Research Of Auto-body Dimensional Variation On Vehicle Wandering Performance

The dimensional variations of auto-body are inevitably produced in manufacturing process, which have a direct effect on vehicle quality and performance. Vehicle driving wandering is a common type of fault and an important index of vehicle. In a relatively short period of time, the new cars’ vehicle driving wandering performance is affected by the dimensional variations of autobody.The main research of this paper is to build the relationship between dimensional variation and vehicle wandering.Based on the research of dimensional variation design process and front wheel alignment parameters theory, a dimensional variation robust design method for front wheel alignment parameters was proposed. Firstly, through the Mc Pherson suspension assembly relations on the vehicle body, a relationship model of front wheel alignment parameters and measuring point was established. Secondly, setting the suspension assembly as the deterministic locating and taking the body as the locating point, based on the deterministic locating method the relation of measuring point and the locator was also established, then using Monte Carlo Simulation method to research the optimal locating scheme. Finally, the sensitivity of auto-body dimensional variations was achieved by the robust of front wheel alignment parameters, and the robust of dimensional variations solution was also achieved. Based on the example of Mc Pherson suspension assembly locator dimensional variation in a body, the validity of models is verified.The mismatch of left-right front wheel toe and camber will make different tyre lateral force and aligning torque, which will destroy the vehicle’s balance and lead vehicle wandering, therefore, taking the toe and camber as the object of study,and a vehicle driving wandering model for front wheel alignment parameters was proposed. Firstly, the coordinate system of the vehicle and the tire was established. Secondly, the vehicle driving wandering model was established based on the classic 2-wheel-2-DOF vehicle model, though taking the left-right tyre lateral force and aligning torque into account. Finally, based on the orthogonal design and response surface methodology, the second order polynomial response surface model of front wheel alignment parameters and the vehicle driving wandering was established and the impact of the front wheel camber and toe angle to the vehicle deviation was analyzed.A multi-parameter self-adaptable differential evolution algorithm was presented to identify the parameters in Magic Formula 2002 tire model. Firstly, according to the experimental data of a tire, the objective function was established. Secondly, the algorithm process was built, the population of control parameters were constructed, multi-control parameters were used in the variation and crossover operation, during the procedures the optimal individual were selected into the next generation, the success evolutionary control parameters were kept,then they were provided to next generation for selecting at random. Finally, twenty groups Magic Formula 2002 tire model parameters were identified, and the results were compared with the NSADE. Through the analysis, we found the multi-parameter self-adaptable differential evolution algorithm has advantages of fast convergence, higher precision and higher success rate.Through the study above, the first-order response surface model of the body dimensional deviation and the vehicle wandering was established based on the orthogonal design and response surface methodology, by means of this RSM model,the sensitivity of auto-body dimensional deviations to the vehicle wandering was analyzed. Finally, the experiment was done to check laboratory repeatability and volatility level of data of the Driving Wandering Text system which is based on machine vision. The data showed that the calculation method of Driving Wandering Text system was flawed, and it is unable to meet the requirement of accurate measurements.