Fault Location Method for Secondary Equipment in Intelligent Substation Based on Improved BA-PNN
Aiming at the problem that the training scale of probabilistic neural networks(PNN)in secondary equipment fault location is large and susceptible to smoothing factor interference,the fault location strategy is presented for secondary equipment of intelligent substation based on improved bat algorithm optimized probabilistic neural network(BA-PNN).Firstly,within the summation layer of PNN,the substitution of Gaussian distribution with Laplacian distribution is implemented,and the optimal smoothing factor is acquired by the utilization of the BA algorithm,and an improved bat algorithm optimized probabilistic neural network(BA-PNN)is proposed.Secondly,based on the feature analysis of secondary equipment of intelligent substation,the fault characteristic quantity is selected and mapped,the fault location model of secondary equipment of intelligent substation based on BA-PNN is established.Finally,taking the fault location of a certain intelligent substation as an example,the BA-PNN neural network is trained with samples to achieve accurate fault location.The simulation demonstrates that the proposed approach effectively minimizes the training scale of neural network,improves the computational performance of neural network and enhances the accuracy of fault location.
improved bat algorithm optimizing probabilistic neural networksecondary systemintelligent substationfault location
万方数据
维普
