FAULT DIAGNOSIS METHOD OF AERO-ENGINE ROLLING BEARINGS BASED ON TIME-FREQUENCY ANALYSIS AND MACHINE LEARNING

As a crucial component of aircraft, aero-engine bearings operate under extreme conditions such as high temperature, high pressure, and high rotational speed, making them highly prone to failure, which seriously affects aviation safety. Traditional bearing fault diagnosis methods suffer from problems such as low diagnostic accuracy and poor real-time performance, and it is difficult to meet the requirements of modern aviation industry for high reliability and safety of engines. With the development of machine learning technology, this paper proposes a fault diagnosis method for aero-engine bearings based on machine learning. Firstly, time-domain and frequency-domain features of vibration data are extracted, and dimensionality reduction processing is carried out through principal component analysis (PCA) to reduce data complexity and retain key information. Subsequently, machine learning models such as logistic regression, K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and decision tree are used for fault prediction, and a comparative analysis is conducted with deep learning models. The experimental results show that the Support Vector Machine (SVM) performs best in the fault classification task, with an accuracy rate of 99%. This research provides an efficient and accurate solution for aero-engine bearing fault diagnosis and has important practical application value.