DCSCA: Detection of Current Secondary Circuit Abnormalities Based on Ensemble Learning

The current secondary circuit is an important component of new advanced power systems, and the determination of its abnormal state has always been of great concern. The measurement anomaly detection of current secondary circuit is of great significance for building a safe and reliable power system and avoiding safety accidents. Therefore, this paper proposes an anomaly detection method for metering current secondary circuit based on Ensemble learning. By integrating three algorithm models of Naive Bayes, support vector machine and XGBoost, a strong learner learning data rule is constructed. The research results show that the proposed model uses integrated learning method to improve the accuracy of measurement anomaly detection of current secondary circuit. Therefore, the proposed model can greatly reduce the maintenance workload of operation and maintenance personnel and improve work efficiency.