CHAPTER 4 DC-DC CONVERTERS AND MAXIMUM POWER POINT TRACKING (MPPT) TECHNIQUES

Transcription

1 61 CHAPTER 4 DC-DC CONVERTERS AND MAXIMUM POWER POINT TRACKING (MPPT) TECHNIQUES 4.1 INTRODUCTION In PV power conversion system, the PV array can be connected to the load either directly or through a DC-DC converter. In a direct connection, the power absorbed by the load will be smaller than the maximum power available at a given illumination and temperature for an operating point on the non-linear I-V characteristic of the PV array. A DC-DC converter can be connected between the PV array and load to make the PV array to operate around the maximum power point under varying solar radiations and temperatures. The DC-DC converter can increase or decrease the output voltage of the PV system depending on the load requirements. Rajesh and Mabel (2015) charted the different types of DC- DC converters as in Figure 4.1. (Source: R. Rajesh and M. Carolin Mabel 2015) Figure 4.1 Types of DC-DC Converters for PV Applications

2 62 The DC-DC converters can either be of isolated type or nonisolated type. The boost converter, buck-boost converter, IBC and floating IBC are non-isolated type converters having simple topology, simple control techniques, compact size and low cost. However, its limitation is that there is need for a large size input inductor to limit the current ripple in the components with a high voltage gain. On the other hand, the isolated type DC-DC converters are suitable for low to medium power applications. The commonly used isolated type converters are half-bridge converters, full-bridge converters, Fly back converters and push-pull converters. The main features of isolated converters include high voltage ratio at the output, providing electrical isolation between the input and the output, improving converter efficiency and reducing the transformer size by increasing operating frequency (Kolli et al 2015). Among the entire DC-DC converters, the buck, boost and canonical switching cells are called as single inductor converters. The Cuk, SEPIC and Zeta converters are named as two inductor converters. 4.2 NON-ISOLATED TYPE DC-DC CONVERTERS A brief study of non-isolated DC-DC converters was carried out by Rajesh & Mabel (2015). Each converter has advantages and disadvantages and the selection depends on the requirement of the application for which the converters are used Buck Converter A buck converter (Figure 4.2) cannot emulate smaller impedances than the load impedance. Therefore, the output current does not reach values near the short circuit current of the PV module. It has a low output ripple

3 63 current and a high input ripple current. The average output voltage is lesser than the average input voltage of the PV array terminals. Figure 4.2 Buck Converter The buck converter has two modes of operation namely continuous conduction mode (CCM) and discontinuous conduction mode (DCM). When the switch is in ON condition and the diode is in OFF condition, the mode of operation is CCM and when the switch is OFF and the diode is ON, the mode of operation is DCM. For an ideal buck converter, the conversion ratio is given by equation 4.1 Vo = Is = R = D (4.1) Vs Io Rs Where the duty cycle (D) is the ratio of the time of conduction (T on ) to the switching period (T s ) as in Figure 4.3. Vo, Vs, Io, Is, R and Rs are the output voltage, input voltage, output current, input current, output resistance and input resistance of the buck converter. Buck converters are used in situations where array voltages are higher than load voltages.

4 64 Figure 4.3 PWM signals to control the output of DC-DC Converter Boost Converter Boost converter depicted in Figure 4.4 cannot emulate greater impedances than the impedance of the load. So it does not reach values near the open circuit voltage of the PV module. The output voltage will be greater than the input voltage. For an ideal boost converter Vo = Is = R = 1 (4.2) Vs Io Rs 1 D where Vo, Vs, Io, Is, R and Rs are the output voltage, input voltage, output current, input current, output resistance and input resistance of the boost converter. Boost converters are used in situations where array voltages are lower than the load voltage. Boost converters can be operated either in CCM or DCM. The boost converters can be controlled either by frequency control or duty cycle control. The control input parameters of the converters may be current, voltage and power.

5 65 Figure 4.4 Boost Converter Cascaded Converters The buck-boost characteristics can be obtained by the cascaded connection of the buck converter and the boost converter. The input-output voltage conversion is a multiple of the conversion ratios of the two converters in cascade. The switches of the two converters must have the same duty cycle Buck-Boost Converter The buck-boost converter can be formed by having buck converter as the first stage. The buck-boost converters shown in Figure 4.5 can sweep the whole range of the I-V curve in CCM from V oc to short circuit current I sc. For a buck-boost converter, the conversion ratio is Vo = Is = Vs IO D (4.3) 1 D where Vo, Vs, Io, Is, R and Rs are the output voltage, input voltage, output current, input current, output resistance and input resistance of the buckboost converter. Buck-boost converters are used in situations where array voltages and load voltages are nearly matched. This converter provides to

6 66 have either higher or lower output voltages compared to input voltages. The buck-boost converter operates through inductive energy transfer. The number of devices used is less in a buck-boost converter compared to a boost-buck converter. But the voltage stress in the switches will be high. It has nonpulsating current characteristics. The buck-boost converter steps down the voltage when D>0.5 and steps up the voltage when D<0.5. Figure 4.5 Buck- Boost Converter Boost-Buck Converter The boost-buck converter can be formed by connecting the boost converter as first stage converter as represented in Figure 4.6. The boost-buck converter operates through capacitive energy transfer. Figure 4.6 Boost- Buck Converter

7 67 The number of devices used will be higher than the Buck-Boost converters. But the voltage stress to the devices will be less Cuk Converter The Cuk converters have low switching losses and the highest efficiency. It can provide better output current characteristics due to the inductor in the output stage. The Cuk converter operates on CCM, discontinuous inductor current mode and discontinuous capacitor voltage mode. Figure 4.7 Cuk Converter For PV applications, the discontinuous capacitor voltage mode is preferred due to the low output voltage of the panel. The Cuk converter has high efficiency and low switching losses compared to other DC-DC converters. The output inductor maintains a continuous output current in the discontinuous mode. The relation between output and input voltage and currents of Cuk converters are given below. Vo = Is = Vs Io D (4.4) 1 D where Vo, Vs, Io, Is, R and Rs are the output voltage, input voltage, output current, input current, output resistance and input resistance of the Cuk

8 68 converter. The Cuk converter also steps down the voltage when D>0.5 and steps up the voltage when D<0.5.The series inductors in the input side and the output side of the Cuk converter, as depicted in Figure 4.7 reduce the current ripple in the input and the output circuits (Duran et al 2008) SEPIC Converter Figure 4.8 SEPIC Converter Figure 4.8 presents the circuit of the SEPIC converter which has non-inverting buck-boost characteristics. It has a simplified gate drive construction because of the grounding of the switch terminal. SEPIC converter has non-pulsating input current. It operates by the energy transfer between C 1 and L 1 and reduces the voltage stress in C 1 much lower than the Cuk converter. This converter provides better input-output isolation and have low input ripple and noise (Duran et al 2008). Vo = Is = D (4.5) Vs Io 1 D where Vo, Vs, Io, Is, R and Rs are the output voltage, input voltage, output current, input current, output resistance and input resistance of the SEPIC converter.

9 Zeta Converter Figure 4.9 Zeta Converter The Zeta converter is an inverse SEPIC converter because it can be formed by the interchange of power input ports and power output ports of SEPIC converter as given in Figure 4.9. This converter operates by the energy transfer between C 1 and L 1. It has non-inverting buck-boost characteristics. The pulsating input current is the drawback of Zeta converter. This converter provides better input-output isolation and has a low input ripple and noise. Vo = Is = D (4.6) Vs Io 1 D where Vo, Vs, Io, Is, R, Rs are the output voltage, input voltage, output current, input current, output resistance, input resistance of the Zeta converter(duran et al 2008) Canonical Switching Cell (CSC) Converter The CSC converter of Figure 4.10 forms the basic block for various converters. Adding of inductors L 1 and L 2 and removal of inductor L makes the CSC converter similar to the Cuk converter. It works on the energy transfer between L and C. The advantage of this converter is the usage of low number of devices. The CSC converter has no pulsating input current. Hence,

10 70 the voltage stresses in C is the same as in a Cuk converter and greater than in a SEPIC converter (Duran et al. 2008). Figure 4.10 CSC Converter Fly Back Converter The fly back converter of Figure 4.11 is also a buck-boost converter in which the inductor is split up to form a transformer. The fly back transformer provides isolation and also the voltage ratios are multiplied by the turn s ratio. Figure 4.11 Fly back Converter The fly back transformer module includes an inductance L m and an ideal transformer with a turn s ratio N 1 /N 2. The leakage inductance and losses

11 71 of the fly back transformer are neglected here. But the leakage inductance affects the switch and the diode transitions. Vo = Is = D Vs Io 1 D N2 N1 (4.7) where Vo, Vs, Io, Is, R and Rs are the output voltage, input voltage, output current, input current, output resistance and input resistance of the fly back converter. N 1 and N 2 represent the number of turns in the primary and secondary winding of the transformer. L m represents the mutual inductance. The magnitude of L m decides the boundary between CCM and DCM. The series connection of switch S 1 with the DC generator results in pulsating input current (Duran et al 2008). Though the various types of non-isolated DC-DC converters are discussed above, the commonly used converters are boost converters for PV applications. Hence the detailed analysis of boost converters is carried out in section ANALYSIS OF BOOST CONVERTER DC-DC boost converters are used at the output of the PV array with MPPT control to provide power to the load through the DC-AC inverter. Boost converters have simple topology, high power density, fast transient response and continuous input current. Therefore, boost converters are usually used in different power electronics applications such as active power factor correction (PFC), photovoltaic power systems and fuel cells Modes of operation of Boost Converter Boost converter operates in two modes i. Mode 1: CCM ii. Mode 2: DCM

12 72 DCM are: The conditions for operation of the boost converter in CCM and I > i L for CCM I < i L for DCM where I represents the inductor current DC component and i L represents ripple peak magnitude (Mohammed H. Rashid 2004). The modes of operation of boost converter are considered under the following assumptions: Switching devices are ideal The load current remains constant AC components in the input current have frequencies less than half of the switching frequency Mode 1: CCM operation In CCM, current flows continuously in the inductor during the entire switching cycle. It starts at zero, reaches a peak value, and returns to zero during each switching cycle. The DC-DC boost converter with ideal switching devices is depicted in Figure The equations for the variation of inductor current and capacitor voltage are obtained by using Kirchhoff s voltage and current law. The converter consists of a DC supply voltage V s, inductor L, switch S, freewheeling diode D, capacitor C and load resistance R. The converter produces an output voltage V o higher than V s.

13 73 Figure 4.12 Boost Converter Circuit CCM has two state of operation namely ON state and OFF state. In the ON state operation the switch S 1 is in ON state and the diode D remains reverse biased as in Figure The inductor current increases the energy stored in the inductor and the load requirement is met out by the output capacitor. Figure 4.13 Boost Converter CCM ON state operation The rate of change of inductor current is given by dil dt = 1 L Vs (4.8) dv dt = 1 C io Vo R (4.9)

14 74 Figure 4.14 Boost Converter CCM OFF state operation As illustrated in Figure 4.14 during OFF state operation, the switch S 1 is in OFF condition and the inductor current reduces linearly. Since the inductor current cannot change suddenly, the reversal in the polarity of the voltage occurs across the inductor and tries to maintain constant current. The current path now is L, D, C and R. The inductor current flows till the switch S 1 turns on by the next gate pulse (Abdullah Abusorrah 2013). The equations for the inductor current and capacitor voltage variations during OFF state operation are given by equations 4.10 and dil dt = 1 L Vs Vo (4.10) dv dt = 1 C il ir Vo R (4.11) The gate triggering signal, output voltage variation, diode current and inductor current are depicted in Figure 4.15 for CCM operation of boost converter.

15 75 Figure 4.15 Boost Converter CCM waveforms The formulae for the design of capacitor and inductor values of boost converter are specified in Equations 4.12 and Components values calculation: C = DV o V o Rf (4.12) L = 1 D 2 RD 2f (4.13) From equation 4.2 it can be understood that in CCM, the voltage conversion only depend on the input voltage and duty cycle and the average inductor current follows the output current, i.e., if the output current decreases, then the average inductor current also decreases. Further the minimum and maximum peak inductor current track the average inductor current exactly Mode 2: DCM operation For DCM there are three states, the ON state during switch S 1 ON and diode D is OFF as in Figure 4.16, the OFF state during switch S 1 OFF

16 76 and diode D is ON as in figure 4.17 and the idle state during which both switch S 1 and diode D are in OFF state as in figure Figure 4.16 Boost Converter DCM ON state operation Figure 4.17 Boost Converter DCM OFF state operation Figure 4.18 Boost Converter DCM IDLE state operation

17 77 The waveform for DCM operation is illustrated in Figure The duration DT represents ON state operation, the switch S 1 conducts and diode D will be in off state, duration D 1 T for OFF state operation during this period the switch S 1 will be in off state and diode D conducts and the duration D 2 T is for idle state operation now both switch S 1 and the diode D will be in off state as air viewed in Figure For DCM, the voltage conversion relationship is a function of the input voltage, duty cycle, power stage inductance, switching frequency, and output load resistance as given by equations 4.14 and vo vi = D 2 K 2 (4.14) K = 2 L R T S (4.15) Figure 4.19 Boost Converter DCM waveforms

18 Simulation of Boost Converter Figure 4.20 Block diagram of Boost Converter Simulation Model The block diagram of PV system employing DC-DC boost converter for impedance matching with load is described in Figure 4.20 and the MATLAB simulation model of the boost converter is illustrated in Figure Figure 4.21 Boost Converter Simulation Model

19 Output Voltage (V) Time (Sec) Figure 4.22 Boost Converter Simulation output (D=0.8) Table 4.1 Variation of Boost Converter output voltage with duty cycle S. No Input voltage (V) Duty ratio (D) Output Voltage (Vo) From the simulation output of boost converter shown in Figure 4.22, it can be seen that the rising time of the boost converter output voltage is sec and settling time is 0.04 sec. The variation in the output voltage with the variation of duty ratio is noted in Table 4.1, which shows that an increase in duty ratio increases the output voltage of the boost converter.

20 Boost Converter Hardware Model A prototype model of the boost converter is constructed with the following hardware specifications as in Table 4.2 and its output voltage and current waveforms revealed in Figures 4.23 and Boost Converter Component Specifications Table 4.2 Hardware specifications of Boost Converter Sl. No Components Specifications Quantity 1 Inductor (L) 540mH 1 2 Capacitor (C) 1000µF 1 3 IGBT IRG15UD Diode 6N Load 1000watts 10A Hardware Results of Boost Converter Figure 4.23 Boost Converter output Voltage

21 81 Figure 4.24 Boost Converter Inductor Current The boost converter input voltage is 24V, output voltage is 90 volts and current is 5A as in Figure 4.23 and the inductor current waveform of the boost converter is shown in Figure Draw Backs of DC-DC Boost Converters To provide high output voltage, a DC-DC converter needs to be operated at an extreme duty cycle which subjects the switching devices to short pulse, high amplitude current, which leads to reverse recovery and Electro Magnetic Interference (EMI) problems and the extreme duty cycle leads to poor dynamic response for line and load variations. Converters with a coupled inductor can provide a high output voltage, less switching voltage stress without extreme duty cycle. But the energy leakage losses in the coupled inductor reduce the efficiency of the converter (Sungsik Park et al. 2011). Resonant or quasi-resonant converters can be used as DC-DC converters. The voltage stress in the switching devices is high for high input DC voltage applications. Passive snubber with Zero Voltage Switching (ZVS)

22 82 control is simple and minimum cost controller. The complexity of circuit topology is difficult to analyze. Auxiliary active snubber circuit helps to reduce the switching losses. For synchronizing the auxiliary switch with the main switch and to trigger the auxiliary switch, additional circuit elements are necessary, which causes switching losses in the auxiliary switch. To overcome these difficulties and to improve the performance of the boost converter, the interleaving technique can be used (Yao-Ching Hsieh et al. 2009). 4.4 BENEFITS OF INTERLEAVED BOOST CONVERTERS For high current and high power applications, DC-DC interleaved boost converters are used which improve the performance in terms of efficiency, size, conducted electromagnetic emission and transient response analysis. Phatiphat Thounthong & Bernard Davat (2010) found that the major advantages of interleaved converters are: Reduction in the size of passive components like inductor and capacitor. Ripple component in the input current and output voltage is reduced. Input and output waveforms ripple frequency increased. Parallel operation of phases of power converters increases reliability. Parallel operation provides large surface area for heat dissipation.

23 83 Power electronic components of reduced current rating can be used. The current rating is proportional to the power rating of the circuit. (Source: Benyahia et al. 2014) Figure 4.25 IBC duty ratio Vs current ripple Benyahia et al (2014) plotted the relationship between the input current ripple and duty ratio of boost converter, two phase interleaved boost converter (IBC-2) and three phase interleaved boost converter (IBC-3) as given in Figure As the number of phases increases the input current ripple reduces and frequency of the ripple current increases. Mounica Ganta et al. (2012) explained that the interleaved converters can be operated in critical conduction mode, CCM and DCM. In critical conduction mode, the critical operating point changes with the change in load which makes the design of the converter difficult. In DCM, the reverse recovery effect will direct to the high input current and the conduction losses in the devices, which is unsuitable for high power applications.

24 84 Hence, CCM is the choice of operation for high power applications. The gate pulses applied to the devices of n-phase interleaved converter is phase shifted by an angle of 360/n degrees for each phase. For a two phase IBC, the phase shift will be 360/2=180 degrees. In interleaved converters the components of the phases should be identical. However in the analysis both CCM and DCM operations were considered and analysed Analysis of Interleaved Boost Converter Figure 4.26 IBC and its Switching waveforms The parallel connection of converters reduces the device stress, increases flexibility and device tolerance. The assumptions made for interleaved operation are: 1.Switching is ideal, 2. Capacitors equivalent resistance is neglected and 3. The two switches operate in interleaved mode. The circuit arrangement and switching pulses are depicted in Figure Modes of Operation of Interleaved Boost Converters Interleaved converter has two modes of operations Mode 1: CCM operation Mode 2: DCM operation

25 85 conduction angle The mode 1 operation can be further divided into 4 stages based on stage 1 from 0-δ 1 stage 2 from δ 1- π stage 3 from π- δ 2 stage 4 from δ 2-2π conduction angle The mode 2 operation can be divided into 6 stages based on stage 1 from 0-δ 1 stage 2 from δ 1 -δ 2 stage 3 from δ 2- π stage 4 from π- δ 3 stage 5 from δ 3 - δ 4 stage 6 from δ 4-2π (Huiqing Wen & Bin Su 2016) Mode 1: stage 1 from 0-δ 1 Figure 4.27 IBC Mode 1 stage 1 operation

26 86 Mode 1 stage 1 operation switching status is represented in Figure Now S 1 is on the current in the inductor L 1 increases and current in L 2 starts to decrease by flowing through the output capacitor C since S 2 is off. The rate of change of current in inductor L 1 is I L1 = di L1 dt = Vs L 1 (4.16) The rate of change of current in inductor L 2 I L2 = di L2 dt v s = L 1 di L 1 dt = Vs Vo L 2 (4.17) (4.18) v s = L 2 di L 2 dt + v 0 (4.19) i diodes = i D2 = i c + i 0 = i L2 (4.20) = C d v 0 dt + v 0 R (4.21) i s = i L1 + i L2 (4.22) d v 0 dt = i L2 c v 0 CR (4.23) Mode 1: stage 2 from δ 1 - π As shown in Figure 4.28 since S 1 and S 2 are off the current in both inductor L 1 and L 2 starts to flow through the output capacitor C and decreases as per the equations 4.24 ad 4.25.

27 87 Figure 4.28 IBC Mode 1 stage 2 operation The rate of change of current in inductor L 1 is i L1 = dil 1 Vs Vo = (4.24) dt L 1 The rate of change of current in inductor L 2 is i L2 = dil 2 Vs Vo = (4.25) dt L 2 V s = L 2 di L 2 dt v s = L 1 di L 1 dt + v 0 (4.26) + v 0 (4.27) i dc = i d1 + i d2 (4.28) = i c + i 0 = i L1 + i L2 (4.29) = C d v 0 dt + v 0 R (4.30) I s = i L1 + i L2 (4.31) dv 0 dt = i L1 +i L2 C v 0 CR (4.32)

28 Mode 1: stage 3 from π-δ 2 Now current in L 1 starts to decrease by flowing through the output capacitor C. Since S 1 is off and S 2 is on the current in the inductor L 2 increases. The devices conduction status is shown in figure Figure 4.29 IBC Mode 1 stage 3 operation The rate of change of current in inductor L 1 is i L1 = dil 1 = dt Vs Vo L 1 (4.33) The rate of change of current in inductor L 2 is i L2 = dil 2 = Vs (4.34) dt L 2 di L 2 dt di L 1 dt = v s v 0 L 2 (4.35) = v s v 0 L 1 (4.36) di L 1 dt = i L1 + i L2 C v s = L 2 di L 2 dt v 0 CR (4.37) (4.38)

29 89 v s = L 1 di L 1 dt + v 0 (4.39) i dc = i d1 = i c + i 0 = i L1 (4.40) = C d v 0 dt + v 0 R (4.41) i s = i L1 + i L2 (4.42) dv 0 dt = i L1 C v 0 CR (4.43) Mode 1: stage 4 from δ-π This mode is similar to mode 2. The waveform for CCM operation of interleaved boost converter is illustrated in Figure Figure 4.30 IBC Mode 1 CCM waveforms

30 Comparison of Energy stored in the Inductor of Boost Converter and IBC Energy stored in Boost converter inductor = 1 2 LI L 2 (4.44) Energy stored in the interleaved Boost converter inductor E interleaved = 1 2 L 1 IL L 2 IL 2 2 = 1 2 L 1 I L 2 Here L 1 =L 2, hence the equation becomes L I L = LI L 2 (4.45) Hence energy stored in the inductor of interleaved converter will be four times less than the energy stored in the boost converter Formulae for Design of IBC Boost ratio Vo = 1 Vs 1 D (4.46) Inductor current peak-to-peak Inductor value Capacitor value I L1, I L2 = VsD L f s (4.47) L = VsD f I L (4.48)

31 91 C = VoDf R V o (4.49) Mode 2 DCM operation Mode 2: stage 1 from 0-δ 1 Figure 4.31 IBC Mode 2 stage 1 operation In this stage the switch S 1 conducting and inductor current i L1 is increasing from zero to the maximum value. Diode D 2 is conducting to follow the current il 2 of inductor L 2 as in Figure Voltage conversion ratio d = V 2 V 1 (4.50) The maximum inductor current I L1 = DV 2 dl s f s (4.51) Mode 2: stage 2 from δ 1 -δ 2 In this stage of operation Diode D 2 is in conduction and it follows current i L1. At the angle δ 2, i L2 decreases and reaches zero. The voltage across L 1 is negative and the current i L1 decreases linearly. The value of the voltage across L 1 is given by

32 92 V L1 = V 2 V 2 = V 2 d d (d 1) (4.52) Figure 4.32 IBC Mode 2 stage 2 operation Mode 2: stage 3 from δ 2 -π Figure 4.33 IBC Mode 2 stage 3 operation In stage 3 operation of Figure 4.33 current i L2 remains as zero. The boost circuit incorporating the inductor L 2, diode D 2 and switch S2 is idle. Diode D 1 is conducting to follow the inductor current i L1. Here M represents the nonzero current period.

33 93 V 2 DT d s = d 1 V 2 (0.5 + M + D)T d s (4.53) M = 2dD d+1 2d 2 (4.54) Mode 2: stage 4 from π-δ 3 Figure 4.34 IBC Mode 2 stage 4 operation In this stage the diode D 1 and the switch S 2 will be conducting. The current levels of the inductor L 1 and L 2 are represented in Figure Mode 2: stage 5 from δ 3 -δ 4 Figure 4.35 IBC Mode 2 stage 5 operation

34 94 In stage five operation both the switches S 1 and S 2 will be in off state and the diodes D 1 and D 2 will be conducting as illustrated in Figure Mode 2: stage 6 from δ 4-2π Figure 4.36 IBC Mode 2 stage 6 operation In stage 6 operation current il 1 remains as zero. The boost circuit incorporating the inductor L 1, diode D 1 and switch S 1 is idle. Diode D 2 is conducting to follow the inductor current i L2 as depicted in Figure Figure 4.37 gives the illustration of switch conduction status, inductor current flow, diode current flow and source current flow waveforms. Figure 4.37 IBC Mode 2 DCM waveform

35 Simulation of Interleaved Boost Converter The block diagram of PV system using interleaved boost converter as impedance matching device between PV source and load are depicted in Figure 4.38 and the simulation of IBC is represented in Figure Figure 4.38 Block diagram of IBC Simulation Model Figure 4.39 IBC Simulation Model

36 Output Voltage (V) IBC output Time (Sec) Figure 4.40 IBC Simulation output (D=0.8) Table 4.3 Variation of Interleaved Boost Converter output voltage with duty cycle S. No Input voltage (V) Duty ratio (D) Output Voltage (Vo) From the simulation output of interleaved boost converter plotted in Figure 4.40, it can be seen that the rising time of the interleaved boost converter is sec and settling time is 0.03 sec. The variation in the output voltage with the variation of duty ratio is noted in Table 4.3, which shows that an increase in duty ratio increases the output voltage of IBC.

37 IBC Hardware Model A prototype model of the boost converter is constructed with the following hardware specifications as in Table 4.4 and its output voltage and current waveforms revealed in Figures 4.41 and Table 4.4 Hardware Specifications of IBC S. No Components Specifications Quantity 1 Inductor (L) 270 mh 2 2 Capacitor (C) 1000µF 1 3 IGBT IRG15UD Diode 6N Load 1000 watts 10A Hardware Results for IBC Figure 4.41 IBC output Voltage

38 98 Figure 4.42 IBC Inductor Current waveform The IBC input voltage is 24V, output voltage is 90 volts and current is 5A as in Figure 4.41 and the inductor current waveforms of the interleaved boost converter is shown in Figure Comparison of Boost and IBC in Terms of Duty Cycle and Output Voltage Table 4.5 Variation of output voltage with duty cycle for Boost Converter and IBC S. No Input Duty ratio IBC converter Boost converter voltage (V) (D) Output Voltage (V o ) Output Voltage (V o )

39 99 The variation of output voltage with the variation of duty ratio for both boost converter and interleaved boost converter are tabulated in Table 4.5 and the same is plotted in Figure It can be seen that the output voltage of IBC is higher than that of boost converter. Figure 4.43 Duty ratio verses output voltage of Boost and IBC Table 4.6 Comparison of Boost and Interleaved Boost Converters in terms components values and operating time Parameters Boost converter IBC Rising time(sec) Settling time(sec) Switching frequency (KHz) 25KHz 25KHz Inductor value.9µh.2µh The boost and IBC are compared in terms of rising time, settling time and inductor values in Table 4.6 and it was clear that the IBC has lesser rising time, settling time and inductor value compared to boost converter.

40 NEED FOR MPPT TECHNIQUES The power transfer from PV array to AC load is carried out through power conditioning systems. Since the efficiency of the solar panel is poor, the power conditioning system should have high efficiency. The power conditioning system includes two conversion stages. A DC-DC converter with maximum power point tracking constitutes the first stage and a DC-AC inverter supplying power to the load constitutes the second stage (Marimuthu et al. 2012). The power output of the PV system can be increased either by increasing the solar irradiation using sun trackers or by MPPT methods. Taghvaee et al. (2013) observed that the direct connection of the PV output to the load extracts only 31% of the solar energy whereas incorporation of MPPT tracks the maximum power of the PV array under all atmospheric conditions and transfers up to 97% of the maximum power generated by the PV array to the next stage. Thus, implementation of the MPPT technique increases the output power from PV array by 30-40% on a clear sunshine day. Figure 4.44 Block diagram of PV System with MPPT control This can be achieved by the utilization of advanced power electronic devices and by the implementation of suitable converter topologies. In DC-DC conversion stage, it is necessary to step up the voltage for most of

41 101 the applications. For this purpose, boost converter is the optimum choice. Figure 4.44 represents the basic block diagram of the PV power generation system Maximum Power Point Tracking (MPPT) Techniques Keeping the operating point of the PV panel at the maximum power point for the corresponding radiation is known as MPPT, which could be achieved by changing the duty cycle of the power electronic converters in the PV system. Mrityunjaya Kappali & Udaya Kumar (2012) discussed that in conventional MPPT methods, the Interruptive type (offline) employs constant voltage control method or constant current control method, which necessitates the measurement of open circuit voltage or short circuit current by delinking the panel from the power electronics converters, which leads to the loss of output power. In non-interruptive (online) type, such as P&O method, modified P&O method and IC method, the power calculation was carried out by the measurement of voltage and current by using sensors, which makes the system a little bit complicated. The duty cycle of the DC-DC converters were adjusted till the maximum power was reached. In this method, there is no need for delinking of the panel from the converters, and hence no loss in the output power. Ali Reza Reisi et al. (2013) underwent a review and found that the soft computing techniques like FLC, adaptive fuzzy logic control, neural network control and Ant Colony Optimization (APO) are flexible, fully digitized and effective in handling non-linear problems and are suitable for the MPPT implementations of PV system. Among these soft computing techniques, FLC can be implemented using simple, less expensive microcontrollers.

42 MPPT Techniques Literature Survey Though there are several MPPT techniques, the most commonly used techniques are P&O, MPPT and Fuzzy MPPT in most of the research articles. The random analysis carried out from IEEE papers for the year 2012, suggest that out of 27 papers, P&O and modified P&O MPPT used in nine papers, Fuzzy MPPT in three papers, incremental conductance in two papers, fuzzy combined with other conventional methods in four papers and all other MPPT techniques in nine papers, were used. From the survey it was found that the simplest and the most commonly used conventional type of MPPT was P&O MPPT and the soft computing type MPPT was fuzzy logic MPPT. Hybrid MPPT controls were formulated by combining the conventional and soft computing techniques for obtaining better performance. Lotfi Khemissi et al. (E MPPT ) technique by (2012) defined the efficiency of MPPT E MPPT = t P 0 actual t dt t P 0 maximum t dt (4.55) Where P actual - measured power of PV system, P maximum - measured maximum power for a given irradiance and temperature P&O MPPT Technique In this algorithm, the array terminal voltage and current were sensed and processed and power output was calculated. The present PV output power was compared with the power of previous perturbation cycle as shown in the flow chart of Figure 4.46.

43 103 The PV voltage and current is perturbed periodically after comparing. If the PV operating voltage varies and change of power is greater than zero (dp/dv>0), the algorithm moves the operating point in the same direction, and if the change of power is less than zero (dp/dv<0), the algorithm moves the operating point in the opposite direction and if the change of power is equal to zero (dp/dv=0) that represents the condition of the maximum power point. The operation of P&O algorithm can be summarized in Table 4.7. Figure 4.45 MPPT Convergence Table 4.7 P&O MPPT duty cycle Change Condition Operating point position Duty cycle change P(k)>0,V(k)> 0 Left of MPP Increased P(k)>0,V(k)<0 Left of MPP Decreased P(k)<0,V(k)>0 Left of MPP Decreased P(k)<0,V(k)<0 Left of MPP Increased

44 104 The advantage of P&O algorithm is a previous knowledge of the PV array characteristics is not required and it is a simple procedure. But the oscillation of the operating point around MPP under steady stated conditions as shown in Figure 4.45 and the inability to respond to the rapidly changing atmospheric conditions are the drawbacks of this system (Rajkumar & Manoharan 2013). Subudhi & Pradhan (2013) suggested that fixing the correct perturbation size is most essential for the correct traction of MPP. Figure 4.46 Flow Chart for P&O MPPT Technique

45 Simulation of P&O MPPT Technique The MATLAB simulation of P&O MPPT technique with IBC was carried out and shown in Figure Figure 4.47 Simulation Diagram of P&O MPPT algorithm Figure 4.48 Simulation output of P&O MPPT at Constant Illumination

46 106 Figure 4.49 Simulation output of P&O MPPT under Varying Illumination The simulation of P&O MPPT was run for both constant illumination and varying illumination conditions. From the graphs plotted, it was seen that for an illumination of 1000W/m 2 the rising time was.001sec and the settling time was 0.025sec as shown in Figure For varying illumination of 500W/m 2, 750W/m 2 and 950W/m 2, the response of the system was plotted in Figure Fuzzy Logic System (FLC) FLC provides an automatic control algorithm by using linguistic variables which may take any value between 0 and 1. Esram & Chapman (2007) investigated that FLC algorithm does not require an accurate mathematical model; hence, the uncertainties such as non-linear operating characteristics and unexpected changes in the operating point can be dealt easily and are therefore more suitable for handling non-linear problems (Algazar et al 2012). Dorin Petreus et al. (2010) and Bendib et al. (2014) carried out a comparison of P&O MPPT and Fuzzy MPPT using MATLAB simulation and found Fuzzy MPPT tracking to MPP is better than P&O for

47 107 dynamic conditions of radiations. Though FLC does not have high accuracy in finding the MPP, it presents the advantage of being the fastest to track the MPP. Figure 4.50 Block diagram of Fuzzy System The FLC system, as demonstrated in the block diagram of Figure 4.50, has three processing stages namely fuzzification, rules inferences and defuzzification. It has a rule Table to store the fuzzy rules and the calculations of FLC are performed by rules inference. The system consists of two linguistic input variables namely error (E) and change in Error (CE) and one output variable duty cycle (D) Fuzzification The input variables of the FLC are divided into five subsets namely positive big (PB), positive small (PS), zero (Z), negative small (NS) and negative big (NB).The partition of the fuzzy subsets and shape of the membership functions are triangular in shape. The input error and CE values are normalized by an input scaling factor such that the input values lie between -1 and 1. To get only one dominant output for a particular input, the triangular shape of membership functions was chosen. The error (E) in Figure 4.51 and change in error (CE) in Figure 4.52 are calculated using the following formulae:

48 108 P k = V k I(k) (4.56) E k = P k 1 P(k) (4.57) CE k = I k 1 I(k) P k 1 P(k) (4.58) Where V (k) - Voltage of the PV panel I (k) - P (k) - E (k) - Current of the PV panel Power of the PV panel Error value CE (k) - Change in error value Figure 4.51 Input variable membership Function Figure 4.52 Change in Error Membership Function

49 109 Figure 4.53 Output Membership Function The voltage and current of the PV panel are measured and the power is calculated. The values of the error (E) and CE are calculated and used by the Fuzzy Inference System (FIS). The inputs are processed by a fuzzy set of rules. The output of the inference system is subjected to defuzzification. The defuzzified output in Figure 4.53 is the change in the duty cycle. The change in the duty cycle is given to the DC-DC converter to track the maximum power point of the PV panel Inference method The Mamdani fuzzy inference method is used in this analysis. Though several composition methods such as MAX-MIN and MAX-SOT are in the fuzzy tool box, the commonly used MAX-MIN method is used in this analysis also. The rule Table for the FLC is given in Table 4.7. Table 4.8 Rule Table of Fuzzy MPPT E \CE NB NS ZE PB PS NB ZE ZE PB PB PB NS ZE ZE PS PS PS ZE PS ZE ZE ZE NS PB NS NS NS ZE ZE PS NB NB NB ZE ZE

50 Defuzzification The FIS output is a fuzzy set. But non-fuzzy value is necessary to control the DC-DC converter. For that, many defuzzification methods are available such as the centroid method, mean of maximum method, first of maxima method and last of maxima method. The centroid method is used in this analysis. The formula for defuzzification using the centroid method is given by the equation D = n j=1 μ Di (Dj) n j=1 μ(dj ) (4.59) Where D is the duty ratio. The major problems using fuzzy logic system are its slow transient response and fluctuations in the output power Simulation of Fuzzy MPPT Technique The MATLAB simulation of Fuzzy MPPT technique with IBC was carried out and depicted in Figure Figure 4.54 Simulation Diagram of Fuzzy MPPT algorithm

51 111 Figure 4.55 Simulation output of Fuzzy MPPT at Constant Illumination Figure 4.56 Simulation output of Fuzzy MPPT under Varying Illumination The Fuzzy MPPT simulation was run for both constant illumination and varying illumination conditions. From the graphs plotted, it was seen that for an illumination of 1000W/m 2 the rising time was Sec and the settling time was 0.015Sec as shown in Figure For varying illumination of 500W/m 2, 750W/m 2 and 950W/m 2, the response of the system was plotted in Figure 4.56.

52 Proposed System (P&O Fuzzy MPPT Technique) Kamarzaman & Tan (2014) reviewed that the conventional MPPT techniques are very efficient for constant solar irradiance conditions and for partial shading conditions stochastic-based MPPT procedures like particle swarm optimization, genetic algorithm, artificial neural networks and fuzzy logic controllers are used. Hence, to get a better MPPT technique, the fuzzy logic control can be entrenched with P&O MPPT technique to change the step size of the voltage such that the MPP tracking can be carried out faster under rapidly changing atmospheric conditions. Previous observations confirm that even for 50% shading, the fuzzy P&O MPPT tracks speedily compared to P&O, which locks in the local peak points. Figure 4.57 Flow Chart for P&O Fuzzy MPPT technique

53 113 The proposed system is a combination of P&O and Fuzzy system, which possess the boon of minimum settling time and reduced oscillations. FLC can be implemented with simple and less expensive controllers like PIC 16F872 and PIC 16C74 and Tricore TC1796 controllers (Salam et al. 2013). The flow chart of the hybrid system is shown in Figure The hybrid system is formed by the two simple and effective methods namely P&O and fuzzy methods. This system is blessed with less settling time, thus improving the performance of the system (Sadek et al & El Khateb et al. 2013) Simulation of P&O Fuzzy MPPT Technique Figure 4.58 Simulation Diagram of P&O Fuzzy MPPT algorithm

54 114 Figure 4.59 Simulation output of P&O Fuzzy MPPT at Constant Illumination Figure 4.60 Simulation output of P&O Fuzzy MPPT under Varying Illumination As in figure 4.58, the simulation for P&O Fuzzy MPPT was run for both constant illumination and varying illumination conditions. From the graphs plotted, it was seen that for an illumination of 1000W/m 2 the rising time was sec and the settling time was sec as shown in Figure

55 For varying illumination of 500W/m 2, 750W/m 2 and 950W/m 2, the response of the system was plotted in Figure Comparison of P&O, Fuzzy and P&O Fuzzy Techniques The performances of P&O, Fuzzy, and P&O Fuzzy MPPT techniques were compared by using MATLAB simulation of figure The results for comparison were obtained by running the simulations under the same illumination levels and the same running time for the simulation Simulation of P&O, Fuzzy and P&O Fuzzy MPPT techniques for performance comparison Figure 4.61 Simulation Diagram of P&O, Fuzzy and P&O Fuzzy MPPT algorithm

56 116 Figure 4.62 Simulation output of P&O, Fuzzy and P&O Fuzzy MPPT at Constant Illumination Figure 4.63 Simulation outputs of P&O, Fuzzy and P&O Fuzzy MPPT under Varying Illumination From the simulation results shown in Figures 4.62 and 4.63 proves that the P&O fuzzy MPPT shows better performance than P&O and Fuzzy MPPT Techniques.

57 117 Table 4.9 Comparison of P&O, Fuzzy and P&O Fuzzy MPPT Techniques in terms of Rising and Settling time S. No Type of MPPT Rising time (Sec) Settling time (Sec) 1 P&O MPPT Fuzzy MPPT P&O Fuzzy MPPT From the simulation output waveforms for both static and dynamic conditions, it can be seen that hybrid P&O Fuzzy MPPT have better performance compared to P&O and Fuzzy MPPT techniques by having minimum rising time of.005 sec and settling time of 0.01 sec. 4.6 CONCLUSION In this chapter, the overview of different types of non-isolated DC- DC converters was carried out. The modes of operation of boost converter and its performance with different duty cycles was carried out by MATLAB simulation. The modes of operation of IBC and its performance with different duty cycles were also carried out by MATLAB simulation. From the analysis, it was found that IBC performance is better than boost converter performances. Both converters were designed for 14V input with a duty ratio of The hardware implementation of boost and IBC were carried out and the performance was analyzed. In the second half of the chapter, the need for MPPT techniques and the different types of MPPT techniques were discussed. Then, P&O MPPT, Fuzzy MPPT and P&O Fuzzy MPPT techniques algorithms were discussed, simulated and the outputs were compared and it was concluded that the P&O Fuzzy MPPT technique with IBC provides better result.