Title :
The component-level and system-level satellite power system health state evaluation method
Author :
Hongzheng Fang ; Hui Shi ; Yi Xiong ; Rui Li ; Ping Wang
Author_Institution :
Beijing Key Lab. of High-speed Transp. Intell. Diagnostic & Health Manage., Beijing, China
Abstract :
The health state evaluation method is one of the key technologies of Prognostic and Health Management (PHM). Through the health assessment of the satellite power system, it can timely find the potential trend of health degradation and hidden danger of influence of health level, thus improving the level of effectiveness, precision and autonomation of the power system management. Firstly, it analyzed the health definition and hierarchy division method of the satellite power system, then it proposed and studied the power supply system-level health state evaluation method based on fuzzy analytic hierarchy process (AHP), and the component-level health condition assessment methods based on the statistical analysis, Bayesian network and Support Vector Machine (SVM) respectively. The experimental results verified the effectiveness of the proposed health state evaluation method.
Keywords :
aerospace instrumentation; analytic hierarchy process; fuzzy set theory; power supplies to apparatus; space vehicle power plants; statistical analysis; support vector machines; AHP; Bayesian network; PHM system; SVM; component-level health condition assessment method; component-level satellite power system; fuzzy analytic hierarchy process; health state evaluation method; hierarchy division method; power supply system-level health state evaluation method; power system management; prognostic and health management system; satellite power system health assessment; statistical analysis; support vector machine; system-level satellite power system; Arrays; Batteries; Bayes methods; Power systems; Satellites; Support vector machines; Vectors; Bayesian networks; fuzzy analytic hierarchy process; health state evaluation; power supply system; satellite; support vector machine (SVM);
Conference_Titel :
Prognostics and System Health Management Conference (PHM-2014 Hunan), 2014
Conference_Location :
Zhangiiaijie
Print_ISBN :
978-1-4799-7957-8
DOI :
10.1109/PHM.2014.6988260
