PD RECOGNITION WITH KNOWLEDGE-BASED PREPROCESSING AND NEURAL NETWORKS

Partial discharge (PD) patterns are an important tool for the diagnosi s of HV insulation systems. Human experts can discover possible insula tion defects in various representations of the PD data. One of the mos t widely used representations is phase-resolved PD (PRPD) patterns. We present a method for the automated recognition of PRPD patterns using a neural network (NN) for the actual classification task. At the core of our method lies a preprocessing scheme that extracts relevant feat ures from the raw PRPD data in a knowledge-based way, i.e. according t o physical properties of PD gained from PD modeling. This allows a ver y small NN to be used for classification. In addition to the classific ation of single-type patterns (one defect) we present a method to sepa rate superimposed patterns stemming from multiple defects. High recogn ition rates are achieved with a large number of single patterns genera ted by stochastic PD simulations. Our network architecture compares fa vorably with a more traditional network architecture used previously f or PRPD classification. These results are confirmed by classification of patterns measured in laboratory experiments and power stations.