Spacecraft Short-Term Fault Prediction Method Based On Echo State Network Time Series

In this paper, the fault prediction method of spacecraft electrical equipment is studied with telemetry data. After analyzing the pattern of object fault and the law of telemetry data change, the fault prediction method of spacecraft electrical equipment based on telemetry data is proposed, and the application of the prediction method is analyzed. The results show that for short-term fault prediction, the short-term fault prediction with echo state network time series is better, and the average relative error of the prediction results is small. It can predict the changing trend of satellite telemetry parameters, and provide an effective tool for the management and decision of spacecraft in orbit.