A Machine-Learning Based Method to Analyze the Correlation between Meteorological Data and Component Outages of Power System

Apart from the operating conditions of power systems, weather conditions have great impact on the reliability performance of the components in power systems. Analyzing the underlying correlation between weather conditions and component outages will be essential to the operating reliability of the systems. This paper proposes a machine-learning based method to analyze the underlying correlation based on real time data. A data enhancement method is also proposed in case the realistic data are insufficient or unavailable. Case studies based on realistic meteorological data of 5 areas in Fujian, China, combined with the IEEE RTS-96 system, demonstrate the effectiveness of the proposed method.