摘 要

随着建筑业的发展，复杂钢结构建筑越来越多，而大部分钢结构构件在现场安装前需进行预拼装，以保证后期施工的顺利进行。然而，由于钢结构构件体积较大，实地预拼装不易，且消耗大量的人力、物力及场地，拼装时，工人作业健康安全保障难，加重环境负担。因此，本文提出离线拼装方法，针对不同构件采取不同方法，避免了实地预拼装，实现了测量、分析、拼装的一体化方案。

本文首先以HSE（健康、安全、环境）理论和EBA（能量隔离分析）理论为基础，以现有预拼装工艺流程及拼装方式为研究对象进行风险分析，探讨其中影响HSE的主要因素、主要事故类型、能量累积及传播途径等，进而阐述风险防范措施，为后续提出离线拼装方法提供理论基础与指导性框架。然后本文构建了离线拼装方法，针对小尺寸及复杂形状构件、大尺寸及深孔构件采用不同测量处理方式。对于小尺寸及复杂形状构件采用三维扫描测量和点云数据处理，形成构件点云模型，进而与BIM模型比对；对于大尺寸及深孔构件，采用便携光笔实时采点测量比对，并用改进TLS迭代算法进行空间点拟合，修正BIM模型。完成测量与数据处理后，以点云模型或修正的BIM模型进行模拟拼装，根据拼装结果及拼装误差值，形成拼装误差报告。对于满足拼装要求的构件，以BIM技术和拼装报告为基础指导现场施工；对于不符合拼装要求的构件，出具矫正方案，指导工厂进行修改。最后，本文将离线拼装方法用于武汉某钢箱梁桥项目，并对离线拼装方法进行效益评价和风险分析。

结果表明，离线拼装方法能有效提高预拼装的效率，改善了传统拼装方法工艺繁琐、耗时长、成本高、质量控制难、作业空间使用率低等不足；有效地保障了工人的健康、安全，减少了对环境的负担，从源头上解决了实地预拼装中存在的风险，为现场施工及钢结构件制作产业的成长发展提供了远景。

关键词：钢结构离线拼装BIM技术三维测量HSEEBA 改进TLS迭代算法

Abstract

With the development of the construction industry, there are more and more complex steel structure buildings. To ensure the success of the construction, most of the steel structure components need pre-assembly in factories before on-site installation. Due to the large size of steel structure, on-site assembly is faced with great difficulty that requires a lot of manpower, resources and space. On the other hand, the unqualified process of installation can have terrible influence on environment and workers' health. In contrast, the offline-assembly method of steel structure components provides another promising solution that avoids pre-assembly on site and achieves detection program integrating measurement, analysis and assembly. This research adopts different measurement and data processing methods for different components to instruct the construction of the steel structure and improve efficiency of steel structure assembly.

Based on HSE (Health Safety Environment) and EBA (Energy Barrier Analysis) theory, the risks of on-site pre-assembly are first assessed to conclude key factors influencing HSE, dominating types of accidents, and ways of energy accumulation and propagation. Then, the risk prevention measures are elaborated to provide the theoretical basis and instructional framework for the proposed offline-assembly method which adopts different measurement and processing methods for small and complex-shaped components, large size and deep-hole components.For the small and complex-shaped components, three-dimensional scanning measurement and point cloud data processing have been used to form point cloud model of components to make comparison with BIM model.For the large size and deep-hole components, the portable light pen is used to measure and compare points on components. The improved TLS iterative algorithm is developed in this study for spatial point fitting to modify the BIM model. After measurement and data processing, assembly results and error reports can be outputted based on the virtual assembly of point cloud model or modified BIM model. For the components satisfying the assembly requirements, the assembly reports with assembly errors is used to guide on-site construction combined with BIM technology. In terms of components that cannot be assembled, components correction can be conducted according to the error reports. Finally, this study verifies this method based on some steel box girder bridge components in Wuhan. The benefits evaluation of offline-assembly method and risks analysis are conducted simultaneously.

The results show that the offline-assembly method can greatly improve the efficiency of pre-assembly, the quality of the project and decrease costs of production and management. Comparing with the transitional techniques which is complicated, consuming more time, high cost, difficult to control error, low usage of space and so on, offline-assembly is more healthy, safely and environment friendly. This method reduces risks fundamentally and provides hopeful prospects for the further development of steel structure production industry.

Key words:Steel structure Offline-assembly BIM technology

Three-dimensional measurement HSE EBA

Improved TLS iterative algorithm