Research On Grid-Connected Control Strategy Of Single-Phase Photovoltaic Inverter

As the core part of the solar photovoltaic power generation system,the quality of the output current and the power generation efficiency of the whole system depend on the performance of the inverter and the stability and effectiveness of its control.Based on this,on the basis of two-stage single-phase photovoltaic inverter system,this paper focuses on two parts,one is the maximum power point tracking technology,the other is to analyze the grid-connected control strategy,and gives its own grid-connected control strategy.Finally,the feasibility of these two points is verified by simulation.Firstly,the traditional MPPT technology is theoretically studied,and it is verified through simulation comparison that the traditional MPPT technology is sufficient for systematic application in the case of no burst,but the photovoltaic panel encounters sudden weather conditions,such as in the local shadow,the traditional maximum power point tracking technology is easy to fall into the problem of local optimization.Based on this,this paper proposes an adaptive particle swarm algorithm MPPT.Quasi proportional resonant(Quasi-proportional resonant——QPR)control can be achieved in a single phase static coordinate system to astatic tracking,but there are big limitations of quasi PR controller parameters on,In this paper,a neural network algorithm is used to optimize the parameters of the quasi-PR controller,form a parameter regulator and combine it with the quasi-PR controller to form a neural network quasi-PR controller.The voltage outer loop adopts PI regulator,and the current inner loop adopts neural network quasi-PR controller to effectively improve the current control,better track the reference signal and increase the resonance suppression performance of the single-phase photovoltaic grid-connected inverter system.The current inner loop is added to the grid voltage full feedforward and capacitor current feedback,which effectively reduces the influence of grid-connected voltage on grid-connected current and suppresses resonant peaks.Based on this,this paper proposes a photovoltaic grid-connected control strategy based on PI and neural network quasi-PR dual-closed-loop control system.A simulation model is built on MATLAB/SIMULINK,and the superiority of the above control strategy is verified by simulation comparison.Finally,based on STM32F103C8T6,an experimental prototype of the photovoltaic part is built,and the experiment verifies that the adaptive particle swarm algorithm can quickly and accurately track the maximum power point under local shadow,which can improve the output and photoelectric conversion efficiency of photovoltaic panels.Based on STM32F103RCT6,an experimental prototype of full-bridge inverter part was built,and the effectiveness of the grid-connected control strategy was verified by experiments.Figure 100 Table 4 Reference 80...