Design of high-efficiency power converter based on renewable energy system

1 Introduction

The development of the global economy has brought about its side effect - the increasing depletion of energy. Various energy sources, especially petrochemical energy, are the most scarce energy sources. In addition to the ecological damage caused by the greenhouse effect, countries have actually long considered the sustainable development of energy. They are not only worried about their own energy problems, but also about the energy problems of future generations. Therefore, the world has reached a consensus on environmental protection and energy regeneration.

The power system established in some clean energy sources needs to use the city grid or use batteries as a backup source, which cannot provide electricity all the time, and cannot make up for the shortcomings of intermittent power generation. Therefore, among many clean energy sources, solar energy has become the most environmentally friendly and sustainable power generation mode, which can effectively replace the current oil energy. Therefore, using solar energy or fuel cells as the initial power provider of the power generation system has become the focus of the country's energy strategy development.

Although the development of new energy sources such as solar cells and fuel cells has become a consensus and has a deep understanding of the importance of everyone. However, in addition to using renewable energy as the initial power, the entire power generation system must also use energy efficiently. As shown in Figure 1, a schematic diagram of a solar power generation system, solar panels absorb light energy and convert it into electrical energy, and then convert the DC voltage they output into DC and AC power through a DC-DC power converter and a DC-AC power inverter to supply different electrical appliances and equipment. Therefore, power conversion technology is an important link among many key technologies in the renewable energy power generation system, and they need to be improved and complement each other.

Power conversion technology has become a fairly mature technology after decades of efforts by researchers and engineers. However, in the past, a large proportion of power conversion needs were to convert 400V DC high voltage power after rectification and power factor improvement circuits, or to use batteries as raw power for low-power applications. Both power conversion applications are easy to handle because their input side is a low current specification, and the conduction loss of semiconductors or other components is relatively low. On the other hand, the input voltage provided by the renewable energy power generation system is a low DC voltage, which needs to be converted into a high-output DC or AC voltage in order to completely replace the current fossil energy power generation system in the future and provide a stable power for most of the currently used equipment. Taking the power system of solar cells and fuel cells as an example, the power conversion of the latter stage needs to deal with a low input voltage with a large range of variation. Assuming that the voltage generated is 16~24V, if the required output power is 5kW, under the minimum input voltage of 16V working conditions, there will be an input current of more than 300 amperes. As long as there is a resistance value of one milliohm on the power conversion transmission line, there will be a loss of more than 90 watts. In addition to reducing the conversion efficiency, the heat dissipation process is the main factor affecting space requirements, reliability and cost. To handle the above-mentioned high current requirements, a single power converter can be used in parallel with multiple semiconductor switching elements or a multi-phase (Multiphase) parallel connection of multiple power converters can be used to achieve the goal of high efficiency and high power density power conversion requirements. Therefore, the challenge facing power conversion technology in this application should be how to improve power conversion technology so that each power converter can more efficiently handle technical problems derived from high step-up ratios, large input currents and high output voltages.

Therefore, this paper proposes a new low output current ripple boost power converter (Boost Converter with Ripple Reduction, BCRR) circuit architecture. Basically, this circuit retains the advantages of this type of circuit because it is a current-fed type. At the same time, it can improve the many challenges faced by low-voltage-high-voltage power converter circuits under high input current and high output voltage working conditions, thereby achieving the design goals of high efficiency and high power density. In addition to introducing the working principle of this power converter circuit architecture, this paper also conducts a prototype circuit experiment with 16-24V low input voltage, 200V output voltage and 320W output power as electrical specifications to verify that this circuit architecture can improve the high withstand voltage and current stress of the components, reduce the current ripple of the pulsating waveform on the high-voltage output side, and improve the efficiency of power generation systems with low voltage as the initial power such as solar energy and fuel cells.

2 Circuit Working Principle

Figures 2 and 3 show a novel low output current ripple boost power converter circuit and its main voltage and current waveforms proposed in this paper. The circuit consists of an input inductor Li, a transformer T1, two semiconductor switch elements Q1-Q2, a clamping capacitor C1, an output capacitor C0, and two pairs of rectifier diodes D1-D2-D3-D4 connected in series. The transformer has two sets of windings P1-P2 on the primary side, two sets of windings S1-S2 on the secondary side, and two sets of secondary leakage inductances marked as L1-L2. The turns ratio of each group is P1:P2:S1:S2=1:1:N:N.

To simplify the circuit analysis, it is assumed that: all semiconductor components are ideal; the input inductor Li is large enough and can be regarded as an ideal current source; the clamping capacitor C1 and the output capacitor C0 are large enough and can be regarded as an ideal voltage source; the leakage inductance L1=L2.

The working principle of this circuit can be divided into four time intervals, as shown in Figure 4 (a)-(d).

(a) T0-T1

As shown in Figure 4 (a), the gate control signal VGS1 is applied to the semiconductor switch element Q1 at T0. Therefore, the semiconductor switch elements Q1 and Q2 are turned on at the same time, and the two windings P1-P2 on the primary side of the transformer are short-circuited, resulting in the primary input voltage across the input inductor Li, which is in a charging state, and the inductor current rises linearly. On the secondary side, because the rectifier diodes D1-D4 cannot obtain the forward bias for conduction, they are all in the off state. At this time, half of the load current is provided by the output capacitor C0, and the other half is provided by the clamping capacitor C1 through the C1 (+)-S1-L1-R-S2-L2-C1 (-) path. Since the clamping capacitor can share the load current required in this time interval, the current ripple of the output capacitor can be reduced to half of the load current. Therefore, a smaller output capacitor can be selected. In addition, because the secondary windings have opposite polarities, the voltages across the two windings cancel each other out, so that the average voltage of the clamping capacitor is clamped to be equal to the output voltage value V0.

(b) T1-T2

As shown in Figure 4 (b), the gate control signal VGS2 is removed at T1. During this time interval, the sum of the primary input voltage and the inductor voltage across the primary winding P1 of the transformer transmits most of the input power to the load through the secondary winding S1 of the transformer and the rectifier diode D1-D2 path. At the same time, part of the input power also charges the output capacitor C0 and the clamping capacitor C1 through the S1-L1-C0-D2-D1-S1 and S2-D2-D1-C1-L2-S2 paths respectively. At this time, the diodes D3-D4 are clamped at the output voltage value V0 due to the conduction of D1-D2.

(c) T2-T3

As shown in Figure 4 (c), the gate control signal VGS2 is applied to the semiconductor switch element Q2 at T2. Therefore, the semiconductor switch elements Q1 and Q2 are turned on at the same time, and the two windings P1-P2 on the primary side of the transformer are short-circuited, resulting in the primary input voltage across the input inductor Li, which is in a charging state, and the inductor current rises linearly. On the secondary side, because the rectifier diodes D1-D4 cannot obtain the forward bias for conduction, they are all in the off state. At this time, half of the load current is provided by the output capacitor C0, and the other half is provided by the clamping capacitor C1 through the C1 (+)-S1-L1-R-S2-L2-C1 (-) path. Since the clamping capacitor can share the load current required in this time interval, the current ripple of the output capacitor can be reduced to half of the load current. Therefore, a smaller output capacitor can be selected. In addition, because the secondary windings have opposite polarities, the voltages across the two windings cancel each other out, so that the average voltage of the clamping capacitor is clamped to the output voltage value V0.

(d) T3-T0

As shown in Figure 4 (d), the gate control signal VGS1 is removed at T3. During this time interval, the sum of the primary input voltage and the inductor voltage across the primary winding P2 of the transformer transmits most of the input power to the load through the secondary winding S2 of the transformer and the rectifier diode D3-D4 path. At the same time, part of the input power also charges the output capacitor C0 and the clamping capacitor C1 through the S1-C1-D3-D4-L1-S1 and S2-L2-D3-D4-C0-S2 paths respectively. At this time, the diodes D1-D2 are clamped at the output voltage value V0 due to the conduction of D3-D4.

From the discussion in the previous section, within half a working cycle, there is a time interval for storing energy and transferring energy. The length of each time interval of Tcharge and Ttransfer can be calculated according to the following formula:

In addition, from the volt-second balance of the transformer, the voltage gain of this circuit can be derived as follows:

The duty cycle, D, should be greater than 50%, and the transformer turns ratio can also be calculated using the following formula:

3 Experimental results show

In order to verify the new low output current ripple boost power converter proposed in this paper, a prototype circuit experiment will be carried out with 16-24V low input voltage, 200V output voltage and 320W output power as specifications to verify the feasibility and electrical performance of this circuit architecture. Its specifications, main selected components and parameters are listed in Table 1.

Figures 5 (a) and 5 (b) show the main voltage and current waveforms of the proposed new low output current ripple boost power converter under high input voltage, light load working conditions and low input voltage, full load working conditions. The third and fourth waveforms in each country are VDS1 and VDS2 voltages, both clamped at 2V0/N (53V). Although the waveforms of diodes D2 and D4 are not shown. However, through the waveforms of the fifth and sixth groups, it can be read or calculated that D1-D4 are clamped at V0 respectively.

According to the circuit analysis results, this circuit has the advantage of low output current ripple, and this feature can be verified from the experiment. Figure 6 shows the experimental waveform under the working conditions of low input voltage 16V and full load 320W. The fourth group of waveforms shows that the output voltage ripple meets the electrical specification of 0.5V. The third and fifth groups of waveforms are the current waveforms IO1 and ICO of the clamping capacitor C1 and the output capacitor C0, respectively, showing that in the input inductor energy storage stage, the output capacitor only needs to provide half of the output current with the assistance of the clamping capacitor. Therefore, the output capacitor value can be selected as 68uF/450V; compared with the traditional full-wave rectification with a center tap, if 330uF/450V must be used to meet the same specifications, it obviously has the advantages of reducing space requirements and reducing costs.

Figure 7 shows the efficiency measurement data of the power stage circuit under different input voltages and different load currents. Due to the application of high input current and high output voltage of this circuit, the efficiency is lower due to increased conduction loss at low input voltage. The maximum efficiency is obtained when working at the highest input voltage and 3/4 full load current, which can reach 92.19%.

4 Conclusion

In view of the broad application prospects of renewable energy power systems such as solar energy and fuel cells in reality, this paper proposes a new type of low current ripple boost power converter. Since this circuit uses the boost function provided by the input inductor, a transformer with a smaller turns ratio (15:2=7.5) can be used to achieve the electrical specification of high output voltage gain (200/16=12.5); as the turns ratio decreases, the stray inductive reactance and capacitive reactance generated by the transformer winding are reduced, thereby improving the performance of the power converter. At the same time, because the clamping capacitor has both the leakage inductance energy recovery and the diode voltage clamping function, the rectifier diode does not generate voltage surges due to the leakage inductance of the secondary side of the high-voltage transformer, so a low-voltage diode can be selected to reduce conduction losses; and because the output current ripple is reduced, a lower-value output capacitor can be selected to save space and cost.