基于响应面法的活动横梁动静态特性多目标优化  

Dynamic and static multi- objective optimization of the activity crossbeam based on response surface method

作　　者：孙康[1] 靳超锋 吴欣[1] 

机构地区：[1]江西理工大学机电工程学院,江西赣州341000

出　　处：《制造技术与机床》2015年第12期71-76,共6页Manufacturing Technology & Machine Tool

基　　金：国家科技重大专项(2009ZX04005-031);国家重点实验室开放基金(GZKF-201023);江西省自然科学基金(20122BAB206030)

摘　　要：为了满足300MN液压机活动横梁在具有高的动态特性的基础上,还使结构具有轻量化的要求,提出一种基于响应面法的动静态特性多目标优化方法。首先采用中心组合试验设计方法(CCD)得到样本试验点和各试验点的试验数据,采用二阶响应面建模的方法建立活动横梁各性能指标的响应面数学模型。将活动横梁的最低固有频率和活动横梁的整体质量作为目标函数,最大应力为约束函数,结合响应面模型,采用Shifted Hammersley序列抽样技术和权衡函数结合的遗传算法对活动横梁进行多目标优化,得到Pareto的优化解集,经过优化可使活动横梁的质量减轻12.2%,固有频率增加27.22%,使活动横梁在具有高的动静态特性的基础上还具有轻量化。In order to satisfy the performance requirement of high dynamic and static characteristics and light weight of the activity crossbeam. a method of multi- objective optimization based on the response surface method driven by dynamic and static characteristics is proposed in this paper. First sample sites and the sites of test data is obtained by the central composite experimental design method（ CCD）,and the response surface mathematical model of the activity crossbeam each performance index was built by adopting the method of quadratic response surface modeling. The overall quality and the lowest natural frequency of the activity crossbeam as the objective function,the maximum stress as the constraint function,Combined with the response surface model,the multi- objective optimization of the activity crossbeam is researched by using Genetic algorithm combined Shifted Hammersley sampling technique with tradeoffs function. The Pareto optimal solution set is obtained,through optimization the mass of the activity crossbeam is decreased by 12. 2% and its Natural frequency is increased by 27. 22%,the performance requirement of high dynamic and static characteristics and light weight of the activity crossbeam was obtained.

关 键 词：300MN液压机 响应面法 优化设计 多目标优化 遗传算法 

分 类 号：TG315[金属学及工艺—金属压力加工]