首页|基于视觉识别的高大模板支撑结构安全监控方法

基于视觉识别的高大模板支撑结构安全监控方法

扫码查看
高大模板支撑结构的坍塌破坏事故时有发生,为保证高大模板支撑结构的施工安全,提出了一套基于计算机视觉和有限元技术的施工智能实时监测方法.根据高大模板支撑结构的杆件排布真实情况建立精细化的有限元模型并进行多种工况下的屈曲分析,得到高大模板支撑结构屈曲破坏时关键节点的竖向位移,获得施工智能实时监控的报警阈值.基于计算机视觉的智能实时监控算法,可实时监测高大模板支撑结构施工过程中的位移变化,当出现不安全位移时算法将自动报警指导排除危险、保障施工安全.施工结束后根据理论计算结果与实测结果,对有限元模型进行参数修正,并对修正后的有限元模型进行屈曲破坏分析,对比修正后屈曲竖向位移与报警阈值大小,说明报警阈值的合理性.最后将该方法应用于余杭智能电网产业基地项目C区块(国网浙江省电力有限公司综合科研试验用房)项目东侧裙房超高大模板支撑结构中,验证了所提出的施工监控新方法的有效性.
A Safety Monitoring Method for High-Formwork Support Structures Based on Computer Vision Recognition
To ensure the construction safety of high-formwork support structures,a set of construction intelli-gent real-time monitoring method based on computer vision and finite element technology was proposed.A re-fined finite element model was established according to the real situation of the rods arrangement of the high-formwork support structure,and the buckling analysis was carried out under a variety of working conditions,so as to obtain the vertical displacement of the key joints during the buckling damage of the high-formwork support structure,and to obtain the alarm threshold value of the intelligent real-time monitoring of the con-struction.The intelligent real-time monitoring algorithm based on computer vision was developed to monitor the displacement changes of the high-formwork support structure in the construction process in real time,and the algorithm will automatically alarm and guide the elimination of dangers and the protection of construction safety when unsafe displacement occurs.After the construction,according to the theoretical calculation re-sults and the actual measurement results,the finite element model was corrected parametrically,and the cor-rected finite element model was subjected to buckling damage analysis,and the sizes of the corrected buck-ling vertical displacement and the alarm threshold were compared to illustrate the reasonableness of the alarm threshold.The successful completion of a high-formwork support structure project for a multi-use building verified the effectiveness of the new construction monitoring method proposed in the paper.

high-formwork supportconstruction monitoringfinite element technologycomputer visionmodel modification

池昌政、庞博、任江涛、鲁寅、李先哲、黄铭枫

展开 >

浙江大学结构工程研究所,杭州 310058

浙江大华建设集团有限公司,杭州 310012

高大模板支撑结构 施工监控 有限元 计算机视觉 模型修正

国家自然科学基金浙江省建设科研项目

521785122022K137

2024

工业建筑
中冶建筑研究总院有限公司

工业建筑

CSTPCD
影响因子:0.72
ISSN:1000-8993
年,卷(期):2024.54(2)
  • 17