Application of a Constant Gain Extended Kalman Filter for In-Flight Estimation of Aircraft Engine Performance Parameters

Abstract

An approach based on the Constant Gain Extended Kalman Filter (CGEKF) technique is investigated for the in-flight estimation of non-measurable performance parameters of aircraft engines. Performance parameters, such as thrust and stall margins, provide crucial information for operating an aircraft engine in a safe and efficient manner, but they can not be directly measured during flight. A technique to accurately estimate these parameters is, therefore, essential for further enhancement of engine operation. In this paper, a CGEKF is developed by combining an on-board engine model and a single Kalman gain matrix. In order to make the on-board engine model adaptive to the real engine’s performance variations due to degradation or anomalies, the CGEKF is designed with the ability to adjust its performance through the adjustment of artificial parameters called “tuning parameters.” With this design approach, the CGEKF can maintain accurate estimation performance when it is applied to aircraft engines at off-nominal conditions. The performance of the CGEKF is evaluated in a simulation environment using numerous component degradation and fault scenarios at multiple operating conditions.