Defect Classification of Power Secondary Equipment Based on XGBoost Algorithm

Abstract

Accurate determination of the severity of secondary equipment defects in power grid can provide an important basis for the operation and maintenance of equipment. Aiming at problems such as large quantity of defect data features, great difficulty in human judgment and error-prone, in this paper, a defect classification method based on XGBoost (eXtreme Gradient Boosting) is proposed to improve the accuracy of defect classification of secondary equipment. Firstly, a series of preprocessing work, such as removing outliers and coding, is carried out on the secondary equipment historical defect data, and the characteristics highly correlated with equipment defects are screened out to establish the feature index set. Then, the XGBoost model is trained and optimized by using historical defect data. Finally, the trained classification model is used to realize the accurate classification of secondary equipment defects. Based on the secondary equipment defect data of a power plant, simulation results are presented to illustrate the effectiveness of the proposed algorithm, comparing with the traditional classifier (decision tree, logistic regression, etc.), simulation results show that XGBoost can accurately judge the defect degree of secondary equipment, so as to assist the maintenance and management of equipment.