Fault detection and reconstruction for micro-satellite power subsystem was achieved by using the principal component analysis (PCA) method. Four typical kinds of sensor failure were identified. The principle to establish the PCA model, diagnose and reconstruct faulty sensors was presented. A model of a satellite's power subsystem, which is consisted of voltage, temperature, and current parameters, was introduced. Using the normal data acquired from the experiment, we established the approximate PCA model for the satellite power system. With the incorrect data introduced to simulate the sensor failure occurrence, the value of squared prediction error (SPE) and sensor validity index (SVI) was detected beyond the confidence limit. Subsequently, the faulty sensor was isolated and reconstructed. The simulation results indicated that the PCA method was probably effective for fault detection and reconstruction in satellite power subsystem.
Fault detection and reconstruction for micro-satellite power subsystem based on PCA
01.06.2010
885800 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
A Review of Fault Detection and Diagnosis of Satellite Power Subsystem
Springer Verlag | 2021
|Ontology-Based Knowledge Modeling for Power Source Subsystem of Satellite Fault Diagnosis
Trans Tech Publications | 2013
|Efficient Chaotic Based Satellite Power Supply Subsystem
AIAA | 2006
|Electrical Power Subsystem of İMECE Satellite
IEEE | 2023
|Next generation GOES satellite power subsystem
IEEE | 2001
|