| Energy is not only the basis of social development, but also the driving force of economic development in the world. Solar energy has become one of the most important clean energy in the world because of its unique characteristics of zero pollution and inexhaustible. The research on photovoltaic power generation system not only provides an effective solution to the sustainable use of energy, but also has important strategic significance to the stable development of society. The maximum power point tracking control technology is an effective way to improve the efficiency of photovoltaic power generation system. In this paper, the photovoltaic power generation system as the research object,the maximum power tracking problem is studied:1. According to the equivalent circuit of the photovoltaic cells derived the mathematical model, simulation platform using Matlab/Simulink to build the model of photovoltaic cells and the photovoltaic power generation system. Using the simulation platform to simulate the output characteristics of photovoltaic cells, and the traditional ?-? method is analyzed and simulated.2. Based on BP neural network, the maximum power point tracking,process error of training data is analyzed, and points out that BP neural network based on least squares (LS-NN) algorithm is the accuracy of the prediction results of MPPT is heavily dependent on the accuracy of training data, a BP neural network based on quasi least squares(QLS-NN) MPPT algorithm. The predicted results of two different algorithms are compared by simulation and experiment.3. In this paper, by using the impedance matching principle of maximum power tracking method was researched. Through the theoretical analysis of the traditional impedance matching MPPT algorithm, it is found that the traditional impedance matching MPPT algorithm is more sensitive to the system parameter setting, and an improved dynamic impedance matching MPPT algorithm is proposed.Compared with the simulation and experiment, the improved dynamic impedance matching MPPT algorithm proposed in this paper can effectively improve the performance of the PV system.This research uses the BP neural network MPPT algorithm based on least squares quasi reduced error of training samples containing process to predict the impact of the maximum power of the neural network,improve the robustness of the system; MPPT algorithm based on improved dynamic impedance matching in traditional impedance matching MPPT algorithm can solve the system parameter setting and output power more effectively the influence of load, so as to provide a reference for the photovoltaic power generation system overall efficiency and stability. |
PDF Full Text Request