PD pattern recognition in oil-paper insulation based on discharge time interval

Abstract

Partial discharge (PD) inside insulation is considered as one major cause of insulation degradation in electrical equipment and attached importance to the safety and reliability of running electrical equipment. Time intervals between consecutive discharges ( Δt ) may even be more efficient to characterize a defect or to differentiate between different defects. Oil-paper insulation is the most important part in transformer. In this paper, five kinds of typical artificial defect models of oil-paper insulation were designed. The time interval distributions of PD pulse were introduced in PD pattern recognition, the 3-D pattern Hn (Δt, ϕ) of discharge phase, time interval and discharge number ϕ-Δt-n distribution was constructed, and the box dimension and information dimension of gray intensity images which are transferred by 3-D pattern were analyzed and extracted. In this scheme, PD fractal dimensions were used as inputs, radial basis function neural network (RBFNN) was used as classifier, five kinds of artificial oil-paper insulation defects were distinguished, recognition rates are all over 90%, and it shows well in noise interference suppression.