索穹顶结构多目标形状优化设计  

Multi-objective shape optimization design of cable dome

作　　者：梁笑天[1] 袁行飞[1] 李阿龙[1] 

机构地区：[1]浙江大学空间结构研究中心,浙江杭州310058

出　　处：《华中科技大学学报（自然科学版）》2016年第7期110-115,共6页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(51278461;51578492);浙江省自然科学基金资助项目(LZ14E080001)

摘　　要：为了得到指定结构跨度下的索穹顶结构形状最优设计方案,对设内拉环的Geiger型索穹顶进行单目标和多目标形状优化设计.以最小单位面积用钢量和最大结构刚度为目标函数,得到满足约束条件的整体预应力系数、拉索和压杆截面面积及结构形状参数的最优值.通过基于遗传算法的单目标形状优化设计以最小单位面积用钢量为目标的可行形状优化设计方案;多目标形状优化设计采用改进的带精英策略的快速非支配排序遗传算法(NSGA-Ⅱ),可得到同时考虑结构用钢量和刚度的Pareto最优解.决策者可根据不同控制目标的偏好要求,选择设计方案.A comprehensive optimization procedure was employed for shape optimization design of Geiger cable dome with inner hoop.Single objective and multi-objective optimization was implemented to obtain the optimum design for the cable dome with specified span length.Design variables of the algorithm determined simultaneously in the search process,were integral coefficient of pre-stressing force,cross section areas of cable and strut and shape parameters.Minimum steel consumption per unit square and maximum stiffness of the structure was chosen as objective functions.The design was firstly acquired for minimum steel consumption per unit square satisfying constraint conditions by using genetic algorithm（GA）of single objective shape optimization.Furthermore,Pareto front solutions which could represent the designs both considering the steel consumption and stiffness of the structure were achieved for multi-objective by using a variant of non-dominated sorted genetic algorithm-Ⅱ（NSGA-Ⅱ）.Final design could be chosen for the designer according to the preference for different target objectives.

关 键 词：索穹顶结构 形状优化 多目标 遗传算法 PARETO最优解 

分 类 号：TU393.3[建筑科学—结构工程]