Annual Conference of the Prognostics and Health Management Society, 2009

Application of Blind Source Separation Techniques for Generation of PHM Useful Information


Bruno P. Leão1 , João P. P. Gomes1 , Roberto K. H. Galvão2 , Takashi Yoneyama2
1Empresa Brasileira de Aeronáutica S.A., S. J. dos Campos, SP, 12227-901, Brasil
2Instituto Tecnológico de Aeronáutica, S. J. dos Campos, SP, 12228-900, Brasil

Abstract:

One of the most important issues when dealing with PHM developments is the availability of adequate sensors to provide measures that indicate the health state of a component or system. Installation of additional sensors for such purpose usually implies increments in costs and weight and reduction of reliability and availability. Sometimes equivalent information can be inferred from other available sources, allowing the design of PHM solutions with no need for additional sensors. The power consumed by a set of components may provide information concerning their health states. These components may be all fed by the same power supply. This paper proposes a novel application of blind source separation techniques to infer the power consumed by the components using only the measurement of the power supply output. The usefulness of such techniques is demonstrated in a real application.

This article is available online here and to CALCE Consortium Members for personal review.

[Home Page] [Articles Page]
Copyright © 2009 by CALCE and the University of Maryland, All Rights Reserved