Power consumer electricity consumption anomaly detection method based on machine learning

Abstract

The invention discloses a power consumer electricity consumption anomaly detection method based on machine learning. The method comprises the following specific steps: 1) cleaning power load data; 2)extracting power load characteristics; 3) performing power load characteristic evaluation; 4) screening data with high reliability based on an isolated forest algorithm; 5) obtaining an inter-class balance data set based on a clustering layered nearest neighbor undersampling technology; and 6) constructing a collaborative forest anomaly detection model based on semi-supervised learning. A hierarchical nearest neighbor sampling algorithm based on clustering is provided, sampling is carried out according to the proportion and the distance, the generalization ability is high, and the accuracy ofunbalanced data set classification is improved; a data set is subjected to preprocessing and feature extraction and evaluation, a time sequence of repeated information is eliminated, and the influenceof a linear relationship between features on a result is eliminated.

Description

technical field [0001] The invention relates to a machine learning-based abnormality detection method for electric power users, belonging to the field of electric power. Background technique [0002] As the basic industry of the national economy, the electric power industry develops rapidly. For a long time, phenomena such as electricity theft and fraud have been repeatedly banned, and they are characterized by intelligence and diversification, which not only endangers the economic interests of the country, but also disrupts the normal order of power supply and endangers the safe operation of the power grid. The current anti-stealing methods mainly include regular surveys by professionals and installation of detection and alarm instruments at the meter box, which increases operating costs and wastes a lot of manpower and material resources. With the rapid development of machine learning, using the power load data of electric energy meters to combine machine learning with in...

AI Technical Summary

Problems solved by technology

At present, most of the abnormal detection for power users is based on unsupervised learning, which makes the accuracy of abnormal detection low and the learning effect lacks effective verification.

Images