混凝土坝水平缝面智能冲毛系统研发及应用  

Development and application of horizontal seam surface intelligent flushing system for concrete dam

作　　者：崔博[1] 程益骄 关涛[1] 王晓玲[1] 谭尧升 陈文夫 CUI Bo;CHENG Yijiao;GUAN Tao;WANG Xiaoling;TAN Yaosheng;CHEN Wenfu(State Key Laboratory of Hydraulic Engineering Simulation and Safety,Tianjin University,Tianjin 300350,China;China Three Gorges Group Corporation,Beijing 100038,China;China Three Gorges Construction Engineering Corporation,Chengdu 610041,China)

机构地区：[1]天津大学水利工程仿真与安全国家重点实验室,天津300350 [2]中国长江三峡集团有限公司,北京100038 [3]中国三峡建工(集团)有限公司,四川成都610041

出　　处：《水利学报》2023年第6期644-653,665,共11页Journal of Hydraulic Engineering

基　　金：中国长江三峡集团公司科研项目(BHT 0836);中国大坝工程协会青年人才资助项目(CHINCOLD-QNRC006);国家自然科学基金项目(51909187)。

摘　　要：新老混凝土水平施工缝面处理是确保坝体混凝土整体性的关键工序。目前混凝土坝施工主要根据经验采用高压水人工冲毛方式对冲毛位置、冲毛时间等进行控制,存在工作强度高、作业主观性强等不足,难以实现冲毛质量的精准控制。针对上述问题,本文研究了混凝土水平缝面智能冲毛系统。基于工业机器人、自动化控制等技术研制智能冲毛装备,实现混凝土坝仓面无人冲毛作业;耦合多源传感器信息,建立智能冲毛装备运动学模型以及路径规划方法,实现智能冲毛装备自主避障;结合机器视觉理论,建立基于卷积神经网络的混凝土冲毛表面图像智能分类模型,实现冲毛质量的客观分析;同时,开发云端监控平台和无线通讯子系统,实现与智能冲毛装备之间的信息传输和多平台互联互通,并进行智能冲毛装备的有效管理和远程控制。此外,通过现场试验验证了智能冲毛系统的可行性和可靠性,研究成果为混凝土坝冲毛施工实现无人化作业、科学化质量管控提供了新的思路和技术手段,具有十分重要的现实意义和应用价值。The treatment of the new and old concrete horizontal construction seam surface is a key process to ensure the integrity of the dam body concrete.At present,the construction of concrete dams mainly uses high-pressure water artificial punching method to control the punching position and punching time based on experience.There are shortcomings such as high work intensity and strong operation subjectivity,and it is difficult to achieve precise control of punching quality.Aiming at the above problems,this paper studies the intelligent punching system of concrete horizontal joint surface.Based on industrial robots,automatic control and other technologies,intelligent punching equipment is developed to realize unmanned punching operation on the surface of concrete dams;multi-source sensor information is coupled to establish a kinematic model and path planning method of intelligent punching equipment to realize the autonomy of intelligent punching equipment obstacle avoidance;combined with machine vision theory,an intelligent classification model of concrete punching surface images based on convolutional neural network is established to realize objective analysis of punching quality;at the same time,cloud monitoring platform and wireless communication subsystem are developed to realize integration with intelligent punching equipment information transmission and multi-platform interconnection and interoperability between them,and effective management and remote control of intelligent punching equipment.In addition,the feasibility and reliability of the intelligent burring system are verified through field tests.The research results provide new ideas and technical means for the realization of unmanned operation and scientific quality control in the burring construction of concrete dams,which have realistic meaning and application value.

关 键 词：混凝土坝 水平缝面 智能冲毛 无人化作业 路径规划 机器视觉 

分 类 号：TV53[水利工程—水利水电工程]