Multi-levelPOMPPTcontrolPVsystemconsideringshadowinfluence

2011 11th International Conference on Control, Automation and Systems Oct. 26-29, 2011 in KINTEX, Gyeonggi-do, Korea 1. INTRODUCTION Solar energy is an important energy source since it is abundant, clean and pollution free. The electrical power supplied by a PV system depends on many extrinsic factors, such as temperature, radiation level. Under uniform radiation and temperature, a PV array exhibits a current-voltage characteristic with a unique point, called MPP, where the PV array produces maximum output power. In order to provide the maximum power, the MPPT algorithm is necessary for PV array [1-2]. The PV MIC system is small capacity power conditioning system (PCS) about 200[W], it attached to the rear of the solar cell. It takes a lot of shadow influence due to its small capacity. The conventional MPPT methods are CV, PO and IC [3-4]. The conventional MPPT methods do not consider the influence of the shadow. This means that may fail to maximum power point tracking when the shadow happened. To solve this problem, this paper proposes MLPO MPPT algorithm considering shadow influence. The MLPO MPPT algorithm is improved conventional PO method. It is robustly control for environment changing because step value of PO method is changed according to output error. The structure of this algorithm is simple and this algorithm is easily realized. This paper is compared with performance of MLPO and conventional MPPT methods and then the validity of the MLPO MPPT algorithm proves through the result. 2. SOLAR CELL MODEL Fig. 1 shows equivalent circuit of solar cell. Where, phI is photo current, dI is diode saturation current. +_Sun lightSolar cellPhIdIShRSRLoadVI LFig. 1 Equivalent circuit of solar cell The short current scI is match photo current PhI in case of ideal and solar open voltage determined diode saturation current can be expressed as follows. Multi-Level PO MPPT Control PV System Considering Shadow influenceJu-Hui Mun 1 , Jae-Sub Ko2 , Jung-Sik Choi3, Sung-Jun Kang4 , Mi-Geum Jang5 , Jin-Gook Lee 6, Dong-Hwa Chung 71 Department of Electrical Control Engineering, Sunchon National University, Suncheon, 540-742, Korea (Tel : +82-61-750-3540; E-mail: challenge211@nate.com) 2 Department of Electrical Control Engineering, Sunchon National University, Suncheon, 540-742, Korea (Tel : +82-61-750-3540; E-mail: kokos22@sunchon.ac.kr) 3 Digital Convergence Research Center, Korea Electronics Technology Institute, Gwangju, 500-480, Korea (Tel : +82-62-975-7036; E-mail: cjs1108@keti.re.kr) 4 Department of Electrical Control Engineering, Sunchon National University, Suncheon, 540-742, Korea (Tel : +82-61-750-3540; E-mail: totalxy@daum.net) 5 Digital Convergence Research Center, Korea Electronics Technology Institute, Gwangju, 500-480, Korea (Tel : +82-62-975-7037; E-mail: kumi0145@keti.re.kr) 6 Department of Electrical Control Engineering, Sunchon National University, Suncheon, 540-742, Korea (Tel : +82-61-750-3540; E-mail: dutedl@nate.com) 7 Department of Electrical Control Engineering, Sunchon National University, Suncheon, 540-742, Korea (Tel : +82-61-750-3543; E-mail: hwa777@sunchon.ac.kr) Abstract : This paper proposes the multi-level (ML) PO MPPT control of the photovoltaic (PV) system considering shadow influence. The output characteristics of the solar cell are a nonlinear and affected by a temperature, the solar radiation and shadow influence. Particularly, PV module integrated converter (MIC) system is very sensitive to the shadow influence because the capacity is very small. In order to increase an output and efficiency of the solar power generation, the maximum power point (MPP) control are necessary. Conventional constant voltage (CV) perturbation and observation (PO) and incremental conductance(IC) are the method finding MPP by the continued self-excitation vibration. The conventional MPPT control is unable to be performed when output is rapidly changing by shadow influence. To solve this problem, the new control algorithm of the MLPO in which the step value changes by output change is presented. The proposed algorithm compares the output error of the conventional control algorithm with the solar radiation, a temperature and shadow influence. In addition, the validity of this algorithm is proved through the response characteristics of the output error. Keywords: PV system, MPPT algorithm, output error, shadow influence, radiation, temperature 428978-89-93215-03-8 98560/11/$15 ICROS??????+= 1lndPhoc IIqkTV (1) Where, ocV is open voltage, k is Boltzmann constant, q is electric charge and T is operating temperature [ K ]. Also, relation equation of short current and open voltage can be expressed as follows. ?????? ??????=kTqVII ocosc exp (2) An equation to obtain the current-voltage characteristic curve of the solar cell can be expressed as follows. )( rctNscph TTISII -+= (3) ??????-?????? += 1)(expAkTRIVqII sLLod (4) ????????????????-???????=crgrcoro TTBkqETTII 11exp3(5) shsLLdphL RRIVIII +--= (6) Where, NS : the radiation per area tI : short current temperature coefficient when surface temperature is increased 1℃ [A/K] cT : solar cell temperature[K] rT : solar cell operating standard temperature[K] B : manufacture constant orI : reverse saturation current at rT [A] gE : energy band gap The output characteristic of solar cell array is obtained using (3)~(6). Fig. 2 shows PSIM model of solar cell array in the 3.2[kW] class considering temperature and radiation. Fig. 2 PSIM model of solar cell array Fig. 3 shows simulation result of solar cell array. To obtain maximum power from solar cell, operating voltage and current must be controlled carefully. Fig. 3 I-V, P-V characteristic waveform of solar cell array The characteristic curve of solar cell is seriously affected to environment factors such as temperature of cell and radiation. Fig. 4, 5 shows output characteristics of solar cell by radiation and temperature. Fig. 4 Output characteristics of solar cell by radiation Fig. 5 Output characteristics of solar cell by temperature 2. CONVENTIONAL MPPT ALGORITHM 2.1 The Perturbation & Observation (PO) method The most commonly used MPPT algorithm is the PO, due to its simplicity of implementation in its basic form. However, it has some drawbacks, like oscillations around the MPP in steady state operation, slow response 429speed due to the change of the solar radiation. PO algorithm is based on the calculation of the PV array output power and the power change by sensing both the PV current and voltage. The tracker operates periodically by comparing the actual value of the power with the previous value to determine the change (incrementing or decrementing) on the solar array voltage or current (depending on the control strategy). If a given perturbation leads to an increase (decrease) in the output power of the PV, then the subsequent perturbation is generated in the same (opposite) direction. The algorithm is summarized in fig. 6. When the MPP is reached, the system then oscillates around the MPP. In order to minimize the oscillation, the perturbation step size should be reduced. However, a smaller step size slows down the MPPT. START)1(V)(V)(V refrefref --=Δ kkkprefref CkVkV -=+Δ )()1(I(k)V(k),Measure)(I)(V)( kkkP =)1(P)(P)(P --=Δ kkk0)( >Δ kP0)( >Δ kVref 0)( (8) MPPofrightdVdP ,0ΔΔ (12) MPPofrightVIVI ,--- kVkVYESYESYES YESNONO NONOVIdVdI // -=VIdVdI // ->0)1()( =-- kIkI0)1()( >-- kIkIReturnFig. 7 Flowchart of IC MPPT Method 3. MULTI-LEVEL (ML) PO MPPT ALGORITHM The conventional MPPT algorithms are not considering influence by shadow. This means that maximum power point tracking may fails when MPP is many by shadow influence. Fig. 8 shows MPPT characteristics of PO and IC with shadow occurrence. If MPPT is normally tracked when shadow is not occurred then operating point is A. If portion of solar cell becomes on bypass-diode by shadow then PV output curve is changed and operating point moves B. B point is local MPP and C is real MPP in fig. 9. PVPPVIPVVReal MPPAReal MPPCBLocal MPPCPBPBVCVBICIVΔFig. 8 Tracking characteristics of PO and IC with partial shadow In case of MPP moves to B from A, conventional PO and IC method is realized radiation changing and it is 430detected MPP through voltage changes as VΔ . B is recognized MPP, because the point( VVB Δ+ , VVB Δ- ) is lower than B. As a result, these methods has self-excitation vibration near the B This paper proposes MLPO method algorithm to increase efficiency of PV system through considering shadow influence. MLPO MPPT method is improvement of conventional PO method that changes step size with environment condition. Fig. 9 shows MLPO MPPT algorithm proposed in this paper. It can be selected one between three acceleration coefficient α using changing value by comparison with existing power and previous power. The new command current refI is expressed as follows. IkIkI refref Δ+-= α)1()( (14) Where, the acceleration coefficient α is controlled that sensitively operated to environment changing such as shadow influence and radiation variation. Also, it is controlled fast tracking when changing range of optimal current and voltage is large and exactly tracked when operating point is near real MPP. 3kkr =Sense ,015.0=α 003.0=α 1=αSTART1kkr -= 2kkr = 4kkr -=RETURNYesNoNoYesYes NoYes No Yes No)(kVPV )( kI PV)()()( kIkVkP PVPVPV ?=)1()(_ --= kPkPP PVPVdiffPVε>diffPVP _2MPPPV PP >0_ >diffPVP)1()( -> kVkV PVPV )1()( -> kVkV PVPV)( _ PVdiffPVr PPkI =Δ)()1()( IkIkI refref Δ+-= αFig. 9 Flowchart of proposed MPPT algorithm The vibration step signal is defined by rk that represents direction for slope signal in VP -characteristic curve and it is as follows. PVdiffPVr PPkI _=Δ (15) )1()(_ --= kPkPP PVPVdiffPV (16) Where, )(kPPV is PV power at k th, )1( -kPPV is PV power at 1-k th. Also, rk is coefficient about four operating mode mixed vibration direction and slope direction of PVPV dVdP / . The vibration cycle, the MPP from the PV power about environment changing is performed repeatedly. The MPPT control is performed using DC-DC boost-converter to track MPP. To perform current-controlled MPPT control structure of the boost converter is shown in fig. 10. Proposed DC-DC converter control algorithm, including input and output, and output of improved PO control algorithm is command current refI corresponding optimal current. S2DS2DS1LmLlkCPVCcCrDo2Do1Cd V o+_S1VPV+_iS1TiS2I PVI LI oMPPTAlgorithmPISawtoothPWMSwitchingDutycycle+ _ILIrefV PVI PVerrDuty cycleDuty cycleIFig. 10 Configuration of DC-DC converter for MPPT control 4. PERFORMANCE RESULT OF SYSTEM Fig. 11 shows PSIM circuit to analyze performance of MPPT algorithms. Fig. 11 PSIM circuit for MPPT control 4.1 Changing of radiation and temperature Fig. 12 shows changing of radiation (400 → 1,000→600 [W/ ㎡ ]) and temperature (28 → 35→ 31[ ℃ ]). 431Fig. 12 Changing of radiation and temperature Fig. 13, 14 and 15 shows output power error of PO, IC and MLPO method. The maximum power error of PO method is about 3.9[%], IC is about 3.6[%] and MLPO is within 1[%]. (a) 400[W/ ㎡ ], 28[ ℃ ]→ 1000[W/ ㎡ ], 35[ ℃ ] (b) 1000[W/ ㎡ ], 35[ ℃ ]→ 600[W/ ㎡ ], 31[ ℃ ] Fig. 13 Output power error of solar cell module (PO method) (a) 400[W/ ㎡ ], 28[ ℃ ]→ 1000[W/ ㎡ ], 35[ ℃ ] (b) 1000[W/ ㎡ ], 35[ ℃ ]→ 600[W/ ㎡ ], 31[ ℃ ] Fig. 14 Output power error of solar cell module (IC method) (a) 400[W/ ㎡ ], 28[ ℃ ]→ 1000[W/ ㎡ ], 35[ ℃ ] (b) 1000[W/ ㎡ ], 35[ ℃ ]→ 600[W/ ㎡ ], 31[ ℃ ] Fig. 15 Output power error of solar cell module (MLPO) 4.2 Considering shadow influence Fig. 16 shows I-V and P-V characteristics curve of solar cell considered shadow influence. The shadow influence is considered 30[%], 50[%] of solar cell area. (a)Characteristic curve of solar cell considering shaded effect (b) Characteristic curve of solar cell considering shaded effect Fig. 16 Characteristic curve of solar cell module considering shaded effect Fig. 17, 18 and 19 shows maximum power error of PO, IC and MLPO method when considering shadow influence. The maximum power error of PO, IC, MLPO at 30[%] shadow is each 15[W], 14.5[W] and within 1[%]. And 50[%] shadow is each 18[W], 18[W], within 4321[%]. The maximum power error of MLPO method proposed in this paper is less than other methods. (a) 30[%] of shaded effect (b) 50[%] of shaded effect Fig. 5.23 Output power error of solar cell module (PO method) (a) 30[%] shaded effect (b) 50[%] shaded effect Fig. 5.25 Output power error of solar cell module (IC method) (a) 30[%] shaded effect (b) 50[%] shaded effect Fig. 5.27 Output power error of solar cell module (MLPO) 5. CONCLUSION This paper proposes MLPO algorithm for PV MIC system considering shadow influence. The MLPO is easily implemented due to based on PO method. Also, perturbation step size is changed according to output variation. Thus, it can be fast tracked MPP when environment factors such as radiation and temperature are changed. And its loss is decreased because self-excitation vibration is small in case of reached MPP. This paper compares with MLPO method and conventional MPPT methods such as PO and IC under various conditions such as radiation, temperature changing and shadow influence. The maximum power error of MLPO is smaller than PO and IC. Therefore, the validity of MLPO proposed in this paper is proved. This work (Grants No. 00048024) was supported by Business for Cooperative R&D between Industry, Academy, and Research Institute funded Korea Small and Medium Business Administration in 2011. REFERENCES [1] H. J. Noh, D. Y . LEE, D. S. Hyun, “ An improved MPPT converter with current compensation method for small scaled PV-applications, ” IEEE IES, V ol. 2, pp.1113-1118, 2002. [2] R. Andoubi, A. Mami, G . Dauphin, M. Annabi, “ Bond graph modelling and dynamic study of a photovoltaic system using MPPT buck-boost converter, ” IEEE ICS, Vol. 3, pp.200-205, 2002. [3] N. Femia, G . Petron, G . Spagnuolo, and M. Vitelli, “ Optimization of perturb and observe maximum power point tracking method, ” IEEE Trans. Power Electron., V ol. 20, no. 4, pp. 963-973, Jul. 2005. [4] P. Wolfs, Q. Li “ A current-sensor-free incremental conductance single cell MPPT for high performance vehicle solar arrays, ” PESC06. pp. 1-7. 433