Research On Model-Based Fault Diagnosis Of Variable Cycle Engine Gas Degradatoin

In order to meet the performance requirements of high unit thrust and low cycle fuel consumption,the variable cycle engine has become an ideal choice for future civil and military aircraft power plant.Its healthy and stable engine operation is an important guarantee for the aircraft to always fly safely all the time.Due to the harsh working state and external conditions of the engine,as the working time increases,the state of the engine may deviate more or less from the stable trajectory,and its performance parameters will gradually degrade and lead to more serious failures.The failure of gas path component performance degradation is the main fault type of variable cycle engines.Therefore,it is of great significance to monitor the performance degradation state parameters of the gas path components of variable cycle engines and perform fault diagnosis.Compared with the data-driven approach,the model-based fault diagnosis method has more practical significance.Because the gas path performance parameters are extremelysensitive tothe engine’s different engine working,namely,different heights,speeds,and fuel quantities,the amount of data required for fault diagnosis will be very large,and the fault diagnosis data cannot cover the faults of the entire flight envelope under different operating conditions.Due to the diverse and complicated working conditions of the engine,the input cost of data collection will be very high,and the credibility of the data obtained at the same time is difficult to make a basis for judgment.Therefore,the model-based gas path fault diagnosis methodisstillan importantbasis for aeroengine faultdiagnosis.This paper first builds a dual-duct variable-cycle engine component-level model on the Matlab / Simulink platform,considers the volume of the engine gas path component connection section,and regards the external duct,combustion chamber,high and low pressure turbine as the volume cavity,regardless of the import and export pressure difference,avoids the iterative calculation caused by the flow balance,and provides a steady-state gas path performance simulation foundation for fault diagnosis of the gas path degradation of the variable cycle engine.Secondly,the mechanism of the performance degradation of each component of the engine gas path is analyzed.By introducing the method of component characteristic flow rate and efficiency degradation parameters,the effects of different component characteristic degradation on the gas path performance parameters are obtained through simulation,so as to select the one that can truly reflect the gas path failure Performance measurement parameters.The state variable model of the engine’s nonlinear model is established,and the residual value of the actual engine gas path performance parameter is taken,combined with the Kalman filter to estimate the characteristic degradation amount of the gas path component.Finally,the similarity analysis of different gas path degradation failure modes is carried out,and the linear variation of measurement parameters caused by different engine component characteristic degradation is studied.In view of the actual situation,the actual number of sensors that can be installed in the engine is less than the number of engine degradation parameters.Based on the SVD-KF method,the problem of fault diagnosis of engine gas path component degradation parameters under the condition of insufficient installable sensors is solved.A real-time performance tracking model and a regularly updated baseline model are designed for fault diagnosis of variable cycle engine gas path degradation parameters.The regularly updated baseline model can identify sensor failures,and the real-time performance tracking model can better track the amount of degradation of the real engine gas path components,and detect the performance degradation caused by sudden engine faults.