Robust Optimization Design For Fatigue Life Of A Car Body Based On Weibull Distribution

As one of the three key automotive assemblies,the fatigue life performance of the body assembly directly determines the service life and safety of the vehicle.Since the fatigue life distribution of a structure is more dispersed,probabilistic statistical methods are commonly used to describe the fatigue life dispersion.The Weibull distribution,which can take different shapes depending on the shape parameter,has become the most used statistical model for fatigue life distribution.In addition,factors such as vehicle design errors,material property deviations and human operation errors can all have an impact on the robustness of the fatigue life of a body.In this paper,a robust optimization design model for the fatigue life robustness of car bodies is proposed based on the Weibull distribution.Firstly,the equivalent Weibull step-bystep method is given,a simulation model of the whole vehicle dynamics and a virtual road surface are established,and the load spectrum is obtained by vehicle kinematic simulation.The fatigue life of the body is then obtained based on the principle of inertial release analysis of car bodies,combined with S-N curves,load spectra and Miner’s linear cumulative damage law.Finally,a genetic algorithm is used to perform robustness optimization using body fatigue life as the study object.The main research elements of this paper are as follows:（1）The equivalent Weibull distribution method is given to obtain the shape,location and scale parameters of the Weibull distribution based on the principle of equality of mean,variance and third-order moments,which in turn gives an equivalent Weibull distribution to the original distribution function.Numerical examples verify that the Weibull distribution is more accurate than the normal distribution in fitting other distributions.（2）Based on kinematic theory,sub-systems such as front and rear suspension,steering system and tire system of the car are established,and then a multi-rigid body motion simulation model of the whole car is established.Then,according to the random road surface unevenness theory to generate B,C,D and E level road surface,so as to establish a virtual simulation test field,the whole vehicle model in the virtual test field for motion simulation,to get the body fatigue life analysis necessary for the time-load history.（3）A finite element model of the body is developed and an inertial release analysis is carried out to obtain the results of the static analysis of the body.Based on the nominal stress method,the fatigue life analysis of the body is carried out by combining the load spectrum,Miner’s cumulative damage rule and S-N curve,and the areas where fatigue failure occurs are indicated.（4）A robust optimization model for body fatigue life is proposed based on the Weibull distribution.The design variables are selected by sensitivity analysis,and the material parameters are used as random variables.The Kriging agent model for body fatigue life is constructed by Latin hypercube sampling,and the body fatigue life robust optimization model is established by considering two aspects,namely the minimum variance of body fatigue life and the concentration of fatigue life distribution around the expected life.Finally,according to the genetic algorithm,the final results of body part plate thickness optimization are obtained and the optimized body life is verified to achieve the robust optimization objective.