换热器清洗机器人管口识别及定位系统研究  

Research on Nozzle Recognition and Positioning System ofHeat Exchanger Cleaning Robot

作　　者：李文军[1] 李彩玲[1] LI Wenjun;LI Cailing(Linfen Vocational and Technology College,Linfen 041000,China)

机构地区：[1]临汾职业技术学院,山西临汾041000

出　　处：《北京石油化工学院学报》2024年第4期54-59,共6页Journal of Beijing Institute of Petrochemical Technology

基　　金：山西省十三五规划课题(HLW-20208)。

摘　　要：随着国家对清洗技术的安全性和效率要求的日益提高,传统人工清洗方式因其低效、高成本及安全隐患而逐渐被淘汰。提出了一种基于视觉识别的换热器清洗机器人识别及定位系统,旨在实现清洗作业的智能化、自动化与高效化。通过融合传统图像处理技术和深度学习算法,采用MATLAB进行机器学习,并创新性地提出HDS-UNet模型,优化图像识别性能,提升管口图像识别的灵敏度和准确度。同时,结合相机标定技术实现从图像像素坐标到世界坐标的精确转换,为清洗机器人提供了精确的路径规划和定位依据。实验结果表明,该系统能够高效、准确地识别并定位换热器管口,为全自动清洗作业提供有力支持,显著提高了清洗效率和安全性,具有广阔的应用前景。With the rapid development of the heat exchanger industry,the country's requirements for the safety and efficiency of cleaning technology have been increasingly improved.Traditional manual cleaning methods are gradually being eliminated due to their inefficiency,high cost,and potential safety hazards.This paper proposes a heat exchanger cleaning robot recognition and positioning system based on visual recognition,aiming to achieve intelligent,automated,and efficient cleaning operations.The system integrates traditional image processing technology and deep learning algorithms,uses MATLAB for machine learning,and innovatively proposes the HDS-UNet model to optimize image recognition performance and improve the sensitivity and accuracy of nozzle image recognition.At the same time,combined with camera calibration technology,it realizes the accurate conversion from image pixel coordinates to world coordinates,providing accurate path planning and positioning basis for the cleaning robot.Experimental results show that the system can efficiently and accurately identify and locate the nozzles of the heat exchanger,providing strong support for fully automatic cleaning operations,significantly improving cleaning efficiency and safety,and having broad application prospects.

关 键 词：换热器 图像识别 视觉识别 智能化清洗 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]