Power distribution network power consumption data abnormity early warning method based on BLS and auto-encoder

The invention discloses a power distribution network power consumption data abnormity early warning method based on a BLS and an auto-encoder, and the method comprises the steps: collecting power consumption data through an information collection node, and obtaining the related characteristics affecting the power consumption condition, namely, the power consumption data amount EC, time information, the external environment information of a region, and the line condition information of the region; then carrying out data preprocessing; predicting the trend of the power consumption data flow based on a sparse BLS model; establishing a power utilization data flow anomaly detection model based on the auto-encoder; and abnormal early warning of the power utilization data flow is realized. According to the method, the efficiency can be improved on the basis of keeping original data distribution, the prediction accuracy is improved, the complexity of the model is effectively reduced, in addition, the weight of the BL