Research on Fault Diagnosis of Electrical System of Medium Transport Aircraft Based on Machine Learning Algorithm

Abstract

The electrical system of medium-sized transport aircraft mainly focuses on electricity. The safe use of electricity must be guaranteed. The reliability of aircraft power supply system is much stricter, because the transportation volume of aircraft is extremely large. At present, two major problems in the research of electrical system fault diagnosis are how to extract signal features and how to establish a diagnostic machine. With the emergence and development of wavelet theory and the increasing maturity of machine learning algorithm, it is an effective and worthwhile solution to preprocess the fault signal by wavelet and then use the machine learning algorithm for fault diagnosis, which provides a new and effective way for fault diagnosis of electrical system. In this paper, a support vector machine (SVM) classification model under the generalized framework is designed, and the parameters of the model are globally optimized by particle swarm optimization. The simulation results show that the fault types can be accurately and orderly identified, thus verifying the effectiveness of the diagnosis model.