基于多种群协同优化的叠合板智能拆分设计  

Intelligent Splitting Design for Precast Composite Slab Based on Multi-population Co-evolutionary Algorithm

作　　者：刘界鹏[1] 张超 郑小磊 齐宏拓 伍洲 李新伟 何亮 许成然 LIU Jiepeng;ZHANG Chao;ZHENG Xiaolei;QI Hongtuo;WU Zhou;LI Xinwei;HE Liang;XU Chengran(School of Civil Engineering,Chongqing University,Chongqing 400044,China;School of Automation,Chongqing University,Chongqing 400044,China;China Construction Science&Technology Co Ltd,Beijing 100070,China;China Construction Science&Technology Co Ltd,East China Branch,Shanghai 201100,China;School of Civil Engineering,Zhejiang University of Technology,Hangzhou 310014,China)

机构地区：[1]重庆大学土木工程学院,重庆400045 [2]重庆大学自动化学院,重庆400045 [3]中建科技集团有限公司,北京100070 [4]中建科技集团华东有限公司,上海201100 [5]浙江工业大学土木工程学院,浙江杭州310014

出　　处：《土木工程与管理学报》2024年第5期1-8,共8页Journal of Civil Engineering and Management

基　　金：国家自然科学基金青年项目(52208219);国家重点研发计划(2021YFF0500903)。

摘　　要：在装配式建筑中,叠合楼板的拆分主要是基于规则的半自动方式完成,其设计过程耗费大量时间和人力,难以获得最优解;同时,采用基于规则的方法存在构件规格多的问题,增加了构件加工的成本和时间。为了解决上述问题,本文提出基于多种群协同优化的叠合楼板智能拆分方法,以提高叠合楼板的标准化和模数化程度。将叠合板规格和数量、接缝宽度规格作为优化目标,每个叠合板位置和尺寸作为优化变量,考虑单构件的限重、限宽等加工和运输要求,建立多目标优化模型。采用变量关联性分析对叠合楼板所有布置区域进行分组划分,原高维优化问题转化为低维优化问题。通过上海某装配式建筑住宅项目的叠合楼板拆分实例应用,表明本文提出的叠合楼板智能拆分方法能够高效、快速地生成全部楼板的规格、数量和位置;相比基于规则的设计方法,本文提出的方法能够一定程度提高叠合板的标准化和模数化水平,且拆分结果满足设计规范和施工要求,从而验证了智能拆分方法的可行性和有效性,有利于装配式建筑的推广。In prefabricated buildings,a rule-based semi-automatic method is widely used in the splitting design of precast composite slab(PCS),which is time-consuming and difficult to obtain the optimal solution.And the rule-based method has the problem of many component specifications,leading to additional manufacturing cost and time.In order to solve this problem,a multi-population co-evolutionary approach is proposed to improve the standardization and modularization of PCSs.The specifications,quantities,and joint widths of PCSs are considered as optimization goals,and the position and dimension of each PCS are regarded as optimization variables.A multi-objective optimization model is established to consider the requirements of manufacturing and transportation,such as weight and width limit of single component.The original high-dimensional optimization problem is transformed into a low-dimensional optimization problem by grouping and dividing all the layout regions of PCSs using variable correlation analysis.An example of composite floor splitting in a prefabricated building residential project in Shanghai shows that the proposed splitting approach can efficiently and rapidly generate the specifications,quantity and position of PCSs.Compared to rule-based splitting,the method proposed in this paper can improve the standardization and modularization of PCSs to some extent,and the splitting results meet the design codes and construction requirements.Therefore,the feasibility and effectiveness of the intelligent splitting method is verified and the proposed method facilitates the promotion of prefabricated buildings.

关 键 词：智能建造 叠合楼板拆分设计 多种群协同优化算法 关联性分析 粒子群优化算法 

分 类 号：TU712.1[建筑科学—建筑技术科学]