Abstract The performance of a photovoltaic (PV) array is affected by temperature, solar insulation, shading, and array

Transcription

1 Two Stages Maximum Power Point Tracking Algorithm for PV Systems Operating under Partially Shaded Conditions Hamdy Radwan 1, Omar Abdel-Rahim 1, Mahrous Ahmed 1, IEEE Member, Mohamed Orabi 1, IEEE Senior Member and Ahmed Alaa El-Koussi 2 1 APEARC, South Valley University, Aswan City, Egypt 2 Cairo University, Cairo, Egypt Abstract-- The performance of a photovoltaic (PV) array is affected by temperature, solar insulation, shading, and array configuration. Often, the PV arrays get shadowed, completely or partially, by the passing clouds, neighboring buildings and towers, trees, and utility and telephone poles. Under partially shaded conditions, the PV characteristics get more complex with multiple peaks. This makes the tracking of the actual maximum power point (MPP) a difficult task. In this paper, a simple technique of MPPT is presented for photovoltaic power generation systems uses two stages of conventional perturb and observe method (P&O) to solve the aforementioned problems. It realizes simple control system to track the real maximum power point even under non-uniform or for rapidly changing insulation conditions. The proposed technique in the first stage searches the I-V characteristic of the PV arrays by scanning the whole range to detect the actual MPP using a P&O with large value of step. The second stage applies the conventional P&O method to oscillate around the obtained point from the first stage using a small step to obtain low oscillation with a higher efficiency. Simulation and some selected experimental results are presented to prove the feasibility of the proposed technique. I. INTRODUCTION Renewable energy sources such as solar energy are acquiring more significance, due to shortage and environmental impacts of conventional fuels. The photovoltaic (PV) system for converting solar energy into electricity is in general costly and is a vital way of electricity generation only if it can produce the maximum possible output for all weather conditions. In general, a PV source is operated in conjunction with a dc dc power converter, whose duty cycle is modulated in order to track the instantaneous MPP of the PV source. The output of the PV depends highly on an insulation condition and a surface temperature of the PV array. Moreover, there are several local maximum power points in the P-V curve under non-uniform insulation, whereas only one MPPT point is exist under uniform insulation for a given temperature and insulation. Several tracking schemes under uniform solar insulation have been proposed [1] [11]. Among the popular tracking schemes are the perturb and observe (P&O) or hill climbing [3]-[4], incremental conductance [7], short circuit current [1], open-circuit voltage [6], and ripple correlation approaches [5]. Some modified techniques have also been proposed, with the objective of minimizing the hardware or improving the performance [6], [8] [11]. However, these methods cannot readily track immediate and rapid changes in environmental conditions or non-uniform insulation of the PV modules. The non-uniform insulation of the PV modules which are connected both in series and in parallel depends on several factors such as the incidence angle of solar ray to the module, shadows full-or-partial, and so forth. Under partially shaded conditions, (e.g., due to clouds, trees, etc.), the P V characteristics get more complex, displaying multiple peaks. There is a need to develop special MPPT schemes that can track the true MPP under these conditions. In order to solve this problem, several methods have been proposed. Some Literatures are based on Fuzzy theory [12], and some others use complicated techniques [13],[14]. Another Literature [15] has proposed a two-stage method to track the MPP. In the first stage, the operating point moves into the vicinity of the MPP on the load line Rpm = Vpm/Ipm, and in the second stage, it converges to the actual MPP. Vpm and Ipm are approximately equal to 80% and 90% of the open-circuit voltage VOC and short-circuit current ISC of the PV array, respectively. However, in this method, if the actual MPP lies on the left side of the load line (Rpm), i.e., Rpm > Ractual (where Ractual = Vactual/Iactual), the operating point is temporarily shifted to 90% of VOC, thereby missing the actual MPP. In this paper, we propose a simple and practical approach for commercial base use. This method is based on two stages; in the first stage the controller searches the I-V characteristic of the PV regularly at certain specified time using large value of step of the P&O method with storing the actual maximum power point and its corresponding duty cycle. As this obtained MPPT point is not the exact one due to the used large step, a second stage of the controller is designed to oscillate around the obtained MPP point by using the conventional P&O algorithm with a small step. In another word, the algorithm during the first stage operates with large step to reduce the scanning time of the I-V characteristic curve, then after obtaining the approximated MPP, the algorithm oscillates around this MPP with a small step (delta) to provide the exact point with a small oscillation. As mentioned above, electrical characteristics of PV module are affected by environmental conditions such as temperature, solar irradiation dusts accumulation and shadow caused by 683

2 bird s clouds and so on. To study the effects of the previous environmental factors a MATLAB model was proposed the model takes into accounts most of the environmental conditions that affect the electrical characteristics of the PV. The control algorithm for extracting maximum power from the cell is proposed by means of the VHDL code and implemented using Xilinx XC3S400 FPGA Board. XILINX FPGA is a programmable logic device developed by XILINX which is considered as an efficient hardware for rapid prototyping. The paper continues as follows: Section II discuses the mathematical model of PV. Section III shows the standard P&O algorithm. Section IV discuses the factors affect the electrical characteristics of PV. In Section V, the proposed MPPT algorithm is described. In section VI, the implementation and the experimental results are described. Finally, conclusions are presented in section VII. II. MODELING OF PV ARRAY Simple equivalent circuit of ideal photovoltaic cell is shown in Fig.1. Equation 1 represents the mathematical representation of PV model as: =, (1), =(, + ) (2) Where, photo is generated current which is proportional to solar irradiation as shown in (2), is the diode current and I is the load current. Diode current can be computed from (2) as: =, ( 1) (3) The reverse saturation current of the diode is finding as:, =, (4) Fig. 3 Characteristic I-V curve of a practical photovoltaic device and the three remarkable points: short circuit (0, I sc), maximum power point (V mp, I mp) and open-circuit (V oc, 0) [1]. P o w e r [ W ] Voltage [V] Fig. 4 Plot of power versus voltage for the simulated PV. Where, q is the electron charge [ C], k is the Boltzmann constant [ J/K], T [K] is the temperature of the p-n junction, is the short-circuit current/ temperature coefficient, K is the open-circuit voltage/temperature coefficient, T=T T, and a is the diode ideality constant. The previous model is an ideal model. Practically, PV cell series and parallel resistance must be added to the model as shown in Fig. 2. The value of R s and R p can be calculated in iterative method [16]. Although the value of R s is small and that of R p is large, they have a big influence on I-V characteristics of PV model. Then, (1) is modified to (5) taking into account the influence of R s and R p. =, (5) Fig. 1 Show ideal photovoltaic cell equivalent circuit Fig. 2 Equivalent circuit of practical photovoltaic cell Figure (3) shows I-V characteristics of a practical photovoltaic cell showing the most important operating points in the curve where (0, Isc) is the short circuit point, (V mp, I mp ) is the maximum power point, and (V oc, 0) is open circuit point. A photovoltaic array consists of a number of photovoltaic cells connected in series and parallel configuration. Cells are connected in series to increase terminal voltage and are connected in parallel to increase terminal current so (6) represents a general equation for an array consist of Ns * N p cells. Where N s is the number of series connected cells and N p is the number of parallel connected cells. 684

3 =,, exp 1 ( + )/( ) (6) III. STANDARD P&O ALGORITHM Figure 4 indicates the characteristic output power curve for the solar cell from simulation under a given temperature and irradiance. MPPT control has been proposed and implemented for extracting the maximum power from the PV cell. The classic P&O algorithm is pretty simple [3], [4]. Figure 5 depicts a flow chart explaining it. It operates by perturbing the PV array voltage (i.e. incrementing or decreasing) and comparing the PV output power with that of the previous perturbation cycle. If the perturbation leads to an increase/decrease in array power, the subsequent perturbation is made in the same/opposite direction. In this manner, the peak power is tracked continuously. But this algorithm has two weaknesses. First with a small increment between two measurement points, the algorithm will be more accurate, but it will result in a slow reaction. Second a larger increment will make the algorithm more reactive but inaccurate. the rest of the modules. It means that the current available in a series connection of PV modules is limited by the current of the PV module which is less illuminated. This can be avoided by the use of bypass diodes which can be placed across a PV module. This is to allow the array current to flow in the right direction even if one of the strings is completely shadowed. Figure 6 shows the characteristics of PV array consist of 2 strings connected in parallel each string consist of four series module under identical conditions. On the other hand, Fig. 7 shows the characteristics of the PV array with two modules of one string are partially shaded, the output PV curve contains multiple peaks which affect the operation of maximum power point controller if exist and also output power is reduced due to the effect of partial shading. As the number of shaded modules increase, the worse output power obtained. The effect of shade on the performance of a PV generator depends on influences such as [17]: - Reduction of insulation (as average value) - Distribution of the shade on the PV generator (geometry of shade) - Modules with or without bypass diodes - Circuit design of PV array (series connection, or strings in parallel). When shading occurs, the reversal of the voltage can be observed in that specific section and now the bypass diode in parallel will conduct the current. The results are: - The current of the un-shaded section flows through the bypass diode and the power/voltage characteristic shows a second local maximum - The shaded cell is only loaded with that fraction of power produced by the remaining unshaded cells of that section - When the number of cells which are bridged by the by-pass diode is not too high, the level of the breakthrough voltage will not be reached. But there are also some draw backs resulting from the by-pass diodes: - Higher cost for the module production and assembly problems of the by-pass diodes. - Losses in the by-pass diode in the case of shading. Fig.5 Flowchart of the P&O algorithm. IV. PV CHARACTERISTICS UNDER SHADING EFFECT. Shading of solar cells not only reduces the cell power but also changes the open circuit voltage ( ), the short circuit current ( ), and the efficiency. Partial shading condition is a common situation due to the shadow of buildings, trees, clouds, and dirties, etc. Under partial shading condition, only one of the series strings of PV modules is less illuminated and which then has to dissipate some of the power generated by V. THE PROPOSED METHOD AND SIMULATION RESULTS. This proposed method consists of two stages, in the first stage the controller searches the I-V characteristic of the PV every certain specified time, and during this survey the controller stored the approximated maximum power point and its corresponding duty cycle. The second stage the controller oscillates around this stored MPP by using the conventional P&O algorithm. Figure 8 shows the flowchart of the proposed method, the algorithm during the first stage operates with large step ( 1) to reduce the time of searching the I-V characteristic, then after take the decision of the approximated MPP, the algorithm oscillates around this MPP with very small step ( 2) to provide the exact MPPT with small oscillation. This proposed method was designed to work in practical applications. This means that s important to ensure a good 685

4 general robustness and to prevent algorithm from working for a long time around an incorrect point (the lower peak of Figure 7) due to partial shading. Therefore, the control has been designed to provide the possibility of doing a reset. In other words, this algorithm periodically restarts its iterations (in practice, periodically means a time between 5 minutes). Simulation was done using PSIM Package to validate the operation of the proposed algorithm. An equivalent circuit for a PV array with some modules is partially shaded has been designed in the simulation. Figure 9 shows the simulation results of the partially shaded PV conditions. It shows the output power of the PV system when the proposed algorithm was applied. As shown in the figure the proposed algorithm was able to extract the real MPP after scanning process. one of them has 20V maximum voltage limit which is corresponding to be the open circuit voltage of the PV model. This represents the string of the PV without shadow (provide P max = 80W ideally). The other one has 10V maximum voltage limit which is corresponding to be the open circuit voltage of the PV model which represents the string of the PV with shadow and the blocking diode connected to the low voltage Setting module. This connection of the different setting modules provides two maximum power point P-V curves. The target of the controller is tracking the actual maximum power point (MPP). FPGA Board receive the PV output voltage and current by using low cost 8-bits analog to digital converter (ADC0804lCN) to convert the analog signal to digital signal. Fig. 6 Characteristics of a PV array under identical conditions. (a) PV array configuration (b) P-V characteristics (c) I-V characteristics. VI. IMPLEMENTATION AND EXPERIMENTAL RESULTS In generally, the design of a digital controller impact is always the implementation of a suitable data acquisition path, so that digital control requires particular care in signal conditioning and analog to digital conversion implementation [18]. A. Experimental Description Figure 10 shows the configuration of the overall system. The DC-DC boost converter (C=220µF, L= 8mH) is boosting The output voltage of the solar cell and in the same time tracks the MPP point. The boost converter extracts maximum power by using (Xilinx XC3S400 FPGA Board) in which the proposed algorithm is downloaded. Two PV model have been built with different ratings. It is used as the solar cell with two strings; each module is energized by DC power supply [19]; Fig. 7 Characteristics of a PV array under partially shaded condition. (a) PV array configuration (b) P-V characteristics (c) I-V characteristic. B. Voltage and current sensing The proposed method requires two sensors. Most of the time, it is easier and more reliable to measure voltage than current. Moreover, current sensors are usually expensive and bulky. Series sense resistor [20] is the conventional technique of sensing the current. It simply inserts a sense resistor in series with the return line of the PV. The output current of the PV is determined by sensing the voltage across it. This method obviously acquires a power loss in R sense. This challenge is overcome by using low resistance with high power rating (0.25ohm, 5W). The voltage across it equals 1.25V when the maximum current equals 5A. But the FSR (full-scale range), of the ADC depends on the amplitude of the voltage reference of the ADC0804lCN (5V). So if the maximum value input to 686

5 the ADC equal 1.25 V this means the FSR is not utilized. To boost this value to the FSR, operational amplifier is used to amplify by a gain of 4 to utilize the FSR. While the performance of the implementation by FPGA is high even if using low resolution ADC [18], this method of current sensing is acceptable and leading to overall price reduction. C. Experimental results The proposed algorithm which is described in the flow chart shown in Fig. 8 has been employed. The update rate of the MPPT algorithms was set to 25msec. This value was found experimentally by setting it as fast as possible without causing instability to the system; too fast update rate may cause the system to be instable due to the relatively long time constant of the power stage. The algorithm during the first stage operates with large step ( 1=5/256). On the other hand, the algorithm during the second stage operates with small step ( 2=2/256). Figure 12 shows the PV maximum output current and voltage and maximum power (channel 2, channel 3 and Math channel respectively). P max equal 80.26W which is the multiply of the V pvmax and I pvmax. This value means the algorithm oscillates around the actual maximum power point (80W ideally). Fig. 8 Flowchart of the proposed algorithm Fig. 10 The configuration overall hardware implementation. Fig. 9 The output power with the proposed algorithm. Fig. 11 capture of the experimentally test. 687

6 Fig. 12 the PV maximum output voltage, current and corresponding maximum power. VII. CONCLUSION In this paper, a simple technique of maximum power generation systems based on two stage algorithm to realize a simple control system to track the real maximum power point even under non-uniform or for rapidly changing insulation conditions. The proposed technique in the first stage implements a survey for the I-V characteristic of the PV arrays to detect an approximated MPP. The second stage determines the exact MPP by applying the conventional P&O method to oscillate around the determined MPP from the previous stage providing MPPT with small oscillation and higher efficiency. ACKNOWLEDGMENT The authors gratefully thank the ministry of Science, Egyptian science and technology development funds (STDF project No 346), for supporting this project. REFERENCES [1] T. Noguchi, S. Togashi, and R. Nakamoto, Short-current pulse-based maximum-power-point tracking method for multiple photovoltaic and converter module system, IEEE Trans. Ind. Electron., vol. 49, no. 1, pp , Feb [2] V. Salas, E. Olias, A. Barrado, and A. Lazaro, Review of the maximum power point tracking algorithms for stand-alone photovoltaic systems, Sol. Energy Mater. Sol. Cells, vol. 90, no. 11, pp , Jul [3] C. Hua, J. Lin, and C. Chen, Implementation of a DSPcontrolled photovoltaic system with peak power tracking, IEEE Trans. Ind. Electron., vol. 45, no. 1, pp , Feb [4] N. Fernia, G. Petrone, G. Spagnuolo, andm. Vitelli, Optimization of perturb and observe maximum power point tracking method, IEEE Trans. Power Electron., vol. 20, no. 4, pp , Jul [5] T. Esram, J. W. Kimball, P. T. Krein, P. L. Chapman, and P. Midya, Dynamic maximum power point tracking of photovoltaic arrays using ripple correlation control, IEEE Trans. Power Electron., vol. 21, no. 5, pp , Sep [6] C. Dorofte, U. Borup, and F. Blaabjerg, A combined two-method MPPT control scheme for grid-connected photovoltaic systems, in Proc. Eur. Conf. Power Electron. Appl., Sep , 2005, pp [7] K. H. Hussein and I. Muta, Maximum photovoltaic power tracking: An algorithm for rapidly changing atmospheric conditions, Proc. Inst. Electr. Eng. Generation, Transmission Distribution, vol. 142, no. 1, pp ,Jan [8] D. Sera, T. Kerekes, R. Teodorescu, and F. Blaabjerg, Improved MPPT method for rapidly changing environmental conditions, in Proc. IEEE Int. Ind. Electron. Symp., Jul. 2006, vol. 2, pp [9] N. Kasa, T. Iida, and H. Iwamoto, Maximum power point tracking with capacitor identifier for photovoltaic power system, Proc. Inst. Electr. Eng. Electr. Power Appl., vol. 147, no. 6, pp , Nov [10] N. Kasa, T. Iida, and L. Chen, Flyback inverter controlled by sensorless current MPPT for photovoltaic power system, IEEE Trans. Ind. Electron., vol. 52, no. 4, pp , Aug [11] M. Veerachary, T. Senjyu, and K. Uezato, Maximum power point tracking control of IDB converter supplied PV system, Proc. Inst Electr. Eng. - Electr. Power Appl., vol. 148, no. 6, pp , Nov [12] T. Senjyu, T. Arashiro and K. Uezato, Maximum Power Point Tracking Control of Photovoltaic Array under partial Shading Conditions, Proc. of the Tenth Annual Conference of Power & Energy Society. IEE of Japan, Volume A, 1, 1999, pp [13] E. V. Solodovnik, S. Liu, and R. A. Dougal, Power controller design for maximum power tracking in solar installations, IEEE Trans. Power Electron., vol. 19, no. 5, pp , Sep [14] Y.Kuo,T. Liang, J. Chen,"Novel maximum-power-point-tracking controllerfor photovoltaic energy conversion system," IEEE Trans. Ind. Electron. 48 (3) (2001) [15] K. Kobayashi, I. Takano, and Y. Sawada, A study of a two stage maximum power point tracking control of a photovoltaic system under partially shaded insolation conditions, Sol. Energy Mater. Sol. Cells, vol. 90, no. 18/19, pp , Nov [16] G. Villalva, J. R. Gazoli, E. Ruppert F., Modeling and circuit-based simulation of photovoltaic arrays, Brazilian Journal of Power Electronics, 2009, vol. 14, no. 1, pp [17] R. E. Hanitsch, Detlef Schulz and Udo Siegfried, Shading Effects on Output Power of Grid Connected Photovoltaic Generator Systems, Rev. Energ. Ren. Power Engineering (2001) [18] Simone Buso and Paolo Mattavelli,"Digital Control in Power Electronics," first addition.1996 [19] A.K. Mukerjee, N. Dasgupta, DC power supply used as photovoltaic simulator for testing MPPT algorithms, Renewable Energy 32 (2007) [20] H.P. Forghani-zadeh, G. A. Rincon-Mora, Current Sensing Techniques for DC-DC Converter, The th Midwest symposium on circuit and systems,2002.mwscas