Study On Maximum Power Point Tracking Of Photovoltaic System Using BP Neural Network Optimized By PSO Algorithm

With the over exploitation of fossil energy such as oil, coal, natural gas, etc, we are facing the problem of energy shortage. So, all the countries pay more attention to the development of renewable energy. As a kind of clean energy, solar has more advantages than other renewable energy. In recent decades, along with the development of science and technology, photovoltaic (PV) system and related industries have become part of the fastest-growing industries. The researches on PV system become more and more important. At present, the promotion and application of PV system are hampered because of low conversion efficiency and high price of PV cells. Therefore, it is necessary to make the out put power of PV system as high as possible.Artificial Neural Networks (ANN) has good ability of nonlinear fitting. And Particle Swam Optimization Algorithm (PSO) performs well in optimization problems. Based on these, the modeling and maximum power point tracking (MPPT) of PV system are implemented.Firstly, the current situation and the developing trend of PV system are introduced in details. Its advantages and characteristics, meanwhile the existing problem are also analyzed in the first chapter.Secondly, ANN and PSO are introduced. The rationale, network structure, interconnection mode, algorithm and training process of ANN are analyzed. Besides, optimal theory is covered also. The works of this thesis are based on the theories above.Thirdly, the photoelectric principle is well introduced. The electrical characteristics and external characteristics are also analyzed in details. Considering the nonlinearity of input and output of PV cells, Back-Propagation Neural Network (BPNN) optimized by PSO algorithm (PSO-BPNN) is used in PV array modeling. Compared with some other methods of PV array modeling, the simulation results show that this method has better performance in convergence rate and simulation precision.At last, the theory of MPPT is presented. And the affecting factors of it are also analyzed. After reviewing the traditional MPPT control method, a new method based on PSO-BPNN is proposed to track the maximum power point. The simulation results show that this method has better efficiency and tracking precision.