DGA fault diagnosis based on the counter propagation neural network optimized by parallel genetic algorithm

Using the counter propagation artificial neural network (CPANN) to diagnose the DGA fault, network structural parameters should be set, such as the training epochs, network size etc. When user to set, it would be affect by the artificial subjective factors. If we use the traversal search way, it would be the consumption of computing and time. So this article employed parallel genetic algorithm to optimize network structure optimization parameters of counter propagation neural network. Genetic algorithm is a simulation Darwin the evolution natural selection and genetic mechanism of biological evolution process calculation model, and a by simulating natural evolution to search the optimal solution. In the GA procedure, the fitness function was defined by the correct ratio combination of the calibration data set and validation data set, as the rules for selecting the optimal network parameters. When selecting the optimal network parameters, the relatively high repeated frequency of chromosome and the optimal fitness function values simultaneously were considered.