Electric power fault signal detection method based on convolutional neural network

The invention provides an electric power fault signal detection method based on a convolutional neural network, and solves the problem that an existing fault signal detection mode cannot give consideration to fault signal pre-judgment and process design conciseness, and the method comprises the steps: firstly collecting an electric power signal with a fault and a known fault position and an electric power signal without a fault; constructing a preprocessing matrix and an output training matrix matched with the preprocessing matrix are constructed, a convolutional neural network model, employing the preprocessing matrix output training matrix to train the convolutional neural network model to obtain an electric power fault signal detection model, and finally, detecting an electric power signal to be detected, wherein the design complexity of the detection process is low, and the detection efficiency is high. Moreover, based on the trained power fault signal detection model, the to-be-detected power signal with