Multiple Control Surface Failure Detection Via On-Line Filter Tuning Method

Abstract

Modern high performance aircraft systems require advanced onboard fault detection, isolation and reconfiguration (FDIR) techniques against system failures. In this paper, we will devise a on-line FDIE (Failure Detection Isolation and Estimation) scheme that could handle multiple and time-varing control surface failure and provide precise health status for future reconfiguration design. This approach combining LQG design with newly developed fuzzy concept would be able to deal with a variety of failure types such as constant failure (hardover), linearly degraded (run-away) and their combination. A big contribution of this approach is unanticipated failure can be successfully detected without prescribing any nominal failure models in memory.