Risk identification of civil aviation engine control system based on particle swarm optimization-mean impact value-support vector machine

Abstract

The actual risk of the engine control system in operation is often higher than expected risk level, which brings hidden dangers to the safe operation of aircraft. Therefore, a PSO-MIV-SVM (particle swarm optimization-mean impact value-support vector machine) model is proposed to identify the risk of engine control system. Firstly, seven characteristic variables are extracted through the analysis of engine unsafe information, and 810 typical engine event samples were selected and normalized. Secondly, the SVM-based engine control system risk identification model is established and optimal kernel function of the SVM model is selected. Thirdly, the SVM-MIV method is used to sort the importance of the seven characteristic variables, and the identification accuracies of different characteristic variable groups are calculated to obtain the optimal combination of characteristic variables. Finally, the parameters of the SVM model are optimized by using the PSO algorithm and the accuracy of the PSO-MIV-SVM model for the risk identification of the engine control system reaches 93.58%.