52 CHAPTER 3 CUK CONVERTER BASED MPPT SYSTEM USING ADAPTIVE PAO ALGORITHM 3.1 INTRODUCTION The power electronics interface, connected between a solar panel and a load or battery bus, is a pulse width modulated DC-DC converter used to extract maximum power from solar PV panel. The I-V characteristic curve of photovoltaic generator is validated based on various DC-DC converters. The nominal duty cycle of the main switch in the DC-DC converter is adjusted to a value, so that the input resistance of the converter is equal to the equivalent output resistance of the solar panel at the MPP. This approach ensures maximum power transfer under all atmospheric conditions. Buck converter cannot emulate smaller impedance than the load impedance, and therefore, it does not reach values near the short circuit current of the PV module. The input resistance of buck converter is given by R i = R / d 2. Boost converter cannot emulate greater impedances than the impedance of load, and therefore, it does not reach values near the open circuit voltage of the PV module. The value of input resistance is given by R i = R (1-d) 2. Buck-Boost derived converters are capable of sweeping the whole I-V curve of a module in CCM, from open circuit voltage (V oc ) to short-circuit current (I sc ). The input resistance of Buck-Boost converter is given by R i = R (1-d) 2 / d 2. But it has discontinuous input current which creates greater harmonic distortion factor in current. Hence, it is also not useful in practice. But Cuk converter is
53 capable of sweeping the whole I-V curve of a module in CCM, from open circuit voltage (V oc ) to short-circuit current (I sc ). Hence, it is inferred that Cuk converter is a good choice to track maximum power from solar PV module. 3.2 DESIGN CONSIDERATION OF CUK CONVERTER The relation between input and output currents and voltage is given in Equations 2.24 and 2.25. mode is given by The duty cycle of the Cuk converter under continuous conduction d = (3.1) V is the forward voltage drop across the diode (D). The maximum duty cycle is given by Equation (3.2). d max = ( ) (3.2) Equation The value of the inductor is selected based on the following L 1 = L 2 = L = ( ) (3.3) I is the peak-to-peak ripple current at the minimum input voltage and f s -switching frequency. The value of C 1 depends on RMS current which is given by I c1(rms) = I out * ( ) (3.4)
P switch = (I Q1(rms) *R DS (ON) *d max ) + ( Vin (min) +V out ) * I Q1(Peak * (Q GD * f S )/ I G (3.8) 54 The voltage rating of capacitor C 1 must be greater than the input voltage. The ripple voltage on C 1 is given by V c1 = ( ) (3.5) The parameters governing the selection of the MOSFET are the minimum threshold voltage V th(min), the on-resistance R DS (ON), gate-drain charge Q GD, and the maximum drain to source voltage, V DS(max). The peak switch voltage is equal to V in + V out. The peak switch current is given by I Q1(Peak) = I L1(PEAK) + I L2(PEAK) (3.6) The RMS current is given by I Q1(rms) = I out v + v ( ) ( ) ( 3.7) The total power dissipation for MOSFET includes conduction loss (as shown in the first term of the above equation) and switching loss as shown in the second term. I G is the gate drive current. The R DS(ON) value should be selected at maximum operating junction temperature and is typically given in the MOSFET datasheet. The output diode must be selected to handle the peak current and the reverse voltage. In a Cuk converter, the diode peak current is the same as the switch peak current I Q1(peak). The minimum peak reverse voltage that the diode must withstand is V RD = V in(max) + V out (max) (3.9)
55 Similar to the boost converter, the average diode current is equal to the output current. The power dissipation of the diode is equal to the output current multiplied by the forward voltage drop of the diode. Schottky diodes are recommended in order to minimize the efficiency loss. Table 3.1 shows the components used in simulation and hardware setup for the power circuit. Table 3.1 Specification of Cuk converter Input inductor L1 Filter inductor L2 Capacitor C1 Filter capacitor C2 500e -6 H 500e -6 H 220e -6 F 220e -6 F Resistive load R Switching frequency Switch : MOSFET Optocoupler Diode 25kHz IRF510 MCT2E MUR450 3.3 CUK CONVERTER-BASED MPPT FOR SOLAR PV SYSTEM Figure 3.1 shows a Cuk converter-based MPPT system consisting of solar PV module, Cuk converter (DC-DC) and load. Cuk converter is capable of sweeping the I-V curve of solar PV module in CCM from open circuit voltage to short-circuit current condition and, hence, Cuk converter is suitable to be employed in designing the MPPT circuits.
56 Figure 3.1 Cuk converter-based MPPT system The closed loop control is obtained by implementing the MPPT algorithm using micro-controller. In the PAO algorithm, the operating voltage of the PV array is perturbed by a small increment, and the resulting change in power, P, is measured. If P is positive, the perturbation of the operating voltage moves the PV array s operating point closer to the MPP. Thus, further voltage perturbations in the same direction (that is, with the same algebraic sign) should move the operating point towards the MPP. If P is negative, the system operating point moves from the MPP, and the algebraic sign of the perturbation should be reversed towards the MPP. To improve the dynamic response without affecting stability the Adaptive Perturb and Observer algorithm (APAO) is used. In APAO algorithm, high perturbation is selected when the operating point is far away from MPP and low perturbation is selected when the operating point is closer to MPP.
57 3.4 SIMULATION RESULTS The closed loop diagram was simulated in MATLAB /Simulink which is given in Figure 3.2 that includes the solar PV module electric circuit subsystem (MATLAB model), DC-DC Cuk converter, and adaptive PAO algorithm. Figures 3.3, 3.4 and 3.5 show the MATLAB models for solar PV subsystem, power sampling and PWM pulse generation using MPPT algorithm. PV module is modeled based on the electrical Equations (2.1) and (2.2) to provide voltage and current to the Cuk converter and the micro controller simultaneously. Using the adaptive PAO algorithm, the duty cycle is adjusted. High perturbation is selected when the operating point is far away from MPP and low perturbation is selected when the operating point is closer to MPP. When the obtained tracked power is equal or nearby actual maximum, the variation in the duty cycle is minimum in such a way that the memory increment value is selected. Figure 3.2 Cuk converter based MPPT system model in MATLAB
58 Figure 3.3 Solar PV module sub-system in MATLAB Figure 3.4 Power sampling circuit model in MATLAB
59 Figure 3.5 Pulse generation scheme using APAO MPPT algorithm Using APAO algorithm, the output is obtained in terms of pulses which may be used to trigger MOSFET. Also the changing irradiation is modeled to study the MPPT algorithm performances. The temperature is constant at 25 C and the illumination level is varying between two levels. Initial irradiation is set as 1000 W/m 2. After 0.02 sec, the irradiation (G) is suddenly changed to 500 W/m 2. The relationship between the duty cycle and tracked solar PV power is shown in Figures. 3.6 and 3.7 They show that the output power at G=1000 W/m 2 and 500 W/m 2 are 36.74 W and 17 W, respectively.
60 Figure 3.6 Change in duty cycle for various irradiation levels The duty cycle of the main switch of Cuk converter-based MPP tracking is 35.3%. The voltage and current through the main switch of Cuk converter is shown in Figure 3.8. Figure 3.7 Change in power for various irradiation levels
61 Figure 3.8 Voltage and current through the switch S of Cuk converter 3.4.1 Study of Effects of Partial Shading A shadow falling on a group of three cells will reduce the total tracked power by two mechanisms: 1) by reducing the energy input to the cell, and 2) by increasing energy losses in the shaded cells. Problems become more serious when shaded cells get reverse-biased. A group of cells under full illumination is connected in series with another group of cells under shaded illumination. Figure 3.9 shows the MATLAB-based equivalent circuit of solar module under partial shaded condition. The photon current of fully illuminated cells I sci is high compared with that of the shaded illuminated cells I scs. If the module current I < I scs, diode Ds is forward-biased and there is no risk for the shaded cells. But if I > I scs, then the diode current I Ds = I scs -I flows through the diode in the reverse direction. Reverse-biased diode Ds offers high resistance which will consume power and significantly reduce the load current (I) itself. The point B will assume negative potential. If the difference in illumination levels is high, Ds may get damaged due to overheating.
62 Figure 3.9 MATLAB-based equivalent circuit when three cells are partially shaded 3.4.2 Parameters Affecting the Performance of Solar PV Module under Partial Shading The photo current generated by the shaded illuminated cell is FI L (or I scs ), where F is the ratio of photo current generated by the shaded cell to that of the fully illuminated cell. F=0 means, fully shaded and F=1 means fully illuminated. When a solar cell in a series array is under shadow, its current output is given by I s = F I L - I o exp 1 - (3.10) where V = V s + R ss I and v - Voltage across the diode in shaded cell.
63 Similarly, the current through the illuminated cell is given by I i = I L - I o exp 1 - (3.11) where V = V s + R si I i and V - Voltage across diode in illuminated solar cell As the shaded and illuminated cells are connected in a series, the same current is forced to flow through both. Hence I s and I i are replaced by the same current I. Therefore, I = F I L - I o exp 1 - (3.12) I = I L - I o exp 1 - (3.13) As the value of F (illumination) decreases from 1 to 0, exp tends to reach zero. Hence, the equation can be simplified as I = F I L - I o - (3.14) The voltage across the shaded cell V s is given by V s = (F I L I ) R Shs I R ss (3.15) In the above equation, I o R shs is neglected in comparison with larger terms. The total module output voltage is the sum of voltages across each cell operating at the same current I. So the module consists of 36 identical series connected cells, the output voltage can be expressed
64 V = V + V (3.16) The power dissipated by the shaded cell is obtained by P= I * V s = I * (F I L I ) R shs I R ss (3.17) Power dissipation in the shaded cell may be substantial leading to increase in its temperature. Due to increased temperature, the cell current gets concentrated in an increasingly small region of the cell, producing the hot spot. The power extracted from the solar cell is reduced when the cells are shaded. The effectiveness of APAO MPPT algorithm to track the maximum power under partial shaded condition is simulated in MATLAB/Simulink Under partially shaded conditions. In APAO MPPT algorithm, large step size is selected when the maximum power point is faraway and small step size is preferred when the operating point is near to MPP. So the APAO MPPT tracks true maximum power. The effectiveness of MPPT algorithm is shown in Figure 3.10 Figure 3.10 Effectiveness of APAO MPPT under partial shaded condition
65 The APAO MPPT tracks maximum power of 37 W under unshaded condition with a voltage of 16.4 V. The tracked power from the solar PV module is lowered when solar cells are shaded. The voltage corresponding to maximum power under shaded condition is 15.6 V with the true maximum power of 33 W. The implemented MPPT algorithm effectively tracks maximum power from solar PV module under shaded and unshaded conditions. 3.5 STEADY STATE PERFORMANCE Figure 3.11 show the duty cycle of the main switch of Cuk converter against different load resistance at the constant irradiation (G) of 1000 W/m 2. Figure 3.12 shows that, for the load range of 1 R L 7, the obtained tracked power for the computer simulation. Figure 3.11 Variation in Duty cycle for the changes in load resistance
66 Figure 3.12 Variation in Tracked power for the changes in load resistance In Figures 3.13 and 3.14, the simulated tracked power from solar PV module by implementing the adaptive PAO algorithm on Cuk converter is illustrated. The obtained results are compared with the traditional PAO algorithm. It shows the effectiveness of the proposed algorithm, resulting in an improvement in the transient response. The settling time is reduced as 0.002sec. Figure 3.13 Tracked power using PAO algorithm
67 The tracking efficiency is 99.3% without considering the efficiency of the solar PV module and converter. The Cuk converter conversion efficiency is 86.266%. This is a good advantage during cloudy days when the working conditions change rapidly. Figure 3.14 Tracked power using Adaptive PAO algorithm 3.6 CONCLUSION The Adaptive Perturb and Observe algorithm with direct control was simulated using Cuk converter in MATLAB Simulink. The PI control loop was eliminated in MPPT system. This is the main advantage of the proposed tracking system when compared to direct control MPPT method which uses two control loops. The adaptive PAO algorithm improves steady state stability, dynamic response and tracks the maximum power of 36.74Watts with tracking efficiency of 99.3% in computer simulation without considering the efficiency of solar PV module / converter. The duty cycle of the main switch of the Cuk converter is 35.3%. The effectiveness of Adaptive Perturb and Observer algorithm is under shaded condition.