Abstract: A multi-objective optimization method based on genetic algorithm is proposed to solve the problem that it is difficult to determine the road condition cycle times of vehicle durability test in the current correlated user test field. Based on the collected user road load data, the damage target of different parts in the whole life cycle of the vehicle is extrapolated. Combined with the basic damage matrix of each road condition on the test field, the “test field-user” equivalent damage correlation model is established. The traditional least square method and genetic algorithm are adopted to solve the damage equivalent model, and an optimal criteria is proposed for multiple non-inferior solutions of the genetic algorithm. The effectiveness of the two algorithms is analyzed by comparing the damage on test field and user target. The results show that cycles of each road condition based on generic algorithm is more reasonably, and the damage level of different parts is better reproduced, while the damage on test field according to the least square method is generally small and only a few parts can reproduce the user damage. The research can provide reference and basis for more reasonable establishment of test field durability test specifications and effective evaluation of vehicle reliability and durability

Key words: user correlated, damage equivalent, rurability test, road cycles, optimization method