桁架结构施工路径优化分析

【作者】 林琳；

【导师】 张云波；

【作者基本信息】 华侨大学 ， 土木结构工程， 2003， 硕士

【摘要】 建筑物的施工顺序对结构的内力、建设成本、安全性等产生很大的影响。因此，存在着一个施工流程(路径)的优化问题。建筑物施工路径的优化解析研究，在优化领域中是个崭新的课题，属于施工力学的优化问题。施工路径方案的制定往往基于以往的业绩和经验。根据试施工中出现的错误作业进行研究和评价。这样，工作量较大，很难判断最终制定的实施方案是否是满足了目标要求的最佳方案。本文将研究用数值解析的方法来寻求满足设定目标的最优施工路径方案。 本文总结了以往关于施工计划的解析研究方法，针对桁架结构，考虑从结构的完成状态逆向探索最佳的施工路径的解析方法，研究施工过程中结构内部参数随时间变化(时变结构)的力学现象，从计算机辅助决策施工计划的角度寻求符合设定目标的最佳施工方案。编制了相应的程序，分析了几个平面桁架结构，并将这种方法拓展运用于空间桁架结构。 遗传算法是近年来兴起的一种寻优方法。本文介绍了遗传算法的基本理论和具体实施，并将此方法应用于桁架结构的施工路径优化分析中。借鉴遗传算法在求解路径优化问题上的成功经验，应用了改进的遗传算子：采用自然数对结构去除构件进行编码，将所求问题的基本信息转换成免疫算子，采用稳态最优保存策略和排序选择方法相结合进行选择，采用适用于本问题研究的特殊的交叉算子和变异算子，从而来讨论在桁架结构施工路径优化问题中使用遗传算法的可能性。编制了相应的程序，与其他优化方法的结果进行了比较，证明此方法的有效性。更多还原

【Abstract】 Construction process planning of structures has an influence upon the cost of the construction, the internal force of the structure, the safety during the construction course etc. So it is very important to find the optimal construction planning. It is a new topic in structure optimization relating to construction mechanics. Construction process is often planned by trial on the basis of previous experience. The work is very tough and could not always give an optimal answer. In this study, our objective is to search for the optimal construction process by a numerical analysis.This paper sums up research in construction planning of structures. An inverse analysis for construction planning by removing members is introduced in order to find an optimum solution of the construction process satisfying given criteria. By studying the mechanical phenomena of structures’ inner parameters varying with time, the optimal construction planning is gained with computer-supported. The corresponding program is applied in several plane truss structures and developed to analyze space truss structures.An optimizing method based on biological evolution-Genetic Algorithm is proposed to the problem of the construction process planning analysis. Adapting the successful experience of path optimization problem, improved genetic operators are adopted: The sequence of numbers, which is arranged by the removal order of members, is assumed to represent a chromosome by natural code; transforming baseline information of this problem into immune operators; combining rank-based model with elitist model; designing special crossover and mutation operators. The mathematical model for computer and special programs are provided in this paper. Based on these improvements, numerical examples are given to show the reliability and efficiency of genetic algorithm in these problems.更多还原