摘要
钢丝绳作为重要的提升、牵引与承载构件,在服役过程中其表面易产生断丝缺陷,严重影响绳系统的运行安全。针对传统视觉检测方法易受钢丝绳表面油污影响,引入脉冲涡流热成像技术,结合钢丝绳金属构件热传导特性,提出一种基于脉冲涡流热成像的钢丝绳表面缺陷红外热视觉检测方法。通过脉冲涡流热成像仿真探明钢丝绳感应加热机理与断丝缺陷的温度分布规律;构建钢丝绳断丝缺陷热视觉检测系统,设计基于脉冲涡流热成像的钢丝绳断丝缺陷表征与信号处理方法;建立钢丝绳表面缺陷红外热像数据集,并设计基于深度目标检测网络的YOLOv5_WR-seg模型对钢丝绳断丝缺陷进行检测识别;研究在役钢丝绳表面油脂油污对红外热视觉检测方法的影响,实现油脂油污覆盖下断丝缺陷的检测,克服油脂油污对断丝缺陷检测的视觉干扰。试验结果表明,所提出的方法可以实现有/无油污覆盖下钢丝绳表面断丝缺陷的准确检测。
As essential lifting,traction,and bearing components,wire ropes(WRs)are prone to surface breakage during service,which seriously impacts the operational safety of the rope system.Given that the traditional visual detection method is susceptible to the oil on the surface of the WR,pulsed eddy current thermography is introduced.Combined with the heat conduction properties of the metal components of the WR,an infrared thermal vision-based detection method is proposed for identifying surface defects.Firstly,the mechanism of induction heating and the temperature distribution of broken wire defects are investigated through pulsed eddy current thermal imaging simulation.Secondly,the thermal vision detection system for wire rope breakage defects is constructed,and the characterization and signal processing method for wire rope breakage defects based on pulsed eddy current thermal imaging is designed.Then,the infrared thermal image dataset of surface defect of the WR is established,and the YOLOv5_WR-seg model based on the deep target detection network is designed to detect and identify the broken wire defect of the WR.Finally,the influence of oil and grease on the surface of the wire rope in service on the infrared thermal vision detection method is studied to achieve the detection of broken wire defects covered by oil and grease and overcome the visual interference of oil and grease on the detection of broken wire defects.The experimental results indicate that the proposed method can accurately detect wire rope surface breakage defects with or without oil contamination.
作者
周坪
周公博
王晗宇
王攀
闫晓东
李远博
ZHOU Ping;ZHOU Gongbo;WANG Hanyu;WANG Pan;YAN Xiaodong;LI Yuanbo(School of Electrical and Mechanical Engineering,China University of Mining and Technology,Xuzhou 221116;State Key Laboratory of Intelligent Mining Equipment Technology,Xuzhou 221116;State Key Laboratory of High-end Mining Equipment(Cultivation Site),Xuzhou 221116)
出处
《机械工程学报》
2026年第4期86-97,共12页
Journal of Mechanical Engineering
基金
中央高校青年教师科研能力创新能力支持(ZYGXQNJSKYCXNLZCXM-E2P)
国家自然科学基金(62301564)
江苏省自然科学基金(BK20231068)
江苏高校优势学科建设工程(PAPD)资助项目。
关键词
脉冲涡流热成像
红外视觉
钢丝绳
断丝缺陷
目标检测网络
pulse eddy current thermal imaging
infrared visual
wire ropes
broken wire defect
object detection network
