索穹顶结构预应力多目标优化的小生境遗传算法  

Multi-objective optimization analysis of cable dome structures prestress based on niche genetric algorithm

作　　者：梁昊庆[1] 董石麟[1] 苗峰[1] 

机构地区：[1]浙江大学空间结构研究中心,浙江杭州310027

出　　处：《建筑结构学报》2016年第2期92-99,共8页Journal of Building Structures

摘　　要：根据索穹顶结构的柔性结构特点和力学特性,探索以质量最小、刚度最大和支座反力最小为目标函数,以预应力整体可行性准则、荷载态应力控制和位移限值为约束条件的多目标优化模型。结合自适应技术、预选择机制和共享函数小生境技术对传统遗传算法进行改进,利用加权系数法将多目标优化问题转换为单目标Pareto最优解问题。利用向量式有限元对不同预应力水平的结构进行静力分析。最后通过两个不同类型索穹顶算例,采用自编程序的计算结果证明所提出的方法是有效的,为索穹顶结构预应力优化分析提供了新的计算方法。According to the flexibility and mechanical property of cable dome structures, a multi-objective optimization model was proposed. The minimum structural weight, maximum stiffness and minimum support reaction were taken as three optimal objectives of the model. The feasible integral structural prestressing, members load stress control and structure displacement limit were taken as three constraint conditions of the model. The traditional genetic algorithm was improved based on the adaptive technique, preselection niche mechanism and sharing function niche mechanism. The weighted coefficient method was applied to transform the multi-objective optimal problem to the singular objective Pareto optimal solution problem. Finally a self-compiled program was applied to solve two given examples. The analysis of optimization results proves the effectiveness and validity of the proposed computing method. The improved niche genetic algorithm provides a new computing method for the stress optimization of cable domes.

关 键 词：索穹顶结构 预应力 多目标预应力优化 小生境遗传算法 加权系数法 向量式有限元 

分 类 号：TU393.3[建筑科学—结构工程] TU311.4