Prediction of Animal-Related Failures in Overhead Distribution Systems Using Neural Network and Wavelet Transform Techniques

Occurrences of faults in overhead distribution systems are a significant factor in distribution system reliability. Prediction of animal-related failures has a practical value since animals cause large number of faults in overhead distribution systems. This paper presents a method to predict weekly animal-related failures in overhead distribution system using neural network and wavelet transform techniques. Discrete wavelet transform is applied to decompose the time series of weekly animal-related failures. Six different neural networks are constructed for each decomposed coefficient series. Finally, the output series of six neural networks are combined according to wavelet reconstruction techniques to get final output for the weekly animal-related failure prediction. Simulation results indicate that this combined method displays good performance.