · TS

2 · Lm+

(n + 1) · Iin

Vo/Vin. (39)

Substituting (35) into (39) yields the DCM operation voltagegain and normalized magnetizing inductance time constant τLm

Vo

Vin·(

Vo

Vin− (n + 1)

)=

Vo · D2 · TS

2 · Lm · Io(40)

τLm =Lm

R · TS=

D2

2 ·(( Vo

V in)2 − Vo

V in· (n + 1)

) (41)

Vo

Vin=

(n + 1) ±√

(n + 1)2 + 2·D 2

τL m

2. (42)


Fig. 7. Voltage gain Vo /Vin as the function of duty ratio D by various turnsratios for the proposed converter under CCM operation.

Fig. 8. Voltage gain MCCM as a function of duty ratio D for the proposedconverter compared with [22] and [23] under CCM operation and n = 4.

When the proposed converter is operating in BCM, the CCM,and DCM operations have equal voltage gains. Boundary nor-malized magnetizing inductance time constant τLmB can beobtained by

τLmB =D(D − 1)2

2(2n + 1)(nD + n + 1). (43)

The plot of voltage gain Vo /Vin as a function of duty ratio D byvarious turns ratios are shown in Fig. 7. The plot of voltage gainVo /Vin as a function of duty ratio D of the proposed converteris compared with three different converters [22] and [23]; allconverters operate at CCM and n = 4, as shown in Fig. 8. Thehigh step-up voltage capability of proposed converter is higherthan [22] and [23] along with duty ratio. Fig. 9 depicts theboundary condition of τLmB of the proposed converter undern = 2–6. With Vo /Vin = 20, Fig. 10 illustrates the magnetizinginductance and turns ratios as a function of duty ratio D. Thischart is convenient to magnetizing inductance design.

Fig. 9. Boundary condition of τLB of the proposed converter under n = 2–6.

Fig. 10. Magnetizing inductance and turns ratios n as a function of duty ratioD when Vo /Vin = 20.

TABLE ISPECIFICATIONS OF THE PROTOTYPE

C. Instantaneous MPPT Accuracy

To estimate the accuracy of MPPT, the instantaneous MPPTaccuracy is defined as

ηMPPT =PPV(t)

PMPPT(t)× 100% (44)

where Ppv(t) is the result of voltage Vpv(n) multiplied by cur-rent Ipv(n); PMPPT(t) denotes the real maximum PV outputpower that is obtained from a PV module or PV simulator.

IV. DESIGN CONSIDERATIONS OF THE PROPOSED CONVERTER

A 300 W SPO prototype is presented to verify the feasibilityof the proposed converter. The basic specifications are shownin Table I. The considerations for component parameter designand selection are described as follows.


Fig. 11. Measurement waveforms of active components at (a) 60 W and Vin = 20 V, (b) 150 W and Vin = 30 V, and (c) 300 W and Vin = 40 V.

A. Duty Ratio and Turns Ratio

The largest voltage gain is 20 (see Table I). The turns ratiocan be set to 2–6 [see (28) and Fig. 7]. When n = 2, the dutyratio is equal to 77.2%. When the duty ratio is larger than 70%,conduction losses significantly increase. If turn ratio n ≥ 5results in a small duty ratio and low magnetizing inductance,but a high peak current over the active switch occurs. Therefore,n = 4 is the appropriate choice. As determined with (28), dutyratio D is 62.5%.

B. Magnetizing Inductor

Substituting the values of duty ratio, turns ratio, and operatingfrequency into (43) yields a boundary magnetizing inductanceof 20.86 μH. This value can also be obtained by using Fig. 10.However, the actual magnetizing inductance is 21.87 μH andthe leakage inductance is about 0.22 μH.

C. Active Switch and Diodes

The highest input voltage is 40 V and its corresponding dutyratio is 35.7%. The voltage stress of diodes D1 to D4 can beobtained by

VD1 =1

1 − D· Vin =

11 − 0.357

· 40 = 62.2V (45)

VD2 = VD3 =n

1 − D· Vin = 62.2 · 4 = 248.8V (46)

VD4 =n + 11 − D

· Vin = 62.2 · 5 = 310V. (47)

The voltage stresses of switch S is shown as follows:

VDS =1

1 − D· Vin = 62.2V. (48)

By considering the parasitic capacitor and inductor effectson the actual components and the PCB, the voltage rating ofMOSFETs is higher than the calculated value. The nominalvoltage of IXFX150N15 drain-source is 150 V, which is denotedas S. Diode D1 uses MBR30100CT, which has a voltage ratingof 100 V. Diodes D2 and D3 are UF3003. The voltage rating is300 V and BYR29-600 is a 600 V diode denoted as D4 .

D. Switched Capacitors

The voltage ripple is set to 2% of the capacitor voltage. Thevoltage across capacitors C1 to Co can be obtained by

C1 =IoTS

ΔVC 1=

0.75 · 20 × 10−6

0.02 · 0.6251−0.625 · 20

= 22.5 μF (49)

C2 = C3 =IoTS

ΔVC 2=

4.5 × 10−6

4∼= 5.6μF (50)

Co =IoDTS

ΔVo=

0.75 · 0.625 · 20 · 10−6

0.02 · 400∼= 1.2μF (51)

Because a high capacitance is linked to low ESR, the actualcapacitance that is considerably larger than that calculated is


Fig. 12. Measured efficiency of the proposed SPO at various load conditions.

Fig. 13. MPPT accuracy distributions of the proposed SPO measured tentimes.

selected; C1 is 68 μF, C2 and C3 are both 220 μF, and Co is100 μF.

V. EXPERIMENTAL RESULTS

Using general power supply for three test conditions, namely,low, medium, and high insolation conditions, are used to verifythe various irradiation circumstances for the SPO. The low in-solation condition, input voltage is 20 V and PV output poweris 60 W. The measurement waveforms of the voltage and cur-rent of active switch S and diodes D1 ,D2 , and D4 , as well asthe voltage waveforms of capacitors C1 , C2 , and Co , are shownin Fig. 11(a). For the medium insolation condition, the inputvoltage is 30 V and PV output load is 150 W; the measurementresults are shown in Fig. 11(b). For the high insolation condition,the input voltage is 40 V and PV output power is 300 W, whichis the full load condition; the measurement results are shown inFig. 11(c). These current and voltage waveforms are consistentwith the operating principle and the steady-state analysis. ThePV simulator Agilent E4360 A is used to simulate various inso-lation conditions for the rest of SPO tests. The maximum SPOefficiency is 96.7%; and the full-load efficiency is about 92.8%at MPP voltage Vmp is 35.8 V; they are shown in Fig. 12. The

efficiency curve, which illustrates the maximum efficiency, isallocated at half-output power and Vmp = 33 V of the PV mod-ule. The MPPT accuracy is defined in (44). Fig. 13 shows thatthe MPPT distribution of the SPO is measured ten times to ver-ify the feasibility and capability of the MPPT control algorithmin the SPO. At an MPP of 60 W, the average accuracy is about96.8%. At an MPP of 150 and 300 W, the average accuracylevels are 97.9% and 97.8%, respectively. The maximum MPPTaccuracy is 99.9% at each irradiation condition. However, theMPPT at low MPP is less accurate than that at half and full PVoutput power.

VI. CONCLUSION

The high step-up SPO uses the coupled inductor with an ap-propriate turns ratio design and switched-capacitor technologyto achieve a high-voltage gain that is 20 times higher than theinput voltage. Because the leakage inductance energy of a cou-pled inductor is recycled and the voltage stress across the activeswitch S is constrained, the low RDS(ON) of active switch canbe selected to improve maximum efficiency up to 96.7%. As aresult, full load efficiency reaches 92.8%. The highest MPPTaccuracy is 99.9% and the highest average accuracy is 97.9% atPPV = 150 W. A 300 W SPO with a high step-up voltage gainand MPPT functions are implemented and verified.

REFERENCES

[1] Y. Fang and X. Ma, “A novel PV microinverter with coupled inductors anddouble-boost topology,” IEEE Trans. Power Electron., vol. 25, no. 12,pp. 3139–3147, Dec. 2010.

[2] A. Ch. Kyritsis, E. C. Tatakis, and N. P. Papanikolaou, “Optimum designof the current-source flyback inverter for decentralized grid-connectedphotovoltaic systems,” IEEE Trans. Energy Convers., vol. 23, no. 1,pp. 281–293, Mar. 2008.

[3] P. Tsao, “Simulation of PV systems with power optimizers and distributedpower electronics,” in Proc. IEEE Photovolt. Spec. Conf., Jun. 2010,pp. 389–393.

[4] D. D.-C. Lu and V. G. Agelidis, “Photovoltaic-battery-powered DC bussystem for common portable electronic devices,” IEEE Trans. PowerElectron., vol. 24, no. 3, pp. 849–855, Feb. 2009.

[5] L. Zhang, K. Sun, Y. Xing, L. Feng, and H. Ge, “A modular grid-connectedphotovoltaic generation system based on DC bus,” IEEE Trans. PowerElectron., vol. 26, no. 2, pp. 523–531, Feb. 2011.

[6] S. M. Chen, K. R. Hu, T. J. Liang, L. S. Yang, and Y. H. Hsieh, “Imple-mentation of high step-up solar power optimizer for DC micro grid appli-cation,” in Proc. IEEE Appl. Power Electron Conf., Feb. 2012, pp. 28–32.

[7] A. Pratt, P. Kumar, and T. V. Aldridge, “Evaluation of 400 V DC distribu-tion in telco and data centers to improve energy efficiency,” in Proc. IEEEInt. Telecommun. Energy Conf., Sep./Oct. 2007, pp. 32–39.

[8] L. Zhang, K. Sun, Y. Xing, L. Feng, and H. Ge, “A modular grid-connectedphotovoltaic generation system based on DC bus,” IEEE Trans. PowerElectron., vol. 26, no. 2, pp. 523–531, Feb. 2011.

[9] S. M. Chen, T. J. Liang, L. S. Yang, and J. F. Chen, “A boost converter withcapacitor multiplier and coupled inductor for AC module applications,”IEEE Trans. Ind. Electron., Early Access Articles, vol. PP, no. 99, p. 1.

[10] A. C. Nanakos, E. C. Tatakis, and N. P. Papanikolaou, “A weighted-efficiency-oriented design methodology of flyback inverter for AC photo-voltaic modules,” IEEE Trans. Power Electron., vol. 27, no. 7, pp. 3221–3233, Jul. 2012.

[11] S. Zengin, F. Deveci, and M. Boztepe, “Decoupling capacitor selection inDCM flyback PV micro-inverters considering harmonic distortion,” IEEETrans. Power Electron., Early Access Articles, vol. pp., no. 99, p. 1.

[12] B. Axelrod, Y. Berkovich, and A. Ioinovici, “Switched-capacitor/switched-inductor structures for getting transformerless hybrid DC-DCPWM converters,” IEEE Trans. Circuits Syst. I, Reg. Papers, vol. 55,no. 2, pp. 687–696, Mar. 2008.


[13] O. Abutbul, A. Gherlitz, Y. Berkovich, and A. Ioinovici, “Step-upswitching-mode converter with high voltage gain using a switched- capac-itor circuit,” IEEE Trans. Syst. I, Fundam. Theory Appl., vol. 50, no. 8,pp. 1098–1102, Aug. 2003.

[14] S. C. Tan, S. Bronstein, M. Nur, Y. M. Lai, A. Ioinovici, and C. K. Tse,“Variable structure modeling and design of switched-capacitor convert-ers,” IEEE Trans. Circuits Syst. I, Reg. Papers, vol. 56, no. 9, pp. 2132–2142, Sep. 2009.

[15] G. Zhu and A. Ioinovici, “Switched-capacitor power supplies: DC voltageratio, efficiency, ripple, regulation,” in Proc. IEEE Int. Symp. Circuitssyst., May 1996, pp. 553–556.

[16] F. L. Luo, “Switched-capacitorized DC/DC converters,” in Proc. IEEEConf. Ind. Electron. Appl., May 2009, pp. 1074–1079.

[17] L. S. Yang, T. J. Liang, and J. F. Chen, “Transformerless DC-DC converterswith high step-up voltage gain,” IEEE Trans. Ind. Electron., vol. 56, no. 8,pp. 3144–3152, Aug. 2009.

[18] B. Axelrod, Y. Berkovich, and A. Ioinovici, “Transformerless DC-DCconverters with a very high DC line-to-load voltage ratio,” in Proc. IEEEInt. Symp. Circuits syst., May 2003, vol. 3, pp. 435–438.

[19] F. L. Luo and H. Ye, “Positive output multiple-lift push-pull switched-capacitor Luo-converters,” IEEE Trans. Ind. Electron., vol. 51, no. 3,pp. 594–602, Jun. 2004.

[20] M. Zhu and F. L. Luo, “Enhanced self-lift cuk converter for negative-to-positive voltage conversion,” IEEE Trans. Power Electron., vol. 25, no. 9,pp. 2227–2233, Apr. 2010.

[21] Y. Jiao, F. L. Luo, and M. Zhu, “Voltage-lift-type switched-inductor cellsfor enhancing DC-DC boost ability: Principles and integrations in Luoconverter,” IET Trans. Power Electron., vol. 4, no. 1, pp. 131–142, Jan.2011.

[22] J. W. Baek, M. H. Ryoo, T. J. Kim, D. W. Yoo, and J. S. Kim, “Highboost converter using voltage multiplier,” in Proc. Annu. IEEE Conf. Ind.Electron. Soc., 2005, pp. 567–572.

[23] R. J. Wai and R. Y. Duan, “High step-up converter with coupled-inductor,”IEEE Trans. Power Electron., vol. 20, no. 5, pp. 1025–1035, Sep. 2005.

[24] S. K. Changchien, T. J. Liang, J. F. Chen, and L. S. Yang, “Novel highstep-up DC–DC converter for fuel cell energy conversion system,” IEEETrans. Ind. Electron., vol. 57, no. 6, pp. 2007–2017, Jun. 2010.

[25] S. M. Chen, T. J. Liang, L. S. Yang, and J. F. Chen, “A safety enhanced,high step-up DC-DC converter for AC photovoltaic module application,”IEEE Trans. Power Electron., vol. 27, no. 4, pp. 1809–1817, Apr. 2012.

[26] Q. Zhao and F. C. Lee, “High-efficiency, high step-up DC-DC converters,”IEEE Trans. Power Electron., vol. 18, no. 1, pp. 65–73, Jan. 2003.

[27] Y. Zhao, W. Li, Y. Deng, X. N. He, S. Lambert, and V. Pickert, “Highstep-up boost converter with coupled inductor and switched capacitor,” inProc. IET Power Electron. Machines Drives Conf., Apr. 2010, pp. 1–6.

[28] W. Yu, J. S. Lai, H. Qian, and C. Hutchens, “High-efficiency MOSFET in-verter with H6-type configuration for photovoltaic nonisolated AC-moduleapplications,” IEEE Trans. Power Electron., vol. 26, no. 4, pp. 1253–1260, Apr. 2011.

[29] C. Restrepo, J. Calvente, A. Romero, E. Vidal-Idiarte, and R. Giral,“Current-mode control of a coupled-inductor buck–boost DC–DC switch-ing converter,” IEEE Trans. Power Electron., vol. 27, no. 5, pp. 2536–2549, May 2012.

[30] K. H. Hussein, I. Muta, T. Hshino, and M. Osakada, “Maximum pho-tovoltaic power tracking: An algorithm for rapidly changing atmosphericconditions,” in Proc. Inst. Elect. Eng., Jan. 1995, vol. 142, no. 1, pp. 59–64.

[31] C. W. Tan, T. C. Green, and C. A. Hernandez-Aramburo, “An improvedmaximum power point tracking algorithm with current-mode control forphotovoltaic applications,” in Proc. IEEE Int. Conf. Power Electron. DriveSys., 2005, vol. 1, pp. 489–494.

[32] F. Liu, S. Duan, F. Liu, B. Liu, and Y. Kang, “A variable step size INCMPPT method for PV systems,” IEEE Trans. Ind. Electron., vol. 55, no. 7,pp. 2622–2628, Jul. 2008.

[33] H. S. Bae, S. J. Lee, K. S. Choi, B. H. Cho, and S. S. Jang, “Current controldesign for a grid connected photovoltaic/fuel cell DC–AC inverter,” inProc. 24th IEEE Appl. Power Electron. Conf., Feb. 2009, pp. 1945–1950.

[34] Q. Mei, M. Shan, L. Liu, and J. M. Guerrero, “A novel improved variablestep-size incremental-resistance MPPT method for PV systems,” IEEETrans. Ind. Electron., vol. 58, no. 6, pp. 2427–2434, Jun. 2011.

[35] A. Safari and S. Mekhilef, “Simulation and hardware implementation ofincremental conductance MPPT with direct control method using Cukconverter,” IEEE Trans. Ind. Electron., vol. 58, no. 4, pp. 1154–1161,Apr. 2011.

Shih-Ming (Orion) Chen was born in Tainan,Taiwan. He received the B.S. and Ph.D. degrees inelectrical engineering from National Cheng-KungUniversity, Tainan, Taiwan, in 2003 and 2011,respectively.

From 1991 to 1995, he was with Lumin Elec-tronics, Tainan, as a Research and Design Engineer.From 1995 to 1999, he was a Section Manager in theDepartment of Engineering, Delta Electronics, Inc.,Taipei City, Taiwan. From 1999 to 2002, he was wasa Manager in-Charge of the Tainan Research and De-

sign Center, Sino-American Electronic Company, Ltd., Tainan. In 2002, hewas a Section Manager at Product Development of LCD-TV Head Division,CHIMEI Optoelectronics Corp. In 2009, he has rejoined Delta Electronics asa Senior Manager in the Division of LCD Display Power. He is currently anAssistant Professor of Green Energy Electronics Research Center (GEERC)in the Department of Electrical Engineering, National Cheng-Kung University.His research interest are dc to dc converters, photovoltaic inverter, switchingpower supply, uninterrupted power system, CCFL/EEFL inverter design, andLED driver and chromatics control.

Tsorng-Juu (Peter) Liang (M’93–SM’10) was bornin Kaohsiung, Taiwan. He received the B.S. degree inelectrophysics from the National Chiao-Tung Univer-sity, Hsinchu, Taiwan, in 1985. He received the M.S.and Ph.D. degrees in electrical engineering from theUniversity of Missouri, Columbia, in 1990 and 1993,respectively.

He is currently a Professor of Electrical Engineer-ing and Director of Green Energy Electronics Re-search Center (GEERC), National Cheng-Kung Uni-versity (NCKU), Tainan, Taiwan. He has authored or

coauthored 50 journal and more than 100 conference papers. He was the Direc-tor of Electrical Laboratories, NCKU, from 2001–2004. His research interestsinclude high-efficiency power converters, high-efficiency lighting systems, re-newable energy conversion, and power ICs design.

Dr. Liang is an Associate Editor of IEEE TRANSACTIONS ON POWER ELEC-TRONICS and the IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS—Part I:REGOULAR PAPERS, and the Technical Committee Chair of the IEEE CAS Sys-tems Power and Energy Circuits and Systems Technical Committee. He is alsoon the Board of Directors for Compucase Enterprise Company, Ltd. and CatcherTechnology Company, Ltd. In 2008, he received the Outstanding Engineer, theChinese Institute of Electrical Engineering, the Kaohsiung Chapter, and the Out-standing Professor Award at Taiwan Power Electronics Conference. In 2010,he received Teaching Excellence Award from NCKU and the Outstanding En-gineers Professor Award from the Chinese Institute of Electrical Engineering,Kaohsiung Branch. He is a member of the IEEE Societies of Power Electronics,Industrial Electronics, Circuits And System, and Industrial Applications.

Ke-Ren Hu was born in Taipei, Taiwan. She receivedthe B.S. degree from Nation Taiwan University ofScience and Technology, Taipei, in 2009, and theM.S. degree from National Cheng-Kung University,Tainan, Taiwan, in 2011.

She is currently an Application Engineer atRichtek Technology Company, Hsinchu, Taiwan,with responsibility in system analysis and testing.Her main research interests include dc/dc power con-verters and resonant power converters.

pv inverter

Documents

power electronics

maximum pv outputpower

maximum energy

power losses

power reduction

amultistring pv array

aphotovoltaic pv panel

individual pv panels