基于多岛遗传算法汽车动力总成悬置解耦优化  

Decoupling Optimization of an Automotive Powertrain Suspension Based on Multi-island Genetic Algorithm

作　　者：孙永厚[1] 杨帅 刘夫云[1] 赵士超 SUN Yong-hou;YANG Shuai;LIU Fu-yun;ZHAO Shi-chao(School of Mechanic and Electronic Engineering,Gulin University of Electronic Technology,Guangxi Guilin 541004,China)

机构地区：[1]桂林电子科技大学机电工程学院

出　　处：《机械设计与制造》2019年第12期155-158,162,共5页Machinery Design & Manufacture

基　　金：国家自然科学基金资助项目(51265006);广西科技开发项目(1598007-51);广西科学研究与技术开发项目(桂科AC16380078)

摘　　要：乘坐舒适性的需求受到人们越来越广泛的关注,动力总成的振动是影响乘坐舒适性的主要因素,动力总成解耦优化已成为了学者们研究的热点。由于传统的优化算法难以满足动力总成复杂非线性关系下的工程需求,并且速度慢、难以得到最优解,该研究主要基于多岛遗传算法,对企业某型号动力总成悬置系统进行解耦优化研究。首先,通过对动力总成动力学理论分析,建立相应动力学模型;然后,基于能量法解耦理论,建立以刚度为设计变量、频率配置为约束条件以及最大解耦率为目标函数的MATLAB数学模型;最后,采用多岛遗传算法进行解耦优化,通过MATLAB与ADAMS进行对比验证。优化结果表明,解耦运算效率和精度得到显著提高,刚度组合能够更好地满足工程实际需求。There are more and more attentions on the ride comfort,whose main factor is the vibration of powertrain which is become a hotspot forscholars.The traditional optimization algorithm is difficult to meet engineering requirements of complex nonlinear relations of powertrain,and its speed is slow,and,difficult to get the optimal solution,so this research is the mainly based on the multi-island genetic algorithm to optimize the decoupling of a certain type of powertrain mounting system.First,analyzing the dynamics of the powertrain establish the corresponding dynamic model;then based on the theory of the energy decoupling to establish the MATLAB mathematical model with stiffness as the design variable,and,the frequency allocation as constraint conditions,the maximum decoupling rate as the objective function;finally,the Multi-island Genetic Algorithm has used to decouple,through the comparison MATLAB and ADAMS.The optimization results show that the efficiency and accuracy of decoupling can be improved obviously,and the combination of stiffness can meet the actual requirements of the engineering.

关 键 词：动力总成 悬置刚度 解耦优化 多岛遗传算法 

分 类 号：TH16[机械工程—机械制造及自动化] U461.4[机械工程—车辆工程]