基于多目标遗传算法的约束阻尼柱壳结构优化  

Optimization of constrained layer damping cylindrical shell based on multi-objective genetic algorithm

作　　者：石慧荣[1] 罗冠炜[1] 高溥[1] 李宗刚[1] 张军平[1] 

机构地区：[1]兰州交通大学机电工程学院,兰州730070

出　　处：《船舶力学》2015年第1期169-175,共7页Journal of Ship Mechanics

基　　金：国家自然科学基金(11172119;11362008);甘肃省创新研究群体计划(1308RJIA006);高等学校博士学科点专项基金(20126204110001);兰州交通大学青年科学基金项目(2012014)

摘　　要：文章为了改进被动约束阻尼柱壳结构的减振特性,解决引入约束阻尼的合理分布问题,提出了一种基于多目标遗传算法的优化设计方法。建立了以振动模态的前两阶损耗因子和引入约束阻尼材料总质量比为优化目标,以约束层、阻尼层厚度和粘弹性材料剪切模量为设计变量的多目标函数。利用遗传算法对Pareto最优进行优化选择,避免了对目标函数值的直接计算。通过分析和比较优化前后结构的模态振动频率变化以及幅频响应,表明引入质量太多会影响柱壳结构固有特性,而优化后结构能够避免太多的质量引入,并且具有更好的减振效果。Based on the multi-objective genetic algorithm, an optimization design method is proposed to im-prove the vibration damping characteristics of passive constrained layer damping cylindrical shell and solve the rational distribution problem of the constrained layer damping. A multi-objective function is es-tablished, in which the objective of optimization is the loss factor of the first two modes and the mass ratio of constrained layer damping material, and the design variables are the thickness of the constraint layer, the damping layer and the shear modulus of the viscoelastic material. The proposed method finds the true Pareto front by using the genetic algorithms and avoids the direct calculation of objective function. By analyzing and comparing the modal frequency changes and the amplitude-frequency response of cylindrical shell be-fore and after optimization, the results show that the inherent characteristics of the cylindrical shell become worse if the introduced mass of damping materials is too much. In contrast, the structure optimized by this method introduces the less of damping materials and achieves the better vibration reduction.

关 键 词：约束阻尼 多目标遗传算法 圆柱壳 优化设计 振动 

分 类 号：O328[理学—一般力学与力学基础]