Intelligent electric energy meter fault prediction method based on device measurement data

The invention discloses an intelligent electric energy meter fault prediction method based on device measurement data, and relates to the technical field of electric energy meter fault prediction, and the method comprises the following steps: obtaining an electric energy meter fault characteristic variable; carrying out data abnormal value elimination, and carrying out normalization processing on the characteristic variable data; integrating the time domain feature information into the normalized feature variable, and constructing a convolutional neural network input matrix related to the time domain information and the fault feature variable; inputting data, and performing forward propagation of the convolutional neural network to obtain a model output result; a loss function of the model is calculated, model parameters are updated, and an intelligent electric energy meter fault analysis model is obtained; and constructing an input matrix, carrying out intelligent electric energy meter fault analysis by adop