A health baseline construction method for key subsystems and stand-alone correlations based on historical telemetry data

Abstract

The present application discloses a key sub-system based on historical telemetry data and a method for constructing a single-machine correlation health baseline, which includes: selecting a set of binary remote reference sequence combinations that are linearly correlated or nonlinearly correlated; The combination of the selected binary remote reference sequence is time-calibrated; the second step is to identify and cut the target working condition for the time-calibrated binary remote parameter sequence, and obtain the binary remote parameter sequence under the target working condition period; the third step , use the binary remote reference sequence data of the normal state target working condition to fit the binary linear correlation health baseline, or use the method based on discrete integration to convert the nonlinear relationship to the linear relationship, and fit the binary nonlinear relationship Relevant health baseline; the binary linear relational health baseline and the binary nonlinear relational health baseline together constitute a relational health baseline library. Through the constructed correlation health baseline library, the quantitative and stable characterization of the key satellite subsystems and stand-alone health status is realized.

Description

technical field [0001] The present application relates to a satellite health monitoring technology, and in particular, to a key subsystem based on historical telemetry data and a method for constructing a single-machine correlation health baseline. Background technique [0002] A satellite is a system with complex functions and composition, including a large number of key subsystems and single units, and a large number of operating status data. By analyzing the operating status data of satellites and monitoring whether they are in abnormal operating state, it is helpful to find and deal with the operating faults of satellites in time, and it is of great significance to ensure the operational reliability of satellites. [0003] To carry out abnormal monitoring of satellites, it is first necessary to accurately, effectively and quantitatively characterize the health status of satellites. The traditional health state characterization methods are mainly divided into three types...

AI Technical Summary

Problems solved by technology

[0005] In view of the problems existing in the current satellite health state characterization methods, such as high dependence on expert knowledge, insensitivity to early anomalies, poor robustness of anomaly detection, and too complex model volume, this application aims to propose a key subsystem based on historical telemetry data and A method for building a health baseline based on a single computer

Images