Prediction method of statistical line loss rate in station area with single data loss based on deep neural network

In view of the lack of single data of station area statistical line loss rate in the collected station area line loss statistical table, which makes it impossible to judge whether the station area has abnormal line loss, a prediction filling method of station area statistical line loss rate based on deep neural network (DNN) is proposed. Firstly, through deep neural network modeling and model training, the missing single data after prediction is obtained; secondly, compared with other regression algorithms, its performance is better than other models; finally, compared with the error, it has a certain prediction accuracy. This method can effectively predict the statistical line loss rate of single data missing station area.