At present, resarch on virtual construction is limited to the construction preparation phase. The level of automation and imformation of the construction and control implementation process is not high, and there is few technical staff who can better grasp of the control theory and methods in most of the construction enterprises, so these construction enterprises can not independently complete the construction of large complex structures. .In this project of application, the desktop environment of virtual construction control is going to be created based on the concept of Augmented Reality. The technology of structrual measuarement based on digital image in the construction process will be studied, the real environment will be integrated into the virtual construction environment. The test data combined with the results of structural analysis of the virtual structure model are to be used to identify the model parameters, and the model will be refined. And the data will then be used combined with the default design state of the structure, to identify the structural control parameters. The strcture control operation will be implemented according to the control parameters, the sensor signal of the control movement in the real environment will be fed back to the virtual environment. .Innovation: based on the combination of the real and virtual situation, the human-computer interaction and the reference of calculate the test data, the deductation and practical operation of the entire process and each stage in the process of the construction control of the engineering structure; Large scal parameters identification with the spacially intensive measurement deformation data in the visible surface of the structure; Impact analysis of various errors of the various stages on the real and virtual environment. .The key issues to be addressed as follows: the distortion of the image detection data for the large scale engineering structure and the coordination of the data; large scale parameters identification; data exchange mutual fusion in the real and virtual environment; the errors impact of the analysis and calculation, measurement and identification in various stages on the real and virtual environment.
目前虚拟施工研究局限于施工准备阶段,施工控制实施过程的自动化信息化程度不高,而且多数施工企业缺少对施工控制理论和方法掌握较好的技术人员,不能独立完成大型复杂结构的施工控制。本项目按增强现实的理念,建立桌面虚拟施工控制环境,研究基于数字图像法的施工检测技术,将真实环境的检测数据溶入虚拟施工环境,结合结构虚拟模型的分析数据识别模型参数,并进行模型修正,结合预设结构状态识别控制参数,在真实环境对结构施工进行控制,传感器将控制动作信号反馈到虚拟环境。创新点为:由虚实结合人机交互、计算测试数据参照,对工程结构施工控制的全程和各阶段进行推演及实操;基于结构可见表面高密度分布点变形数据的规模化参数识别;施工过程的各阶段各种误差对虚实环境的影响分析。需解决的关键问题为:大体量结构的图像检测数据失真和相互协调;规模化参数识别;虚实环境的数据交换及相互溶合;各阶段分析计算、检测、识别误差对虚实环境的影响。
本项目按增强现实的理念,针对连续箱梁桥顶推施工、钢桁架桥悬臂拼装施工和斜拉桥悬臂拼装施工,建立桌面虚拟施工环境,将施工各阶段的结构分析计算数据输入虚拟施工环境预演施工过程,按预演剧本开始真实施工,由数字图像法结合传统的检测方法实施施工监测,将真实环境的检测数据溶入虚拟施工环境,比较结构在施工过程中变形实测与预演结果,修正施工剧本,调整控制下阶段的施工。在真实环境按虚拟剧情进行结构施工并予以监测和控制,在虚拟环境里按分析和实测结果单一或迭加表现结构施工预演和实测状态。完成了连续梁桥顶推施工、钢桁架悬拼施工虚实结合控制的仿真和模型试验研究,部分完成了斜拉桥模型悬拼施工试验。虚拟的意义在于现实的应用,本项目成功地将虚拟施工与真实施工在全过程各阶段实现交互融合。. 研发了边缘检测的正交多项式拟合法及二维多项式拟合法,模型试验验证,图像检测结构(长度8米以内)变形精度与百分表相当,精确检测保证了精确施工,连续梁顶推各阶段实测与预演结果最大绝对误差小于1毫米相对误差小于8%。基于数字图像法的结构变形检测,一维长度在10米以内的结构监测精度约0.1毫米,满足工程要求,大型桥梁长度往往大于100米,可采用多通道联机监测设备。图像采集桥梁变形常须采用遥测,须解决长焦远摄的抖动问题。在精确检测的基础上,结构施工虚实控制不难实现。. 研发了梁、弦振动检测的视频法,识别了连续梁及斜拉桥模型在施工各阶段的频率阻尼及振型。由索前几阶识别频率计算的索力与力传感器监测结果吻合较好,识别的弦振型与理论解吻合较好。采用加速度监测索振动时,识别振型较困难,因为传感器多点安装困难,同时传感器惯性质量对弦频率有影响,这两点正好是视频非接触检测的优点。加速度幅值是位移幅值乘以圆频率的平方,弦的n阶频率是其1阶频率的n倍,n阶频率对应的加速度幅值是位移幅值的n的平方倍,在识别高频振动的灵敏度方面加速度传感器明显优于视频检测。
{{i.achievement_title}}
数据更新时间:2023-05-31
演化经济地理学视角下的产业结构演替与分叉研究评述
环境类邻避设施对北京市住宅价格影响研究--以大型垃圾处理设施为例
端壁抽吸控制下攻角对压气机叶栅叶尖 泄漏流动的影响
基于ESO的DGVSCMG双框架伺服系统不匹配 扰动抑制
基于全模式全聚焦方法的裂纹超声成像定量检测
基于多孔硅微阵列芯片数字图像的高灵敏度生物免标记检测技术研究
基于动静结合的内存泄漏自动检测技术研究
结构工程施工拟实技术
基于综合特征的数字图像盲取证技术研究