基于非劣排序遗传算法的三代轮毂轴承多目标优化  

Multi-objective Optimization for Third Generation Wheel Hub Bearing Based on NSGA-II

作　　者：林棻[1] 王伟[1] 张尧文 朱为文 

机构地区：[1]南京航空航天大学能源与动力学院,南京210016

出　　处：《南京航空航天大学学报》2013年第6期865-870,共6页Journal of Nanjing University of Aeronautics & Astronautics

基　　金：国家自然科学基金(10902049)资助项目;中国博士后科学基金(2012M521073)资助项目

摘　　要：将某型轿车第三代驱动轮轮毂轴承作为研究对象,以疲劳寿命、磨损寿命、旋滚比为优化目标,引入惩罚函数处理约束条件,将原约束优化问题转化为极小化的无约束优化问题,采用带精英保留策略的非劣排序遗传算法(NSGA-II)进行轮毂轴承多目标优化。通过有限元仿真对优化前后的结构应力情况进行了对比。分析结果表明:在满足规定约束的条件下,提出的优化方案实现了3个目标函数整体性能的同时提升;内法兰、外法兰、内圈、滚珠的应力集中情况均有改善,模型整体最大等效应力较优化前降低8.61%。The third generation driving wheel hub bearing of certain vehicle is considered as research object. Fatigue life, abrasion life and spin roll ratio are considered as the optimization targets. Penalty function is introduced to deal with constraints. Multi-objective optimization is carried out by NSGA-II. Through the finite element analysis, the stress of the pre-optimization structure is compared with the optimal one. The analysis results show that： Under the condition of meeting all the required con- straints, these performances of wheel bearing are improved in different degrees. The stress concentra- tion of inner flange, outer flange, inner ring and balls is reduced. Overall maximum stress of the model decreases by 8. 61% compared with that of the pre-optimization one.

关 键 词：轮毂轴承 多目标优化 非劣排序遗传算法 有限元 

分 类 号：U463.2[机械工程—车辆工程]