Intelligent fault diagnosis and early warning method and system for low-orbit constellation

Abstract

The invention discloses an intelligent fault diagnosis and early warning method and system for a low earth orbit satellite. The method comprises a fault diagnosis step and / or a fault early warning step. The fault early warning step comprises the following steps: S1, setting a prediction time length, and forming a first prediction diagnosis result sequence by a current diagnosis result and a historical diagnosis result; S2, inputting the mth prediction and diagnosis result sequence into a fault early warning model, and outputting a prediction and diagnosis result of the mth telemetry data acquisition period in the future by the fault early warning model; S2, if m is greater than or equal to M, entering S3, if m is less than M, adding a prediction diagnosis result of mth prediction to the tail end of an mth prediction diagnosis result sequence and removing head end data of the mth prediction diagnosis result sequence to obtain an (m + 1) th prediction diagnosis result sequence, and if m is equal to m + 1, returning to S2; and S3, forming a prediction sequence by M prediction diagnosis results in the prediction time duration. The performance of the constellation in a period of time in the future can be predicted, constellation fault early warning is realized, and the complexity of a prediction model is reduced.

Description

technical field [0001] The invention relates to the technical field of satellite fault diagnosis, in particular to a low-orbit satellite intelligent fault diagnosis and early warning method and system. Background technique [0002] As one of the most widely used spacecraft, satellites have complex physical structures, and suffer from harsh working environments and unknown space interference during their entire operation cycle. These factors pose challenges to their reliability and safety. . Once a satellite fails, the mission will be downgraded, and the satellite will be out of control or even scrapped; it will not only bring huge property losses to the country and the company, but also bring a certain degree of security loss to the space environment. Therefore, it is necessary to monitor the operating status of satellite equipment in orbit, and timely and accurately discover satellite faults and diagnose fault information, so as to improve the reliability of satellite equi...

AI Technical Summary

Problems solved by technology

The main performance is as follows: 1) Due to the large differences in the telemetry parameters of different low-orbit satellite systems, the expert systems for processing and analyzing satellite telemetry parameters are mostly specific, and the labor cost of setting the threshold is large. ; 2) The coverage of telemetry parameters is not comprehensive, and the monitoring method is still limited to the monitoring of a single telemetry parameter itself, and there is little consideration for the correlation between parameters; 3) Telemetry parameters will occur when satellites operate in a dynamically changing space environment. The phenomenon of concept drift, the distribution of telemetry data will change with time, and the threshold will also change to a certain extent. However, conventional expert systems do not have the ability to learn independently from experience, and the threshold cannot be properly corrected in time; 4) The failure early warning ability is weak, and the alarm cannot be issued before the failure occurs

Images