Research On Mode Tracking Of Grid-connected Inverter Based On Hybrid System Theory

The research subject is supported by the National Natural Science Fund :Research on Intelligent Fault Diagnosis Method of the Inverter in Wind Power Generation System"(Project Number 61364010). It is urgent for development and application of new energy, especially the renewable energy, facing with the problem of energy shortage. In order to connect to the grid, the requirements of dynamic control performance of inverter are increasing. So the research for grid-connected wind power inverter has been paid more and more attention.Mode tracking for grid-connected inverter is to track and identify the system’s operation modes, which can achieve the good real-time tracking results. So the dynamic performance and stability of the system based on the hybrid system control theory can be improved. The purpose of the research on mode tracking methods for grid-connected inverter is to solve the practical problems of difficult fault diagnosis which is caused by the complex and variable structure characteristic of the grid-connected wind power generation inverter system.According to the above discussion, this subject studies the theory and methods for high power inverter based on model in the view of energy to realize the modeling and simulation, mode tracking and identification of the grid-connected inverter based on hybrid system theory. The hybrid bond graph model of the high power grid-connected inverter for the wind power generation system is established and complete tracking and identification the discrete modes of the grid-connected inverter based on the Global Analytical Redundancy Relations and the Gauss-Newton algorithm. The problems of operation modes tracking of high power grid-connected inverter, the offline study of components’ fault detectability and isolability for the grid-connected inverter based on the mode tracking technology and the online accurate identification and location for the components’ structural failures are resolved.This paper broadens the idea of mode tracking and fault diagnosis for inverter based on hybrid system and bond graph theories and also lays a foundation for intelligent fault diagnosis for grid-connected inverter.