Anomaly detection of orbit satellite telemetry sequence based on two-window mode

Aiming at the problem of low detection accuracy and high false alarm rate in satellite telemetry sequence anomaly detection, a new anomaly detection algorithm of orbit satellite telemetry sequence based on Two-Window mode is proposed. The telemetry dynamic average sequence in the learning window is used as the benchmark template, to obtain the template learning ability of the latest telemetry sequence. Adaptive detection threshold is established based on the five-distribution-number in detection window, to detect the telemetry abnormal sequence when the satellite on orbit is changing. The experimental results of anomaly detection of a satellite power system telemetry data show that the algorithm can adapt to the complex changes of satellite telemetry sequence in orbit, reduce noise and outliers effectively, improve the accuracy of detection, realize the on orbit satellite remote sensing anomaly detection effective sequence and achieve effective detection of satellite telemetry sequence anomalies on orbit.