Power Communication Network Fault Diagnosis Method Based on WNN-SCNN

Aiming at the problems of poor timeliness and low accuracy of current power communication network fault diagnosis methods, this article proposes a type of power communication network failure diagnosis method based on the wavelet neural network combination to improve the convolutional neural network (WNN -SCNN). First of all, filter alarm data and select valuable attributes for diagnosing faults. Then use the wavelet neural network model to obtain different alarm types and information for the fault diagnosis. Finally build a convolutional neural network containing skip-connections convolution structure, and make full use of the correlation between front and rear layer of data to improve diagnostic accuracy. The experimental results show that this method can effectively diagnose faults in power communication networks, with good timeliness and high accuracy of diagnostic results.