Cable intermediate joint is a crucial accessory that connects two cable ends.The accurate identification of its insulation status and fault types is a key task for operation and maintenance.Therefore,typical fault samples of intermediate joint were designed and made,and a 10kV cable intermediate joint fault type identification method based on the ultrasonic partial discharge testing method and DWT method for denoising the test data,combined with the LeNet-5 convolutional neural network architecture,was proposed.The results show that the proposed method can effectively identify different types of cable intermediate joint faults with an accuracy rate of up to 94％,provi-ding a new method for the operation and maintenance of cables.
维普
万方数据