Classification of different types of partial discharge based on acoustic emission techniques

The major cause of failure of high voltage apparatuses is the presence of partial discharge in insulation structure. Different types of partial discharge existed in power apparatus, which will lead to breakdown of insulation. The purpose of the research work presented in this paper constitutes the issue of effective and efficient recognition of different types of partial discharges. First, partial discharge activity under AC is studied using acoustic emission technique. The wavelet technology is used to analyze acoustic signal caused by PD. Then a hybrid model which combined wavelet transform with wavelet network is proposed to classify and characteristic different types of signals. By using this method the surface, plane-plane, pin-plane, and floating discharge signal is classified effectively. Finally, to demonstrate the effectiveness of the proposed classified method, this study investigates its identification ability using 150 sets of acoustic signals generated by the PD model in insulation oil. The experiment results show that the proposed method is efficient and reliable.