1 Introduction
Traditional centralized, string-type grid-connected photovoltaic (PV) power generation systems connect PV panels in series and parallel to effectively increase the bus voltage before supplying power to a grid-connected inverter to transmit electricity to the grid. Their simple structure and high conversion efficiency make them particularly suitable for power plant systems with good sunlight. However, in eastern urban and rural areas, cloud cover, building and tree obstructions, and single panel failures can severely reduce the overall system's power output. Micro-inverters, installed behind each PV module, independently control each module to operate at its maximum power point, significantly improving the system's resistance to localized shading and overall power generation. Although their cost is relatively high, their modular architecture, high reliability, high power generation, and ease of installation make them an important direction for distributed PV power generation.
This paper details the design, analysis, and control strategy of a certain type of quasi-single-stage interleaved parallel microinverter. High-frequency inverter technology not only achieves a large boost ratio matching between the microinverter's input and output voltages, but also addresses the leakage current problem in non-isolated systems where primary-secondary electrical isolation cannot solve the issue. Furthermore, based on active clamping technology to absorb leakage inductance energy, it achieves ZVS (Zero Voltage Switching). The system control block diagram and flowchart demonstrate that the variable-step-size perturbation-observation method can achieve MPPT (Multi-Level Perturbation Test), and the input voltage feedforward method can solve the bus voltage collapse problem in quasi-single-stage microinverters.
2. Main Circuit Topology
2.1 Topology Selection
The quasi-single-stage flyback inverter has only one stage of power conversion [4] , and its topology is simple, making it particularly suitable for low-cost applications. In discontinuous mode (DCM) and critical continuous mode (BCM), it exhibits current source characteristics, and the control system design is simple, making it the ideal topology for current photovoltaic micro-inverters. Since the output power of the flyback converter is limited, the interleaved parallel connection technology shown in Figure 1 is adopted in the micro-inverter system structure: the inputs of the two flyback converters are connected in parallel, the outputs are connected in parallel, and the main tubes of the primary side are interleaved by 180 degrees to reduce the input and output current ripple. At the same time, a common set of output polarity switching bridges is used. Considering the existence of the leakage inductance of the flyback transformer, active clamping technology is further adopted to recover the leakage inductance, and ZVS of the main tube and auxiliary tube is realized, which effectively reduces switching losses and improves circuit efficiency.
Figure 1. Topology of interleaved parallel flyback microinverter
At this point, the photovoltaic module undergoes high-frequency SPWM modulation by the flyback converter's main switch, resulting in an output current with a unipolar power frequency sinusoidal half-wave envelope. The AC-side power frequency commutator bridge timing track tracks the grid voltage, flipping the preceding unipolar power frequency sinusoidal half-wave into a sinusoidal grid-connected current, which is in phase and frequency with the grid voltage.
2.2 Working Mode Analysis
Based on whether the transformer's magnetic flux is continuous, flyback converters can be classified into three operating modes: Continuous Inductor Current (CCM), Direct Current Current (DCM), and Basic Current Current (BCM). Flyback inverters in CCM mode have relatively poor stability and require careful handling. Currently, mainstream flyback inverters primarily use DCM and BCM. However, since frequency conversion control is required in BCM mode, making calculations and control more complex, DCM is used here. Compared to BCM and CCM, DCM's advantages include constant frequency operation, simple control, and elimination of the secondary diode reverse recovery problem; its disadvantage is that the magnetizing inductance is smaller than in CCM, resulting in higher peak current stress on the devices.
To ensure the converter operates in DCM ( Direct Current Management), its primary inductance Lp (magnetic inductance) must be less than the critical continuous inductance value. The power frequency period Tgrid is defined as 2k times the high-frequency switching period, and dp is defined as the maximum duty cycle. Since the input current is proportional to the duty cycle, the duty cycle for each switching cycle is also a sinusoidal pulse dp sin( iπ/k ). Therefore, the average value of the transformer primary current idc is:
Simplify to get
Substituting P <sub>in</sub> = U <sub>dc </sub> * I<sub> dc</sub>,avg into the above equation, we can obtain the primary inductance of the transformer:
3 Control System
3.1 Control Block Diagram
Quasi-single-stage microinverters need to simultaneously perform MPPT, phase-locked loop (PLL), islanding detection, and grid-connected current control [5][6] . As shown in Figure 2, the reference amplitude Io of the grid-connected current is calculated by MPPT to ensure that the photovoltaic modules transmit energy to the grid at maximum power. PLL provides phase information of the grid-connected current to ensure that the grid-connected current is in phase and frequency with the grid voltage. Islanding detection is a necessary function of grid-connected inverters. In case of grid abnormality, the inverter is shut down to ensure the safety of personnel and equipment. Grid-connected current control is the core control part of the grid-connected inverter. Here, high-quality grid-connected current is ensured by sampling the output current closed-loop control (theoretically, under DCM, open-loop control can realize grid connection of the current source, but its total harmonic content of the grid-connected current is relatively high).
Figure 2 Control System
3.2 Quasi-single-stage system MPPT and DC bus voltage control
MPPT uses a specific algorithm to continuously adjust the grid-connected current reference and the inverter output power, thereby regulating the output power of the photovoltaic modules to maximize their output power.
The perturbation-observation method is simple in principle and easy to implement, and is one of the most commonly used methods in the MPPT algorithm. Its algorithm principle is to compare the output power of the current time with the output power of the previous time. If P(k+1) > P(k) , then the photovoltaic output voltage reference will continue to be perturbed in the same direction as this change. Conversely, if the output power decreases, the direction of perturbation will be changed in the next cycle. This process of perturbation and comparison will be repeated until the output power of the photovoltaic system reaches its maximum. The algorithm flow is shown in Figure 3. The step size of the perturbation-observation method determines the tracking speed of the algorithm and the amplitude of the system oscillation near the highest point. Therefore, this paper adopts a perturbation-observation method with a variable step size [7] . Specifically, when the power is relatively small, the perturbation value C is increased; when the power is relatively large, the perturbation value C is appropriately reduced.
Figure 3. Algorithm flow of the perturbation-observation method
In quasi-single-stage grid-connected inverter systems, a simple MPPT loop cannot guarantee good dynamic performance and achieve system stability. When sudden changes in external conditions or program misjudgments occur, the DC bus voltage will oscillate violently or even collapse. As shown in Figure 3, adding an input voltage loop to the original control system can prevent violent oscillations of the DC bus voltage during MPPT misjudgments, effectively preventing bus voltage collapse and achieving stable system operation.
4 Experimental Results
To verify the above-mentioned interleaved parallel quasi-single-stage high-frequency photovoltaic grid-connected micro-inverter scheme, a 220W micro-inverter prototype based on DSP28035 control was developed in the laboratory. The front-end DC input voltage Vpv = 35VDC, the grid voltage Vo = 220VAC, the grid frequency fac = 50Hz, the main switch V1 switching frequency fs = 135kHz, and the filter inductor L1 = 1mH. The photovoltaic modules and AC grid were simulated using a photovoltaic simulator and an AC power supply. Figures 4a and 4b show the output waveforms of the grid-connected current io under light and full load conditions, respectively. It can be seen that io and ug are in phase and frequency, and the waveform quality of io is good. As shown in Figure 5c, the drain-source voltage of V1 is zero before switching on and off, achieving ZVS for V1. Figure 4e shows the waveforms of the transformer primary voltage up, secondary voltage us, and currents is and ug, verifying the feasibility of the power frequency switching bridge.
(a) Light load output
(b) Full load output
(c) Main switch waveform
(d) Clamping tube waveform
(e) Transformer primary and secondary voltage waveforms
Figure 4 Experimental waveforms
Figure 5 shows the MPPT effect tested by the photovoltaic simulator, with an MPPT efficiency of 99.5%.
Figure 5. IU and PU curves
Figure 6a shows the efficiency test curve, which further demonstrates that the micro-inverter achieves high efficiency across the entire load range, with a maximum full-load efficiency of 94%. Figure 6b shows the power analyzer test results without considering auxiliary power loss, with a maximum efficiency of 95% and a grid-connected current THD of only 1.5%, verifying the feasibility of the micro-inverter solution.
Figure 6 Efficiency curve and THD test
5. Conclusion
This paper introduces the design, analysis, and control strategy of a quasi-single-stage interleaved parallel microinverter. This microinverter features the following characteristics: based on high-frequency link inverter technology, it effectively achieves primary-secondary electrical isolation, solving the leakage current problem in non-isolated systems; it employs active clamping technology to absorb leakage inductance energy, achieving zero-voltage switching of the switching transistors and reducing switching losses; it uses a variable-step perturbation observation method to achieve maximum power point tracking, and solves the bus voltage collapse problem of the quasi-single-stage microinverter based on the input voltage feedforward method; the 220W prototype achieves a maximum power point tracking efficiency of 99.5%, with a maximum full-load efficiency of 94%. Without considering auxiliary power supply, the highest efficiency is 95%, and the total harmonic distortion of the grid-connected current is less than 1.5%.
Source: Power Electronics Technology, Issue 6, 2014
References
1. Zhang Xing, Cao Renxian. Solar Photovoltaic Grid-Connected Power Generation and Its Inverter Control [M]. Beijing: Machinery Industry Press, 2011.
2. Woo-Young Choi, Jason Lai. High-efficiency grid-connected photovoltaic module integrated converter system with high-speed communication interfaces for small-scale distribution power generation[J]． Solar Energy, 2010, 84(4): 636–649
3. Chen Daolian. DC/AC Inverter Technology and Its Applications [M]. Beijing: Machinery Industry Press, 2003.
4. Ji Baojian, Wang Jianhua, Zhao Jianfeng. A high-efficiency H6 structure non-isolated single-phase photovoltaic grid-connected inverter [J]. Proceedings of the CSEE, 2012, 32(18): 9-15
5. Gu Junyin, Wu Hongfei, Chen Guocheng, Xing Yan. Soft-switching interleaved anti-laser grid-connected inverter [J]. Proceedings of the CSEE. 2011, 31(36):40-45
6. Wang Lu, Research on Micro Photovoltaic Grid-Connected Inverters [D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2012.
7. D.Lalili, A.Mellit, N.Lourci, et al. Input output feedback linearization control and variable step size MPPT algorithm of a grid-connected photovoltaic inverter[J]. Renewable Energy, 2011, 36(12):3282-3291.
