This paper describes an experimental study of the discrimination of partial discharge (PD) signals from external noise in a crosslinked polyethylene (XLPE) cable by using a neural network (NN) system. Measurement of PD signal and external noise was carried out with a PD pulse recorder for a 66kV XLPE cable with an artificial defect and a drill. The NN was a three-layer artificial neural system with feedforward connections, and its learning method was a back-propagation algorithm.Its input infomation was a combination of the discharge magnitude, the number of pulse counts, and the phase angle of applied voltage.The NN was able to discriminate unknown input patterns with 89% correct responses after learning PD patterns that included external noise.In this case, the NN could correctly discriminate all unknown input patterns for a signal-to-noise ratio greater than or equal to unity. The duration, including the measurement time, required for the NN to discriminate an input pattern was about 45 s.

No abstract

No abstract