Three-phase full-controlled rectifier circuit fault diagnosis based on optimized neural networks

The problem that how to use hierarchical genetic algorithm to determine the structure and parameters of neural networks was studied. The two grade coding structure of the hierarchical genetic algorithm is utilized to solve the ancient problem that when optimize the neural networks´ structure, connection weights, threshold at the same time, the efficiency was low. Furthermore, we compare the networks´ capability that optimized by the improved hierarchical genetic algorithm with the ones optimized by other algorithm, and prove the algorithm´s credibility through the simulation. At last, the improved adaptive genetic algorithm is used in the fault diagnosis of three-phase full-controlled bridge rectifier circuit, and the simulation result show the method is correct and applied.