A Novel Prognostics Library for Electromechanical Systems

Fault diagnosis and prognosis is very important in health monitoring or condition based maintenance. If one can measure the degradation of a component before it actually fails, then it will provide ample time for maintenance engineers to schedule a repair,and to acquire replacement components before the components actually fail.IAI and its STTR partner, Georgia Institute of Technilogy, propose a library of prognostics tools that contain several innovative ideas. First, a Principal Component Analysis (PCA) based prognostics tool is proposed. The PCA based tool was applied toassess the health status of a gearbox at Penn State University. We were able to detect a gearbox failure five hours before the tooth broke. Second, a Hidden Markov Model (HMM) based approach is proposed to predict the degraded state of the system. The HMMbased tool was applied to a rotating shaft system in IAI's laboratory. Third, an adaptive prognostics tool is proposed to predict the remaining life of a system. The adaptive prognostics tool was demonstrated by using a bearing testbed at Georgia Tech.Again accurate remaining life was correctly predicted. The new library is highly relevant to this topic since we plan to apply the techniques to the Airborne Laser (ABL) system. The technology can be used in any military and commercial applications, where electromechanical systems are used. The proposed system may also be extended to many other applications, such as bearings, pumps, gearboxes, motors, etc. The Joint StrikeFighter (JSF) program is our first target Phase 3 application. In addition, IAI has developed health monitoring systems and/or components for Boeing, Ford, and Motorola. Recently, IAI has completed a contract with Boeing to develop health monitoring andprognostics technology for the Future Combat System (FCS) of the Army. The research results of this project will lay down a solid foundation for future commercialization effort with these and other companies.