Fault diagnosis for aero-engine gas path considering operating condition

Abstract

There are various of failure modes and operating conditions of an aero-engine gas path system. They are correlated with each other, making the fault diagnosis for aero-engine gas path difficult. This paper proposes a fault diagnosis method for aero-engine gas path considering operating condition. First, we develop a multilayer perceptron (MLP) model to identify the operating condition based on Mach number and altitude. Then, we adopt a Convolutional Bidirectional Long Short-Term Memory (CNN-BiLSTM) model to diagnose the failure mode under the corresponding operating condition based on monitoring parameters of each cross-section state. Finally, the effectiveness of the proposed gas path fault diagnosis method was validated by the simulation data generated by gas turbine simulation program (GSP).