基于改进的Pareto遗传算法的车身气动多目标优化  

Multi-Objective Aerodynamic Optimization of Car Body Using Improved Pareto Genetic Algorithm

作　　者：韦甘[1] 杨志刚[1] 李启良[1] 

机构地区：[1]同济大学上海地面交通工具风洞中心,上海201804

出　　处：《汽车工程》2014年第10期1243-1247,共5页Automotive Engineering

基　　金：国家"973"重点基础研究发展计划(2011CB711203)资助

摘　　要：把气动性能和空间性能作为优化目标,用改进的Pareto遗传算法求解得到流线型车型和普通车型无轮车身的Pareto波阵面。通过二维车身优化算例验证,以小家族为单位进行进化的邻点交叉法、分象限外推法和单目标预测法配合使用,可以有效提高子代的分布性能和进化效果,解决三维车身气动优化中计算量过大的问题。流线型车型与普通车型的Pareto波阵面相比,在中低阻区,前者的综合性能更优秀,在高阻区则反之。Improved Pareto genetic algorithm is used to conduct a simulation with the aerodynamic and spatial performances of car body as objectives,and the Pareto fronts of wheel-less car bodies with both conventional and streamline styling are obtained. A 2D car body optimization example verifies that the adoption of neighbouring-point crossing and quadrant allocation /extrapolation combined with single objective prediction can effectively enhance the distribution performance and evolution effects of offspring,and solve the problem of heavy computational efforts in3 D car body aerodynamic optimization. In the comparison between the Pareto front of car body with streamline styling and that with conventional styling,the formers overall performance is better at low-and middle Cdregion,and is just the opposite at high Cdregion.

关 键 词：PARETO最优解 遗传算法 气动优化 多目标优化 

分 类 号：U463.82[机械工程—车辆工程]