| With the increasing popularity of the car,car safety has been a hot issue in our life.Most of our studies focus on the driver safety system,while the rear occupant's safety system receives little attention.Until mid-2012,China's C-NCAP has not taken the rear occupants' injury situation into the scope of the score.Compared with the front row,the rear row occupant's restraint system is less sophisticated with worse material and less appropriate param-pattern.Based on the car whose rear occupant's restrain system is not qualified for the C-NCAP,the verified vehicle finite element model was established.After that,we presented the WIC to evaluate the comprehensive injury of the dummy based on the scoring system of C-NCAP and E-NCAP.Through the sensitivity analysis,the stiffness of the seat belt,the height of the seat belt's fix point,the limit force of the retractor and the preload of the pretensioner were selected as the design variables.Next,the orthogonal experiment design method was adopted to arrange the experiments and build response surface models.Finally,SQP method was used to optimize WIC.As a result,WIC was reduced by the ratio of 20.74% and all the dummy response were qualified.After the deterministic optimization,the reliability of the optimization was analyzed.It was found that the reliability of the neck bending moment was very low when the design variables were surged.To solve this problem,6? robustness optimization design was adopted to optimize original problem.The goal of 6? robust optimization is the minimum value of weighting of WIC and its standard deviation,and the constraint response is at least 6? away from constraint boundary under the fluctuation of design variables.The robust optimization result shows that the WIC is reduced by 9.56% compared with the primary,while all the dummy's responses are qualified in 6? range.The robust optimization plays a greater role in practical application and has higher engineering application value than deterministic optimization. |
PDF Full Text Request