基于遗传算法的混合动力汽车动力总成悬置系统的优化设计研究  

Optimization design for powertrain mounting system of a hybrid electric vehicle via genetic algorithm

作　　者：庄伟超 王良模[1] 殷召平[2] 叶进[2] 吴海啸[2] 

机构地区：[1]南京理工大学机械工程学院,南京210094 [2]南京依维柯汽车有限公司,南京210028

出　　处：《振动与冲击》2015年第8期209-213,共5页Journal of Vibration and Shock

基　　金：江苏省教育厅高校科研成果产业化推进工程项目(JHB2012-3)

摘　　要：为了提高并联式混合动力汽车动力总成五点悬置系统的隔振性能,建立了动力总成五点悬置的动力学模型,以动力总成六自由度能量解耦与固有频率的合理分配为优化目标,五个悬置点的各向刚度为设计变量,采用遗传算法对悬置系统进行优化。应用上述方法对某并联式柴电混合动力汽车悬置系统进行了优化,动力学仿真与实车试验结果表明,悬置优化后消除了整车怠速工况时方向盘抖动,验证了所提方法的合理性。同时,遗传算法克服了序列二次型规划算法(SQP)易收敛于局部最优解的缺点,得到的悬置系统解耦性能优良,优化结果稳定可靠。A method to optimize a powertrain mounting system was developed to improve the vibration isolation performance of the mounting system for a parallel hybrid electric vehicle.The optimization was based on the genetic algorithm by taking 6-DOF energy decoupling of the powertrain mounting system and the reasonable allocation of the natural frequencies as the objectives,and taking the stiffnesses of 5 mounting points as the design variables.This method was applied to deal with the shaking of steering wheel for a parallel hybrid electric vehicle in its idling process.The results of dynamic simulation verified the effectiveness of the method.Furthermore,it was shown that compared to the sequential quadratic programming （SQP）,the genetic algorithm overcomes its shortcoming of converging to local optimal solution, the decoupling features of the mounting system obtained from the optimization are better and reliable.

关 键 词：动力总成悬置 能量解耦 优化 遗传算法 

分 类 号：U461.1[机械工程—车辆工程]