文章摘要
王晓玲,王栋,任炳昱,陈文夫,谭尧升,关涛.高拱坝混凝土振捣机器人系统研发及应用[J].水利学报,2022,53(6):631-643,654
高拱坝混凝土振捣机器人系统研发及应用
Development and application of concrete vibrating robot system for high arch dam
投稿时间:2022-01-06  
DOI:10.13243/j.cnki.slxb.20220010
中文关键词: 高拱坝  混凝土振捣  振捣机器人系统  机器视觉  改进ResNet  表面图像分类
英文关键词: high arch dam  concrete vibration  vibrating robot system  machine vision  improvedResNet  surface image classification
基金项目:国家自然科学基金项目(51879186,51909187);中国长江三峡集团公司科研项目(BHT/0835);天津市研究生科研创新项目(2020YJSB095)
作者单位E-mail
王晓玲 天津大学 水利工程仿真与安全国家重点实验室, 天津 300072  
王栋 天津大学 水利工程仿真与安全国家重点实验室, 天津 300072  
任炳昱 天津大学 水利工程仿真与安全国家重点实验室, 天津 300072 renby@tju.edu.cn 
陈文夫 中国三峡建工(集团)有限公司, 四川 成都 610041
中国长江三峡集团有限公司, 北京 100038 
 
谭尧升 中国三峡建工(集团)有限公司, 四川 成都 610041
中国长江三峡集团有限公司, 北京 100038 
 
关涛 天津大学 水利工程仿真与安全国家重点实验室, 天津 300072  
摘要点击次数: 164
全文下载次数: 175
中文摘要:
      常态混凝土浇筑人工振捣施工存在工作强度高、作业主观性强等不足,现有振捣监控方法主要进行振捣位置、振捣时间等参数监控,且均依靠人工经验评估振捣是否密实,难以实现混凝土振捣质量的精确评估。针对上述问题,研发了融合表面图像机器视觉识别的混凝土振捣机器人系统,该系统包括振捣机器人本体、云端监控平台以及通讯子系统。基于工业机器人、自动化控制、机器视觉等技术研制振捣机器人本体,实现高拱坝仓面混凝土无人振捣作业;同时,建立基于改进残差网络(Residual Network,ResNet)的混凝土振捣表面图像的分类模型,实现振捣密实性的科学分析。此外,基于B/S架构研发云端监控平台,并通过无线通讯子系统实现与振捣机器人本体之间信息的传输与交互,实现振捣作业信息的可视化与远程控制。现场试验验证了振捣机器人系统的可靠性和可行性。研究成果为高拱坝混凝土振捣实现无人化作业、精细化质量控制提供了新的思路和技术手段,也可用于其它大体积混凝土施工中,具有十分重要的应用和推广价值。
英文摘要:
      The traditional manual vibration is faced with the shortcomings of high labor intensity and strong subjectivity during normal concrete placement,and the existing vibration monitoring methods mainly focus on the vibration parameters such as vibration position and vibration time.Those methods mainly rely on manual experience to evaluate whether the vibration is consolidated,which is difficult to realize the accurate evaluation of concrete vibration quality.To solve those problems,a concrete vibrating robot system integrating surface image recognition using machine vision method is developed,which includes vibrating robot,cloud monitoring platform and communication subsystem.The vibrating robot is designed to realize the unmanned vibration of concrete on the storehouse surface of high arch dam using advanced technology including the industrial robot,automatic control and machine vision.Meanwhile,the classification model of concrete surface image based on improved residual network (ResNet) is established to realize the scientific analysis of vibration quality.Moreover,the cloud monitoring platform is developed based on the B/S architecture,and the information transmission and interaction with the vibrating robot are realized through the wireless communication subsystem,which realizes the vibration information visualization and remote control.The field tests have been conducted to verify the reliability and feasibility of the proposed vibrating robot system.The research results provide a new idea and technical way for realizing unmanned vibration and refined quality control of concrete placement of high arch dam and could be used for other mass concrete construction,which have significant application and popularization value.
查看全文   查看/发表评论  下载PDF阅读器
关闭