基于CLPSO算法的混合变量桁架形状优化  

A CLPSO algorithm for truss structure shape optimization with mixed variables

作　　者：许锐[1,2] 马安峰[3] 谢鹏 高福如 薛松涛[2,5] 

机构地区：[1]长安大学地质工程与测绘工程学院,陕西西安710054 [2]同济大学结构工程与防灾研究所,上海200092 [3]陕西省地矿局西安中勘工程有限公司,陕西西安710016 [4]中国有色金属工业西安勘察设计研究院,陕西西安710016 [5]日本东北工业大学,日本仙台982-8577

出　　处：《燕山大学学报》2012年第6期547-555,共9页Journal of Yanshan University

基　　金：国家自然科学基金资助项目(50708076)

摘　　要：为了解决混合变量桁架形状优化问题中离散截面面积和连续节点坐标的变量耦合给优化带来的困难,将一种新型智能优化算法——基于"综合学习策略"的粒子群算法(ComprehensiveLearning Particle SwarmOptimization,CLPSO)应用于桁架混合变量形状优化问题中。给出了考虑离散截面面积和连续节点坐标两类不同性质的设计变量的混合变量桁架结构形状优化的数学模型,并对经典桁架结构进行混合变量的形状优化,将所得结果与其他优化算法结果进行了比较。分析结果表明了该方法进行混合变量桁架形状优化设计的有效性。In order to overcome the difficulties encountered by the coupling of two distinct types of design variables, the discrete section area and continuous node coordinates, in the shape optimization of truss structures with mixed variables, a novel intelligent optimization method, comprehensive learning particle swarm optimization （CLPSO） is introduced in this paper. The basic principle of CLPSO algorithm is presented in detail first, and then mathematical model for shape optimization of truss structures is presented, in which two distinct types of design variables, the discrete section area and continuous node coordinates, are considered simu- ltaneously. Several classical problems were solved for shape optimization with mixed variables, and the results are compared with those using the other optimization methods. The effectiveness of the proposed method is evaluated through the numerical analysis.

关 键 词：CLPSO算法 形状优化 桁架结构 变量耦合 离散变量 混合变量 

分 类 号：TU323.4[建筑科学—结构工程] TU311