Research on Fault Diagnosis of Flight Control System Based on SVM Optimization

This paper designs a fault diagnosis model based on the SVM algorithm. The data is obtained through simulation and preprocessed to obtain training data. Then, based on the SVM multi-class model, different kernel functions are used for analysis, and various types of kernel functions are analyzed and compared difference between. After that, the advantages of the Adaboost classification algorithm were analyzed, and the advantages of the SVM classification algorithm were complementary. A hybrid model of SVM-Adaboost fault diagnosis was proposed, and simulation experiments were performed on the obtained model. It is confirmed that the model does improve the accuracy of fault diagnosis.