Research recognition of aircraft engine abnormal state

An aircraft engine is known as the "heart" of the aircraft, directly affects the safety of the flight. In order to identify aircraft engine lubrication system abnormal state, a method based on principal component analysis (PCA) and support vector machine (SVM) is presented in this paper. Firstly, a fault monitoring model is established by using PCA based on the engine normal samples in order to not only monitor the running condition of the aircraft engine, but also extract fault features. Then, a classifier is built by using SVM based on the score vectors selected as fault feature vectors which is used to identify the engine fault once a fault occurs. The performance of fault diagnosis is tested by the lubrication system of an aircraft engine. The experimental results show that PCA and SVM fault diagnosis method can effectively identify the engine fault and has a good application value.