桁架结构可靠性优化设计的微分演化算法  

Differential evolution algorithm of reliability optimization for truss structure

作　　者：胡长远[1] 唐和生[1,2] 薛松涛[1,3] 邓立新[1] 

机构地区：[1]同济大学结构工程与防灾研究所,上海200092 [2]同济大学土木工程防灾国家重点实验室,上海200092 [3]东北理工大学,日本仙台982-8577

出　　处：《江苏大学学报（自然科学版）》2013年第2期234-238,243,共6页Journal of Jiangsu University：Natural Science Edition

基　　金：国家自然科学基金资助项目(51178337,50708076);科技部国家重点实验室基础研究项目(SLDRCE11-13-01)

摘　　要：针对不确定性往往成为结构设计和可靠性分析的关键因素,并左右设计决策,在结构优化设计中考虑物理参数、几何尺寸和外荷载的不确定性,将其视为随机变数,提出了一种基于微分演化算法的求解不确定性问题的结构可靠性优化设计方法.以节点坐标和杆件截面面积为设计变量,结构重量极小化为目标函数,建立了基于可靠度约束的桁架形状优化数学模型,并对典型桁架进行形状优化,分析了不同可靠度指标和变异系数对优化结果的影响.结果表明:该分析方法能够充分发挥微分演化算法的优势,可以有效地进行基于可靠度的桁架结构形状优化设计.To solve the uncertainties for structural engineering design and reliability analysis and avoid undermining engineering decisions, the structual optimization design was conducted with the consideration of the uncertainties of materials, geometric size and loading. Taking the uncertainties as random variables, a reliability optimization design of truss structure was proposed by differential evolution （DE） algorithm to solve uncertain problems. The node coordinates and the cross-sectional areas of truss were introduced as design variables. The objective of the structural weight was considered to formulate a mathematical model of the shape optimization problem under reliability-based constraints. The typical truss shape optimization problems were conducted, and the influences of reliability index and variance factors on optimum results were also investigated. The results of the numerical example indicate that the proposed method can full play the advantage of the differential evolution algorithm, and can be applied for reliability based truss structure shape optimization design effectively.

关 键 词：桁架结构 微分演化 可靠性约束 形状优化 不确定性 

分 类 号：TU323.4[建筑科学—结构工程] TU311