In-Flight Isolation of Degraded Engine Components by Shape Comparison of Transient Outputs

Abstract

A direct method of aircraft engine health monitoring is introduced that can isolate a degraded engine component in-flight. The method utilizes continuous wavelet transformations of transient engine sensory data to represent their shape attributes in the time-scale domain. This enables contrasting the shapes of the current engine outputs with those previously collected from the engine. Continuous wavelet transforms also provide enhanced delineation of the engine transients in the time-scale domain. This enables identification of minute differences between the outputs affected by component degradations as well as between the sensitivities of modeled outputs with respect to the health parameters or components. The presence of these differences is used in this method as evidence of identifiability of the degradation by the outputs as well as of the effect of parameters or components on the outputs. The effectiveness of the proposed method is evaluated in engine simulations. The results indicate that with the suite of outputs currently available on-board 70% to 96% of the degraded components simulated can be isolated for new and older engines.