The demand of electric power is on rise and is anticipated to increase exponentially in future. Among main sources of energy, thermal sources such as fossil fuels contribute to bulk amount of power generation. But these fuels are expensive due to their bulky demand around the world. Also their reserves are limited and running short. Besides this, their combustion generates pollutants gases which affect the environment in various forms. Due to these factors, the renewable sources of energy are getting attention and the photovoltaic systems (PV) are also becoming more prominent during the last few decades. The photovoltaic module provides a low DC voltage with wide range according to various operating conditions   . A boost converter is required to step up this low DC voltage to the required DC link voltage. The conventional boost converters cannot offer such a high voltage gain. Even in case of extreme duty ratio, the conversion efficiency is declined due to losses. This also increases the ratings of devices like output diode and other issues such as severe electromagnetic interference and reverse recovery  . The quadrature converters also offer a high voltage gain, but these have equal voltage stress across the switch   . In particular cases, the line frequency transformer can also be used, but it increases the size and weight of photovoltaic system   . The grid-connected photovoltaic systems also need the inverters for power conversion, grid interconnection and control optimization   . Pulse width modulation (PWM) voltage source inverters are most commonly used in PV systems to invert DC voltage into AC before feeding to the grid or AC loads. The connection of inverters to the grid specifically is dependent on the control scheme which is adopted by these inverters. Different control schemes have been under research for grid-connected inverters in the past decade   . The phase locked loop technique is being the conventional. There are other PLL based grid synchronization techniques like power based PLL, quadrature PLL (QPLL). A PLL based new technique for synchronization with the grid has also been developed in  .
This paper describes grid-connected photovoltaic system with a proposed high voltage conversion ratio DC-DC converter. This system mainly includes two processing stages: a high voltage conversion ratio converter which converts low DC PV module voltage into the required DC link voltage. The second stage is the inverter which converts it into AC voltage. A low pass filter is designed to filter the inverter output before it is applied to the grid. The DC-DC converter is capable of maintaining the magnitude of inverter’s output voltage and inverter guarantees the phase and frequency synchronization of its output. The proposed system finds applications not only in low power, but can also be extended to some extent to large scale allowing the parallel operation of PV modules.
The grid-connected system is shown in Figure 1 with the main constituents as proposed topology of DC-DC converter, inverter and a low-pass filter. As the PV module normally generates low voltage, so a boost converter is needed to step up this DC voltage to a higher value. This output of DC-DC converter is applied to an H-bridge inverter which has four MOSFET switches S1 to S4. The discrete pairs of switches (S1, S4) and (S2, S3) are operated to produce positive and negative half cycles of AC voltage respectively. For the operation of inverter, the pulse width modulation method is used in which a sinusoidal pulse width modulated signal is achieved by the comparison of a sine wave and saw-tooth wave. This sinusoidal PWM signal is then compared with a square wave to get driving signals for the switches (S1, S4). Similarly, the sinusoidal PWM signal is compared with inverted square wave to get driving signals for the switches (S2, S3). The output of inverter comprises of ripples and is filtered out using an LC filter as shown in Figure 1.
Figure 1. The grid-connected system.
2. Operation of the Proposed Conveter
The proposed converter is shown in Figure 2 which includes two stages. An inductor L1, capacitor C1, switch Sa and diode D1 constitute the first stage. A coupled inductor and capacitors C2, C3 and C4 form the second stage. Switches Sa and Sb are used to control the operation of first and second stage respectively.
2.1. Operation of First Stage
When switch Sa is turned on, L1 inductor is storing the energy. During this mode, diode D1 is reverse biased and capacitor C1 supplies current to the second stage which acts as a load. When switch Sa is turned off, diode D1 becomes on and the current iL flows through the load and capacitor C1. These operating modes are shown in Figure 3.
The voltage conversion ratio of first stage is same as boost converter and is given by
where, and are the output voltage, input voltage and duty ratio of first stage respectively.
2.2. Operation of Second Stage
Figure 4 shows the second stage of the proposed topology. In order to accomplish mode analysis of second stage converter, coupled inductor is displayed as Lm (magnetizing inductance), leakage inductances Lk1 and Lk2 on the primary and secondary side respectively and a lossless transformer having n as turn ratio. The switch Sb controls the operation of this stage. The steady state waveforms of
Figure 2. Proposed DC-DC converter.
Figure 3. Topologies of first stage (a) Sa is on (b) Sa is off.
Figure 4. Second Stage of proposed converter.
Mode 1: The switch Sb is initially turned on as shown in Figure 6(a). D2 becomes reverse biased and the applied voltage causes the current on primary side to increase. In this mode, the current flowing through secondary winding is zero and there is increase in energy on the primary side.
Figure 5. Steady state waveforms of the second stage.
Mode 2: In second mode, switch Sb is turned off and diode D2 is forward biased which charges capacitor C3. The secondary winding is also receiving some of primary current via capacitor C2 during this mode. The energy in magnetizing inductance decreases and there is increase in secondary current. Diode D3 remains in off condition. Capacitor C3 becomes charged at the end of this mode.
Mode 3: In this mode, switch Sb remains off and D2 becomes reverse biased. Primary, secondary windings and capacitor C2 are series connected across the source. This allows the direct transfer of energy to the output from capacitor C2, coupled inductor and source.
Mode 4: The switch Sb is turned on and current on the primary side starts increasing there by storing energy in magnetizing inductance. Diode D3 becomes forward biased and capacitor C2 is charged by capacitor C3. The current on secondary winding decays to zero and D4 becomes reverse biased. The voltage across C3 becomes equal to the voltage across C3 causing diode D3 to become reverse biased at the end of forth mode and the circuit proceeds to its initial condition.
Some assumptions are made for the derivation of voltage conversion ratio. The coupling co-efficient of coupled inductor is unity and the diodes are ideal. Let, , and represent the voltages across magnetizing inductance, secondary winding, input and output of second stage respectively. is the duty ratio of second stage.
During on and off modes the inductor Lm voltage is respectively given by:
Figure 6. Topologies of second stage (a) Mode 1; (b) Mode 2; (c) Mode 3; (d) Mode 4.
Application of inductor volt-second balance to Equation (2) and Equation (3) yields:
Generally the relation between primary and secondary voltages is given the following equation:
The inductor Lm voltage can be written from Equation (5) in the following form:
During Mode 2, the voltage across capacitor C3 is given by the following equation:
Substitution of Equation (6) into Equation (7) leads to the following equation:
As capacitor C2 is charged by capacitor C3 and at the end of mode 4, so
During Mode 3, apply KVL
This is the voltage gain of second stage and the voltage conversion ratio of proposed topology is obtained by the product of voltage gains of both the stages i.e.; the product of Equation (1) and Equation (10) yields:
Equation (11) is used to find the voltage conversion ratio of the proposed DC-DC converter. This equation illustrates that for a selected value of turn’s ratio n, the required DC link voltage can be achieved at lower values of duty ratios.
3. Grid Connected Inverter
In order to synchronize the phase and frequency of inverter voltage with the grid, a signal is taken as reference from grid voltage. This feedback signal is converted into square wave in a comparator as shown in Figure 7. The positive and negative half cycles of the grid voltage appear as high and low voltage respectively at the output of the comparator. The rising and falling edges of this wave correspond to the start of positive and negative half cycle of grid sine wave respectively. The square wave actually determines the phase and frequency of the output of H-bridge inverter. The comparison of sinusoidal PWM signal for half of the period (10 ms) with same square wave generates gate drive signals for one pair of switches (S1, S4). Similarly the comparison of sinusoidal PWM signal with inverted square wave for half of the period generates gate drive signals for the other pair of switches (S2, S3). The transition from low to high in square will make switches S1 and S4 to turn on and S2 and S3 to turn off. Accordingly, the inverter voltage changes its polarity at exactly the same time as the grid voltage. Any shift in grid voltage is followed by the square wave generated accordingly. The same shift is adopted by the gate drive signals and hence the same shift occurs in the output voltage of inverter making it in phase with connected grid.
The inverter in this design operates at constant modulation index (m) and the magnitude of grid voltage and its output voltage should be identical. This is accomplished by changing the duty ratio of DC-DC converter switches (Sa, Sb) by a µ-controller. The grid voltage is stepped down by a simple voltage divider circuit and a measured value is given to the µ-controller which reads the input voltage () of converter and the voltage of grid (). The µ-controller calculates the desired output voltage () of DC-DC converter using the grid voltage by the equation  .
Figure 7. Frequency and phase synchronization.
Equation (11) is used to compute the duty ratio (d). Ton and Toff are computed from the duty ratio. The frequency of switching signal is taken to be 20 KHz and two timers are used to produce switching signals of this selected frequency and duty ratio for the operation of DC-DC converter.
4. Simulation Results
The system specifications and its design parameters are given in Table 1. The proposed topology has been validated by performing its simulation. The results demonstrate a good accordance with the analysis. Figure 8 shows the required voltage of 350 V is achieved at considerable low duty ratio and the output is almost ripples free. The converter has been simulated for the lowest value of input voltage, Vin = 30 V, still it operates excellently to give the required voltage at low value of ~0.45. Figure 8 also demonstrates that the proposed converter has less voltage stresses across the switches. The switch Sa has less voltage stress which is about 60 V. The voltage stress across Sb is less than 110 V which is far less than
Figure 8. DC-DC converter output.
Table 1. System specifications and parameters.
the output voltage (350 V). These less stresses allow the selection of switches in the design procedure to have low voltage ratings. This lowers the cost of the proposed PV system. Comparison has been made been made in Figure 9 among the proposed topology, simple boost, flyback, cascaded boost flyback and another high voltage gain converter presented in  . This comparison has been made for the same input voltage. The voltage conversion ratio of cascaded boost flyback convert can be increased by selecting the higher value of turn’s ratio. But this will increase the size of the system. Figure 9 shows that the voltage conversion ratio (M) of proposed topology is very high in comparison with the other converters.
The required DC voltage can be achieved at low duty ratio of ~0.4 in the proposed converter. The lower value of duty ratio decreases the conduction losses and hence improves the efficiency of the proposed system to some extent. The H-bridge inverter has been also simulated. Figure 10 shows the simulated results of inverter’s output. The output of inverter is square wave and not suitable for most of the applications, so a low pass filter has been designed and simulated. This filter eliminates most of the ripples from the inverter’s output. The output of low pass LC filter is also shown in Figure 11. This simulation result shows that the output voltage is almost sinusoidal having low ripples. This can be used for most of the electrical loads. This output is applied to grid to meet the present day demand of energy.
This paper proposes the grid-connected photovoltaic system with a proposed topology of DC-DC converter. The converter with a high voltage conversion ratio can easily boost up the lower PV voltage to the required DC voltage at low duty ratio. As the gain is high, this allows the parallel operation of PV modules and also eradicates the need of power transformer to achieve the required grid
Figure 9. Comparison of DC-DC converter with other topologies.
Figure10. Inverter output voltage.
Figure 11. Inverter output using filter.
voltage. This results in high efficiency and reduced size of this system. The converter has low voltage stresses across the switches with decreased cost of the system. The system is beneficial in terms of unit saving as it can meet the demand of load. It may be implemented for a three-phase system.