Modeling And Fault Diagnosis Of Micro Turbojet Engine

Fault diagnosis is one of the main contents in the research of aero-engine health management system. The effective diagnosis system is the important prerequisite to achieve the condition-based maintenance, reduce the operation cost, and ensure the flight safety. Mathematical model and, failure diagnosis and isolation methods of gas path component and sensor were studied. In order to verify the effectiveness of the diagnostic algorithm for gas path component failure, rig test platform of simulating the inlet blockage fault was designed.First of all, nonlinear component-level model above idle point, and small deviation linear model of the micro turbojet engine were established. The effectiveness of the models was proved through engine test data and model simulation.Secondly, based on robust H∞ filter, three different diagnostic systems were emphatically researched and designed for gas path component failure, sensor failure, and the situation with the both types of failure occurring simultaneously. Detection, isolation and estimation of gas path component failure were realized by the gas path failure diagnosis system composed of two independent modules. For sensor failure diagnosis, different filters were designed for every sensor, and the diagnosis was achieved by comparing the norm of filter residuals with threshold. Joint diagnostic system was designed in consideration of gas path component failure and sensor failure occurring simultaneously. Diagnosis and distinction were realized by augmenting health parameters into state variable to remove its influence on filters outputs to find out the failed sensor, and then estimating component health parameters through state-observer corresponding to the failed sensor. The effectiveness of the three diagnostic systems was also validated by the simulation results.Finally, rig test experiment with inlet blockage as a type of gas path component failures was designed, by adding an air-inlet box and an area controlled baffle at the box inlet before the engine intake. The inlet blockage failure was diagnosed successfully through analyzing and processing of rig test data from both the qualitative and quantitative view, and in result, the algorithm for gas path diagnostic system was verified.