Optimization of low-floor vehicle dynamic performance based on Kriging model and NSGA-II algorithm

Abstract

Echoing the pressing need to improve optimization efficiency and change the method of vehicle dynamic performance optimization by varying single parameter, a new method based on the Kriging approximate model and NSGA-II genetic algorithm is proposed to obtain more accurate results and shorten the design cycle. Accordingly, taking the suspension parameters as the variables, the dynamic performance indexes as the responses to fit the Kriging approximate model, the dynamic model of the 70% low-floor vehicle has been established. Subsequently, multi-parameter, multi-objective optimization of vehicle dynamic performance is achieved by virtue of the approximate model. The results show that, compared with the original method, all indexes are improved except the vertical comfort and stability. In particular, the optimization efficiency is significantly enhanced, and the effectiveness and feasibility of applying the optimization method to vehicle dynamic performance by virtue of the Kriging model have been proved.