In order to improve the accuracy of cable partial discharge detection,this paper proposes a cable partial discharge detection method based on wavelet denoising and LSSVM.Using wavelet transform to remove noise from cable partial discharge signals,singular value spectrum is used to extract cable partial discharge features,and the extracted results are used as support vectors for LSSVM.Based on this,an LSSVM classifier for partial dis-charge is established.The simulation analysis using actual examples shows that the detection results of the cable partial discharge detection method proposed in this paper only show two errors,with a detection accuracy of 95%.The detection effect is better than other methods,verifying the correctness and effectiveness of the proposed method.
关键词
电缆/局部放电/检测/小波变换/最小二乘支持向量机
Key words
cable/partial discharge/detection/wavelet transform/least squares support vector machine
维普
万方数据