An Optimal State Estimation Observer for Fault Detection of Gas Turbine Engine

This paper presents a new scheme based on state estimation to diagnosis an actuator or plant fault in a class of nonlinear systems that represent the nonlinear dynamic model of gas turbine engine. An optimal nonlinear observer is designed for the nonlinear system. By utilizing Lyapunov's direct method, the observer is proved to be optimal with respect to a performance function, including the magnitude of the observer gain and the convergence time. The observer gain is obtained by using approximation of Hamilton -Jacobi -Bellman (HJB) equation. The approximation is determined via an online trained neural network (NN). Using the proposed observer, the system states and the fault signal can be estimated and diagnosed, respectively. The proposed approach is implemented for state estimation and fault detection of a gas turbine model subject to compressor mass flow fault. The simulation results illustrate that the proposed fault detection scheme is a promising tool for the gas turbine diagnostics.