To apply pattern recognition technology to cable defects under oscillating voltage

four kinds of artificial defect models of 10 kV XLPE cable joints were made according to common cable defect types. Oscillating voltage was applied to them to test the partial discharge signals. Statistic operators of positive and negative half-waves of the partial discharge signal were taken as the inputs of pattern recognition

and four different defects were recognized by support vector machine. The recognition result of the support vector machine was compared with that of artificial neural network to verify its effectiveness. It is shown that taking the statistic operators of positive and negative half-waves as inputs of pattern recognition can perfectly reflect the features of partial discharge information

and the proposed pattern recognition method based on support vector machine can effectively identify the partial discharge patterns of different defects with higher recognition rate and running speed than artificial neural network.