Faults Diagnosis of H-bridge Bi-directional DC-DC Converter Based on Cuckoo Search Algorithm and BP Neural Network

Abstract

In energy storage device, H-bridge bi-directional DC-DC converter bear the ability of bidirectional transportation of energy, but it has large amount of fault information. A novel fault diagnosis method for H-bridge bi-directional DC-DC converter is proposed in this paper, based on BP neural network (BP) as classifier optimized by Cuckoo Search algorithm (CS). Neural networks have the strong classify ability which can be used to fault diagnosis. But it is very easy to trap the local minimum if the initial network weights are randomly generated. To solve this problem, we use the CS algorithm to optimize the initial weights of the neural network. The simulation results show that the method can effectively diagnose the H-bridge bi-directional DC-DC converter.