Research On Trend Prediction Of Gas Path Parameters Of Aeroengine

Flight safety is the theme of the Civil Aviation forever. Aeroengine which provides the needed impetus to energy is known as the heart of aircraft, the working reliability of engine is the key to ensure flight safety. It is important significance of using the state of the engine parameters to predict and determine engine gas system trends, which can earlier monitor the state of aeroengine and help engineers more scientifically project the use and maintenance of aircraft engine. It can support engineers in completing the fault prediction and health management of aeroengine.This paper first introduces the gas path condition monitoring of aeroengine, which includes the parameters trend monitoring on cruising and the EGT margin on take off. According to the CFM56 series aeroengine adopts N1 speed as the parameters of the aeroengine thrust, author takes the gas path parameters(EGTM、ΔEGT、AFF、ΔCORE SPEED) as the research object.In order to ensure the accuracy of the prediction, preprocessing the performance parameters of aircraft engine according to the characteristics of aeroengine performance parameters, which includes identification and processing about abnormal data, in addition, the secondary exponential smoothing method is used to make the performance parameters of aircraft engine more smooth.After preprocessing the performance parameters of aircraft engine, this paper uses the ARIMA model of time series method and least squares support vector machines(LS-SVM) method to predict the performance parameters of aeroengine, compared with the original data and two predicted data, we can find the method of LS-SVM more accurate.Based on the above research results, designing the prediction system for gas path performance parameters of aeroengine based on JAVA, and introducing the modules and functions of the system.