Aircraft electrical power systems prognostics and health management

Military and commercial aircraft, spacecraft and ground vehicles are increasingly dependent on electrical power. It has become common place for vehicles to rely on electrical power, in whole or in part, for all systems, including critical systems such as flight control and fuel delivery. Microprocessors embedded in digitally controlled power distribution systems, as well as in the digital controllers within these systems, provide an unprecedented, affordable and inherent opportunity to monitor an electrically powered vehicle´s systems health. Data transmitted to and from these controllers can be used to characterize the system and component operating signatures, thereby enabling advanced diagnostic and prognostic capabilities. These capabilities will ensure a high mission reliability rate as well as reduce life cycle ownership costs. The aircraft electrical power systems prognostics and health management (AEPHM) program, presently being worked by Air Force Research Laboratories (AFRL), Boeing, and Smiths Aerospace, has developed and demonstrated health management (diagnostics, prognostics and decision aids) algorithms. The first phase of the program, which ended in July of 2005, addressed electric actuation, fuel pumps/valves and arc fault protection. The second phase is addressing power generation. Algorithm development is based on data collected from seeded and accelerated run-to-failure laboratory testing. The AEPHM architecture supports system level fusion of evidence and state information from multiple sources to improve estimates of degradation. The robustness of health management as a function of possible data sources and data rates is being determined. The product of the research will be adaptable to a range of platforms, including military, space and commercial vehicles. Phase I of the program was completed with an end to end, hardware-in-the-loop (electric actuator, fuel pump, fuel valve, arc fault, and power distribution unit) demonstration with on-line - - data generation to show the integration of the technology into a realistic setting