Research On Control Strategies For Grid-connected Photovoltaic Generation System Based On Dynamic Impedance Matching And MFAC

Currently, the photovoltaic power generation with great prospects has become a research hotspot in recent years. In this thesis, a single-phase grid-connected PV control system is regarded as the research object. Deep researches on the problems such as the imprecise of tracking accuracy, the difficulty of building the accurate model, the complexity of controller design and unmodelled dynamics in the maximum power point tracking part, DC/DC and DC/AC converters are made. Then some improved control algorithms are proposed and a power electronics semi-physical simulation platform is designed to verify the practicality of the control algorithm.An improved dynamic impedance matching algorithm is proposed in this thesis because of the limitations in tracking speed and steady state accuracy for the traditional maximum power point tracking algorithms. The dynamic impedance matching error is taken as the target to judge and adjust.The proposed algorithm has been compared with the traditional methods and the existing methods, the results show that the proposed algorithm is help to reduce blindness and improve the tracking performance.Because DC/DC and DC/AC converters of the PV system are very difficult to establish the precise mathematical model, and the traditional model-based control methods have the shortcomings such as model mismatch, unmodelled dynamics and so on, so a data-driven model-free adaptive control (MFAC)method with tight format is proposed in this paper and is used to power electronic converters.The comparison study among state feedback linearization control, PID control and MFAC shows that MFAC is easy to design like the traditional control, and its accuracy control effect is the same as the modern methods, what’s more important is that it can effectively overcome model mismatch.In order to verify the effectiveness and practicality of the proposed method, a power electronics semi-physical simulation platform is designed. The semi-physical simulation platform can implement the online optimization and adjustment of controller parameters, and it can quickly verify the actual application effects of control algorithms.The superiority and practicability of the control algorithm is validated by using Simulink simulation and power electronic platform in this thesis.