基于点云和BIM的大跨钢桁架虚拟预拼装应用研究  

Intelligent Virtual Preassembly Application of Long Span Steel Truss Based on Point Cloud and BIM

作　　者：黎翔 呼辉峰 陈天国 刘界鹏[1] 余鹏 赵国军 Xiang Li;Huifeng Hu;Tianguo Chen;Jiepeng Liu;Peng Yu;Guojun Zhao(School of Civil Engineering,Chongqing University,Chongqing 400045,China;School of Civil Engineering,Changsha University of Science and Technology,Changsha 410114,China;China Construction Steel Structure Corporation Ltd.,Chongqing 402160,China;Power China Road Bridge Group Corporation Ltd.,Chongqing 402260,China)

机构地区：[1]重庆大学土木工程学院,重庆400045 [2]长沙理工大学土木工程学院,长沙410114 [3]中建钢构股份有限公司,重庆402160 [4]中电建路桥集团有限公司,重庆402260

出　　处：《钢结构（中英文）》2023年第10期10-15,共6页Steel Construction（Chinese & English）

基　　金：国家自然科学基金项目(52208184)。

摘　　要：大型复杂钢结构具有构/部件数量多,焊接、栓接工作量大,施工工序多等特点。传统的预拼装方法成本高、效率低,难以满足大型复杂空间结构的要求。针对该问题,提出基于BIM和三维激光扫描技术的大型复杂钢结构虚拟预拼装方法。以西部(重庆)科学城科学会堂项目为工程背景,对屋面圆管桁架和大跨度楼面桁架智能化施工流程进行研究,提出了技术路线和方法流程。针对工程常见的圆管桁架,提出采用基于随机采样一致性算法、拉普拉斯-滚球混合检测算法的拼接控制点提取方法;针对形状规则的楼面工字钢桁架和牛腿,提出采用有向包围盒法获取拼接控制点;为获取施工质量检测数据,提出采用基于k近邻算法、主成分分析算法和基于区域准则的边缘检测算法完成对接口角点的智能提取。结果表明:将BIM和三维激光扫描技术相结合的技术路线对圆管桁架和工字钢桁架的虚拟预拼装均取得了较准确结果;采用提出方法获取的圆管桁架基准点,配合基准点全排列算法进行粗配准,再通过迭代最近邻算法进行精配准,从而获得较高的配准质量。Large and complex steel structures have the characteristics of a large number of structural components,a large workload of welding and bolting,and multiple construction processes.Traditional preassembly methods not only have high costs and low efficiency,making it difficult to meet the requirements of large and complex spatial structures.To address this issue,a pre-assembly method for large and complex steel structures based on BIM and 3D laser scanning technology is proposed.Taking the Western(Chongqing)Science Hall project as the engineering background,the intelligent construction process of roof circular pipe truss and large-span floor truss was studied,and the technical route and method flow were proposed.A benchmark extraction method based on random sampling consistency algorithm and Laplacian rolling ball hybrid detection algorithm is proposed for common circular pipe trusses.A directed bounding box method is suggested to obtain the registration reference points for the regular shaped I-shaped steel truss and bracket point cloud.To obtain construction quality inspection data,k-nearest neighbor algorithm,principal component analysis algorithm,and edge detection algorithm based on region criteria to intelligently extract interface corners are suggested.The results indicate that the combination of BIM and 3D laser scanning technology has achieved relatively accurate results in virtual pre assembly of circular tube trusses and I-shaped beam trusses;The cylindrical truss reference points obtained using the proposed method,combined with the full arrangement algorithm of the reference points and the iterative nearest neighbor algorithm,can achieve high registration quality.

关 键 词：大型复杂钢结构 点云数据 智能虚拟预拼装 施工尺寸质量检测 

分 类 号：TU17[建筑科学—建筑理论] TU758.11