Firstly

using Haar wavelet and DB16 wavelet

the original exhaust gas temperature data series were denoised.Further analysis on the denoised data series indicates the existence of a chaos feature.Then

by using chaos theory

chaotic forecasting arithmetic was established to forecast the exhaust gas temperature data series.Finally

by testing the stable level of data series and comparing the series with the red line

the condition of aero-engine was defined.The proposed arithmetic was verified through some types of aircraft aero-engine exhaust gas temperature data series obtained from actual flight

and then compared with adding-weight one-rank local-region arithmetic and auto-regressive and moving average (ARMA) arithmetic.The result shows that the proposed arithmetic has a better forecasting accuracy.It can be used as a supporting model in the decision-making of aero-engine fault forecasting.