Electric Power Cable Fault Recognition via combination of wavelet transform and optimized artificial neural network by using bees algorithm

Fault detection and diagnosis of underground cables is one of the basic parts of maintenance and repair of power cables, and without using of modern diagnosis approaches, the electric power distribution companies cannot provide reliable services to industries and public consumptions. So, power cable diagnosis requires widespread support and attention. In this paper by means of case based methods and also fault detection and fault classification algorithms, the common faults in the underground cable transmission system will be detected. The neural network has been used for classification, in the proposed method. Also, to optimize the performance of neural network, the wavelet transform as effective input, and the bee algorithm for finding the optimal values of neural network control parameters have been used. The simulation results show that the proposed method has high detection capability and shows good performance, so that can separate about 100 percent of faults successfully. For this reason, the overlap matrix has been presented in analyses.