Research On Grid-connected Control Of Photovoltaic Power Generation System Based On Fuzzy Neural Network

With the consumption of traditional energy and the continuous deterioration of the environment,the demand for new energy at home and abroad has reached an unprecedented height.With the development of photovoltaic storage and gridconnected power generation technology,the photovoltaic power generation system has gradually become an important branch of the power system.Taking advantage of the high energy density of the energy storage unit,combining the photovoltaic system with the energy storage unit not only improves the problem of unbalanced photovoltaic power,but also acts as a voltage regulator for the DC bus.The photovoltaic grid-connected inverter,as the core component in the gridconnected process,can improve the problem of poor power quality during the grid-connected process by optimizing the control method.Therefore,focus on the research on the voltage regulation performance of the energy storage unit and the fuzzy neural network control algorithm of the grid-connected inverter,to make up for the lack of photovoltaic power during the grid-connection process,and to achieve high-performance,stable and reliable grid-connected operation.The overall research of the article is as follows:Firstly,starting from the principle of photovoltaic power generation,establish its related mathematical model through equivalent circuit,and do corresponding simulation analysis of its photovoltaic characteristics,and expound the basic principle of Maximum Power Point Tracking(MPPT)in photovoltaic cells.,and compared and analyzed the three commonly used tracking control algorithms,built a simulation model for analysis,and verified the superiority of the disturbance observation method.Secondly,an energy storage system with nickel-hydrogen battery as a unit is established,the equivalent circuit model of nickel-hydrogen battery is analyzed,and the DC/DC conversion circuit is adopted,and the energy management of the energy storage system based on constant voltage control and battery SOC droop control is designed.The control strategy is used to control the energy absorption and release of the energy storage system,which verifies the reliability of the control strategy.Finally,for the grid-connected inverter in normal mode,the optimization algorithm of fuzzy neural network is added to the traditional double closed-loop control of voltage and current.Centralized optimization and self-learning to predict the optimal PID control parameters,so that the grid-connected voltage and current can be fast at the same frequency and phase.It can be seen from the results that the accuracy and response speed of the designed control algorithm are better.