Online Model Parameter Estimation of Jet Engine Degradation for Autonomous Propulsion Control

摘要

Jet engine components are subject to degradation over their lifetime of use, and this can lead to a deterioration in thrust performance of the engine. For autonomous propulsion control, it is desirable for the engine control system to maintain a nominal level of propulsion system thrust performance in an engine subject to changes in dynamics caused by aging and degradation. In this paper, two adaptive control techniques are investigated to recover the thrust performance of a degraded engine so that it is as close as possible to the thrust performance of a nominal (new) engine. The first technique consists of an adaptive onboard linear engine model tuned by parameter estimation using a Kalman filter, and used in closed loop with a PI controller to maintain nominal thrust performance. In the second technique, the nominal PI controller gains are adapted with a least squares method of controller parameter optimization in the frequency domain, so that the closed loop frequency response of the degraded engine and adaptive PI controller matches the closed loop frequency response of a nominal engine and nominal PI controller. Use of the new adaptive controller recovered the closed-loop transient thrust response of the degraded engine to nominal levels.