| Ride comfort is of great importance to the overall performance of a car,and in today’s rapidly developing automotive industry,the importance of car ride comfort research is becoming increasingly apparent.In order to better study the smoothness of the car,this paper takes a certain car as the research object and conducts smoothness simulation tests and optimisation for the car with the help of virtual prototype technology and the integration of ADMAS/Car,MATLAB and ISIGHT software.According to the multi-body dynamics theory,the system of the study car was simplified using ADAMS/Car software according to the analysed smoothness problem.The main subsystems of the model formed are: body,front suspension,rear suspension,steering system,stabiliser bar,powertrain,etc.The harmonic superposition method was used to establish a stochastic input model in MATLAB software,and stochastic road smoothness simulations were carried out separately at different vehicle speeds according to the relevant evaluation criteria.The pulsed input pavement model file was also built in ADAMS pavement modeller and the pulsed pavement smoothness simulation was carried out with this pavement.Based on the simulation results,the smoothness of the studied car was evaluated in conjunction with the evaluation criteria of the smoothness test.The paper is based on the ISIGHT integrated optimisation platform and integrates ADAMS/Car and MATLAB into ISIGHT by writing the relevant documentation.The root mean square value of the weighted acceleration at the driver’s seat,a smoothness evaluation index,is used as the objective function,the influence of the suspension variables on the objective function is determined by applying the optimal Latin superelevation method,the suspension parameters are matched and studied,and these four suspension parameters are used as design variables.Then the constraints in the optimisation process and the optimisation algorithm were analysed in detail,and the selected design variables were optimised and calculated using the multi-island genetic algorithm algorithm.The smoothness of the optimised car was improved to a certain extent,but the efficiency of applying this method for smoothness optimisation was not high.In response to the low efficiency of these methods,the approximate model optimisation method and the combined optimisation method are proposed as the starting point for improving the optimisation model and the optimisation algorithm respectively,and the improvement of the smoothness optimisation method is carried out.The approximate model optimisation method uses the response surface method to fit the whole vehicle model,while the combinatorial optimisation method combines a multi-island genetic algorithm with a sequential quadratic programming algorithm.The two optimisation methods were used to optimise the smoothness of the car,and the efficiency of the optimisation was significantly improved,with the approximate model method being highly efficient and the combined optimisation method being highly accurate.The combined optimisation method was used to optimise the smoothness of the car,and the final design of the suspension parameters was obtained.The optimised suspension spring stiffness and damper damping data were then substituted back into ADAMS/Car software for smoothness simulation again,and the before and after effects were compared,and the total weighted acceleration RMS value at the driver’s seat was reduced by 18.47%.It is verified that this paper is effective in optimising the suspension parameters on a multi-software integrated optimisation platform,and that the application of the combined optimisation method for smoothness optimisation also effectively improves the computational efficiency,and the smoothness of the car is significantly improved after applying the final optimisation scheme. |
PDF Full Text Request