The IA Tech, Inc. and Institute for Scientific Research, Inc. (ISR) team proposes an innovative concept for data driven prognostics that includes a library of predictive algorithms and an architecture with which these algorithms can be selected, implemented, visualized, and monitored. When systemic changes are indicated by the data collected and mined, the prognostics system refines the previously developed algorithm. This approach leverages the successful data mining and predictive algorithm development that ISR has developed for the Fixed Head Star Tracker (FHST) subsystem of the Hubble Space Telescope, with test results indicating a potential 60% decrease in FHST failures. ISR is also participating in the Intelligent Flight Control System program sponsored by NASA Dryden Flight Research Center. This ongoing effort includes developing, for flight demonstration, a flight control concept that uses neural networks to optimize aircraft stability in both normal and failure conditions. The expertise, tools, algorithms, and lessons learned are all part of the technology ISR will transfer and share with IA Tech for the approach and solution to the Airborne Laser and similar data driven prognostics problems. The resulting library application/implementation architecture will be portable to different platforms and extensible to accommodate advances in data driven prediction and prognostics. Anticipated Benefits/Commercial Applications: Effective data driven prognostics can reduce system maintenance costs and downtime, extend the life of a system, and improve its reliability, safety and operational performance. This capability and its attendant benefits are applicable to the Airborne Laser and a number of commercial sectors, including for example, aircraft, power, manufacturing, processing, transportation, and medical systems.
Keywords: prognostics, data mining, predictive algorithm, data driven, failure signature, computer-controlled equipment, Airborne Laser
