Parameter identification for inflight detection and characterization of aircraft icing

Increasing interest in aircraft icing has motivated the proposal of a new i ce management system that would provide inflight monitoring of ice accretio n effects. Since these effects are manifested in the flight dynamics, param eter identification is a critical element of ice detection. In particular, identification must provide timely and accurate parameter estimates under n ormal operational input in the presence of disturbances and measurement noi se. This paper evaluates a batch least-squares algorithm, an extended Kalma n filter, and an H-infinity algorithm in the context of icing detection. Si mulation results show that only the H-infinity method provides a timely and accurate icing indication. (C) 2000 Elsevier Science Ltd. All rights reser ved.