Condition Monitoring and Operation Analysis of Power Generation Metering Device Based on Multi-Source Information Fusion

In this paper, a condition monitoring system of power generation metering device based on multi-source information fusion is proposed. By integrating the equipment running time, environmental parameters (such as temperature and humidity), electrical parameters (such as voltage and current) and other multi-source data, the system uses Kalman filter, Bayesian estimation and machine learning algorithm for intelligent comprehensive analysis, and realizes the accurate assessment of equipment status and fault warning. The experimental results show that the health status evaluation accuracy of the system is more than 95%, and it can give early warning before the equipment fails, and the accuracy rate is more than 85%. The system significantly improves the reliability of power metering equipment and trade fairness.