Application of Online Partial Discharge Detection of Swithchgear Based on Ultra-High Frequency (UHF) Method

Ultra-High Frequency (UHF) method is one of the most effective methods, which is widely applied for partial discharge (PD) detection of HV switchgear owning to its high sensitivity and strong resistance to external interference. Despite of these advantages, it is still uneasy to realize the accurate identification of the fault types of HV switchgear because of the lack of a systematic research and analysis on the partial discharge characteristics of different defects in switchgear. In this paper, three typical partial discharge defect models, including needle discharge, suspended discharge, and internal discharge were designed. Then the ϕ-q-n patterns of different partial discharges were obtained by application of UHF method. The results demonstrated based on the characteristics of ϕ-q-n patterns of various PD and differences among them, the type of single or combined PD could be effectively recognized by extracting the characteristic variables of the ϕ-q-n patterns. Moreover, the accuracy of locating PD in switchgear could be improved by use of iterative least square method. Therefore, UHF provides an effective way to evaluate and predict the defect type and position of PD of HV switchgear.