Arc-Fault Recognition Based on BP Neural Network

The back-propagation algorithm has been used widely as a learning algorithm in a feed-forward multilayer neural network. In this study, fault detection was carried out using the information of the arc current. After collecting the actual data, wavelet transformation were adopted in order to obtain the sideband or detail value characteristics under healthy and various faulty operating conditions. The feature vectors consisted of the values obtained by wavelet transformation and turned into the input of network. They were trained with BP neural network system, and the fault detection algorithm was verified using the test data. The experiment shows that the method has better performance for arc-fault recognition.