Reliability Study of Critical Components of Urban Rail Vehicle Based on Improved SCSO Algorithm

The existing reliability model of bogie components of urban rail vehicles has shortcomings such as low estimation accuracy, poor versatility, and weak stability, resulting in the model being unable to predict the failure time of vehicle components accurately. To solve this question, a reliability parameter estimation method for train critical components based on the improved sand cat swarm optimization (ISCSO) algorithm is proposed in this paper. The original SCSO algorithm has problems such as unstable optimization results and easy falling into local optimality. The algorithm’s global optimization ability is improved by introducing the chaos map, spiral search strategy, and sparrow early warning mechanism. Using the failure data of the component of train bogie system as a case, the reliability curve of each component is solved, which verified the model’s feasibility and the algorithm’s superiority. The research outcome describes that the parameter estimation model of train critical components based on the ISCSO algorithm has high accuracy and fast convergence speed. Compared with other methods, the presented algorithm fits the reliability of different components of the train to a high degree. It is more versatile, providing a theoretical basis for the reliability analysis of train critical components.