Satellite health assessment method based on Gaussian mixture model

Abstract

The invention discloses a satellite health assessment method based on a Gaussian mixture model, and the method comprises the steps: respectively carrying out the preprocessing feature extraction of obtained normal operation multi-dimensional remote parameter data and current operation multi-dimensional remote parameter data, and obtaining a normal high-dimensional feature vector and a current high-dimensional feature vector; performing training processing on an initial Gaussian mixture model by using the normal high-dimensional feature vector to obtain a trained normal Gaussian mixture model,and performing training processing on the trained normal Gaussian mixture model by using the current high-dimensional feature vector to obtain a trained current Gaussian mixture model; calculating theoverlapping degree of the trained normal Gaussian mixture model and the trained current Gaussian mixture model, and performing normalization processing on the overlapping degree to obtain a current health degree index value; and performing evaluation processing on the satellite health state by using the plurality of current health degree index values.

Description

technical field [0001] The invention relates to the technical field of spacecraft, in particular to a satellite health assessment method based on a Gaussian mixture model. Background technique [0002] The satellite system is composed of many components, the structure is complex, and due to the influence of uncertain factors that may exist in the orbital environment, it is difficult to ensure that no faults will occur during the orbital operation. Satellite system failures occur in various forms, and the functions of each component in the system are related. The propagation of failures may cause a chain reaction. If timely intervention is not possible, it will lead to serious consequences. Therefore, there is an urgent need for effective health management of satellites. [0003] Satellite health management mainly includes anomaly detection, fault diagnosis, health assessment and life prediction. Due to the variety of satellite failure modes and the lack of failure samples, ...

AI Technical Summary

Problems solved by technology

[0005] The technical problem solved by the solution provided according to the embodiment of the present inv

Images