矿井提升机钢丝绳外观缺陷视觉识别技术研究

王国锋; 王守军; 陶荣颖; 李南; 罗自强

doi:10.13272/j.issn.1671-251x.2024010080

矿井提升机钢丝绳外观缺陷视觉识别技术研究

doi: 10.13272/j.issn.1671-251x.2024010080

淮河能源控股集团煤业公司，安徽淮南　232095

基金项目: 安徽省自然科学基金项目（1808085QE130）。

详细信息

作者简介:
王国锋（1967—），男，安徽淮南人，高级工程师，研究方向为矿井机电运输装备信息化、智能化，E-mail：huamuna@sina.com

中图分类号: TD532
计量
- 文章访问数: 193
- HTML全文浏览量: 36
- PDF下载量: 42
- 被引次数: 0
出版历程
- 收稿日期: 2024-01-23
- 修回日期: 2024-05-15
- 网络出版日期: 2024-06-13

Research on visual recognition technology for appearance defects of steel wire rope in mine hoist

Huaihe Energy Holding Group Coal Company, Huainan 232095, China

摘要

摘要: 针对多根钢丝绳检测部署困难、钢丝绳图像采集质量较低、视觉检测法适应性差、准确性不高等问题，提出了一种基于计算机视觉和深度学习的矿井提升机钢丝绳外观缺陷视觉识别方法。首先构建矿井提升机钢丝绳在线监测系统；其次由地面移动巡检平台和井下本安高速相机采集钢丝绳图像，建立钢丝绳图像数据集；然后考虑井下粉尘影响、相机镜头易受污染、光照不均、钢丝绳高光反射等问题，采用基于Retinex算法的图像去噪方法和基于同态滤波的图像去噪方法对钢丝绳图像进行去噪处理，处理结果表明，基于色彩增益加权的多尺度Retinex（AutoMSRCR）算法为较优方案；最后缺陷检测过程以卷积神经网络为基础，构建基于YOLOv5s的缺陷检测模型，为降低人为因素影响、调参工作量，在YOLOv5s中加入Focus结构对其进行优化，并将改进的YOLOv5s模型作为钢丝绳缺陷检测的预训练模型，以进一步降低模型内存占用率，提高模型加载和检测速度。实验结果表明，所提方法对钢丝绳2处断丝的检测误差分别为1.61%，1.35%，对钢丝绳4处磨损的检测误差分别为2.43%，3.44%，2.11%，3.39%。针对淮河能源控股集团顾北煤矿主井提升机原有钢丝绳安全监测系统的检测精度无法满足现场需求的问题，采用所提方法对原系统进行改进，现场应用效果表明，钢丝绳断丝检测准确率由80%提升至96%，损伤定位误差由500 mm降低至300 mm范围内，损伤定位准确率由75%提升至98%，损伤实时检出率由76%提升至90%，尾绳畸变检出率由70%提升至85%。
- 矿井提升机 /
- 钢丝绳外观缺陷 /
- 断丝 /
- 表面磨损 /
- 视觉识别 /
- 图像去噪处理 /
- Retinex算法 /
- 改进YOLOv5s
Abstract: A visual recognition method for appearance defects of mine hoist steel wire ropes based on computer vision and deep learning is proposed to address the problems of difficult deployment for detecting multiple steel wire ropes, low image acquisition quality of steel wire ropes, poor adaptability and accuracy of visual detection methods. Firstly, an online monitoring system for the steel wire rope of the mine hoist is constructed. Secondly, the steel wire rope images are collected by the ground mobile inspection platform and the underground intrinsic safety high-speed camera, and a steel wire rope image dataset is established. Considering the effects of underground dust, susceptibility of camera lenses to contamination, uneven lighting, and high light reflection of steel wire ropes, image denoising methods based on Retinex algorithm and homomorphic filtering are used to denoise the steel wire rope images. The processing results show that the automated multi-scale Retinex with color restoration (AutoMSRCR) algorithm based on color gain weighting is the optimal solution. The defect detection process is based on convolutional neural networks, and a defect detection model based on YOLOv5s is constructed. In order to reduce the influence of human factors and the workload of parameter tuning, a Focus structure is added to YOLOv5s for optimization. The improved YOLOv5s model is used as a pre training model for steel wire rope defect detection to further reduce the memory usage of the model and improve the loading and detection speed of the model. The experimental results show that the proposed method has detection errors of 1.61% and 1.35% for wire breakage at 2 positions of the steel wire rope, and detection errors of 2.43%, 3.44%, 2.11%, and 3.39% for wear at 4 positions of the steel wire rope. In response to the problem that the detection precision of the original steel wire rope safety monitoring system for the main shaft hoist of Gubei Coal Mine, Huaihe Energy Holding Group, cannot meet the on-site requirements, the proposed method is adopted to improve the original system. The on-site application results show that the accuracy of wire rope breakage detection is increased from 80% to 96%, the damage positioning error is reduced from 500 mm to within 300 mm. The damage positioning accuracy is increased from 75% to 98%, the real-time detection rate of damage is increased from 76% to 90%, and the tail rope distortion detection rate is increased from 70% to 85%.
- mine hoist /
- appearance defects of steel wire ropes /
- wire breakage /
- surface wear /
- visual recognition /
- image denoising processing /
- Retinex algorithm /
- improved YOLOv5s

HTML全文

图 1 矿井提升机钢丝绳在线监测系统组成

Figure 1. Online monitoring system composition of mine hoist steel wire rope

下载: 全尺寸图片幻灯片

图 2 移动巡检平台

Figure 2. Mobile inspection platform

下载: 全尺寸图片幻灯片

图 3 Retinex算法处理结果

Figure 3. Processing results of Retinex algorithms

下载: 全尺寸图片幻灯片

图 4 同态滤波算法处理结果

Figure 4. Processing results of homomorphic filtering algorithms

下载: 全尺寸图片幻灯片

图 5 钢丝绳缺陷检测实验流程

Figure 5. Experimental flow of wire ropes defect detection

下载: 全尺寸图片幻灯片

图 6 改进YOLOv5s网络结构

Figure 6. Improved YOLOv5s network structure

下载: 全尺寸图片幻灯片

图 7 钢丝绳断丝检测结果

Figure 7. Breakage detection results of steel wire rope

下载: 全尺寸图片幻灯片

图 8 钢丝绳磨损检测结果

Figure 8. Surface wear detection results of steel wire rope

下载: 全尺寸图片幻灯片

表 1 不同方法的峰值信噪比

Table 1. Peak signal-to-noise ratio of different methods

去噪方法	AutoMSRCR	MSR	高斯滤波	指数滤波
PSNR	13.25	12.63	11.79	11.18

下载: 导出CSV

表 2 断丝检测定位结果

Table 2. Breakage detection and positioning results of steel wire rope

方法	断丝实际位置/m	检测位置/m	误差/%
文献[9]方法	10.56	10.18	3.60
文献[9]方法	11.12	10.69	3.87
文献[11]方法	10.56	10.29	2.56
文献[11]方法	11.12	11.46	3.06
本文方法	10.56	10.39	1.61
本文方法	11.12	10.97	1.35

下载: 导出CSV

表 3 磨损检测定位结果

Table 3. Surface wear detection and positioning results of steel wire rope

方法	磨损实际位置/m	检测位置/m	误差/%
文献[9]方法	5.75	5.41	5.91
	6.11	未识别	−
	6.62	6.31	4.68
	6.79	未识别	−
文献[11]方法	5.75	5.50	4.35
	6.11	未识别	−
	6.62	6.37	3.78
	6.79	未识别	−
本文方法	5.75	5.89	2.43
	6.11	6.32	3.44
	6.62	6.48	2.11
	6.79	7.02	3.39

下载: 导出CSV

表 4 效果对比

Table 4. Effect comparison

比较项目	改进前	改进后
断丝检测准确率/%	≥80	≥96
损伤定位误差/mm	≤500	≤300
损伤定位准确率/%	≥75	≥98
损伤实时检出率/%	≥76	≥90
尾绳畸变检出率/%	≥70	≥85

下载: 导出CSV

参考文献(22)

[1]	朱真才,李翔,沈刚,等. 双绳缠绕式煤矿深井提升系统钢丝绳张力主动控制方法[J]. 煤炭学报,2020,45(1):464-473. ZHU Zhencai,LI Xiang,SHEN Gang,et al. Wire rope tension active control of double-rope winding deep well hoisting systems[J]. Journal of China Coal Society,2020,45(1):464-473.
[2]	李腾宇,寇子明,吴娟,等. 超千米深井提升机可视化监测系统应用[J]. 煤炭学报,2020,45(增刊2):1069-1078. LI Tengyu,KOU Ziming,WU Juan,et al. Monitoring system of the hoist in the over kilometer deep shaft[J]. Journal of China Coal Society,2020,45(S2):1069-1078.
[3]	王红尧,田劼,张艳林,等. 矿用钢丝绳在线监测教学实验装置关键技术[J]. 煤矿安全,2021,52(6):177-182. WANG Hongyao,TIAN Jie,ZHANG Yanlin,et al. Key technologies of teaching experimental device for on line inspection of mining wire rope[J]. Safety in Coal Mines,2021,52(6):177-182.
[4]	ZHANG Guoyang,TANG Zhaohui,FAN Ying,et al. Steel wire rope surface defect detection based on segmentation template and spatiotemporal gray sample set[J]. Sensors,2021,21(16). DOI: 10.3390/s21165401.
[5]	ZHOU Ping,ZHOU Gongbo,HE Zhenzhi,et al. A novel texture-based damage detection method for wire ropes[J]. Measurement,2019,148(12). DOI: 10.1016/j.measurement.2019.106954.
[6]	刘钰,康爱国,李良辉,等. 基于TMR传感器的矿用钢丝绳断丝缺陷检测系统[J]. 煤矿安全,2019,50(5):122-125. LIU Yu,KANG Aiguo,LI Lianghui,et al. Broken wire defect detection system in mine wire rope based on TMR sensor[J]. Safety in Coal Mines,2019,50(5):122-125.
[7]	田劼,田壮,郭红飞,等. 矿用钢丝绳损伤检测磁通回路优化设计[J]. 工矿自动化,2022,48(3):118-122. TIAN Jie,TIAN Zhuang,GUO Hongfei,et al. Optimization design of magnetic flux circuit for mine wire rope damage detection[J]. Journal of Mine Automation,2022,48(3):118-122.
[8]	叶辉,乔铁柱. 矿用钢丝绳在线检测系统[J]. 煤矿安全,2018,49(8):131-134. YE Hui,QIAO Tiezhu. Research on on-line detection system of mine wire rope[J]. Safety in Coal Mines,2018,49(8):131-134.
[9]	李金华,夏黎明. 图像识别技术在矿用钢丝绳检测中的应用[J]. 山西焦煤科技,2022,46(4):16-18,21. doi: 10.3969/j.issn.1672-0652.2022.04.005 LI Jinhua,XIA Liming. Application of image recognition technology in mining wire rope detection[J]. Shanxi Coking Coal Science & Technology,2022,46(4):16-18,21. doi: 10.3969/j.issn.1672-0652.2022.04.005
[10]	姜泓宇,董增寿,贺之靖. 基于机器视觉的钢丝绳表面缺陷检测[J]. 太原科技大学学报,2023,44(5):434-439,446. JIANG Hongyu,DONG Zengshou,HE Zhijing. Surface defect detection of wire rope based on feature fusion and IWOA-SVM[J]. Journal of Taiyuan University of Science and Technology,2023,44(5):434-439,446.
[11]	刘晓磊,吴国群,阚哲. 基于深度学习的煤矿钢丝绳缺损检测方法研究[J]. 煤炭工程,2023,55(11):148-153. LIU Xiaolei,WU Guoqun,KAN Zhe. Research on defect detection method of coal mine wire rope based on deep learning[J]. Coal Engineering,2023,55(11):148-153.
[12]	吴东,张宝金,刘伟新,等. 强噪声背景下钢丝绳损伤信号降噪方法[J]. 工矿自动化,2022,48(1):58-63. WU Dong,ZHANG Baojin,LIU Weixin,et al. Noise reduction method for wire rope damage signal under strong noise background[J]. Industry and Mine Automation,2022,48(1):58-63.
[13]	阮顺领,刘丹洋,白宝军,等. 基于自适应MSRCP算法的煤矿井下图像增强方法[J]. 矿业研究与开发,2021,41(11):186-192. RUAN Shunling,LIU Danyang,BAI Baojun,et al. Image enhancement method for underground coal mine based on the adaptive MSRCP algorithm[J]. Mining Research and Development,2021,41(11):186-192.
[14]	朱海平. 矿井提升钢丝绳表面损伤在线视觉检测系统研究[D]. 徐州:中国矿业大学,2023. ZHU Haiping. Research on online visual detection system for surface damage of mine hoisting wire rope[D]. Xuzhou:China University of Mining and Technology,2023.
[15]	郭永坤,朱彦陈,刘莉萍,等. 空频域图像增强方法研究综述[J]. 计算机工程与应用,2022,58(11):23-32. GUO Yongkun,ZHU Yanchen,LIU Liping,et al. Research review of space-frequency domain image enhancement methods[J]. Computer Engineering and Applications,2022,58(11):23-32.
[16]	BHATT P M,MALHAN R K,RAJENDRAN P,et al. Image-based surface defect detection using deep learning[J]. Journal of Computing and Information Science in Engineering,2021,21(4):1-23.
[17]	HUANG Xinyuan,LIU Zhiliang,ZHANG Xiuyu,et al. Surface damage detection for steel wire ropes using deep learning and computer vision techniques[J]. Measurement,2020,161(12). DOI: 10.1016/j.measurement.2020.107843.
[18]	李鑫. 基于机器视觉的钢丝绳直径在线检测方法研究[D] . 西安:西安石油大学,2023. LI Xin. Research on online inspection method of wire rope diameter based on machine vision[D]. Xi'an:Xi'an Shiyou University,2023.
[19]	LIU Shiwei,SUN Yanhua,KANG Yihua. A novel e-exponential stochastic resonance model and weak signal detection method for steel wire rope[J]. IEEE Transactions on Industrial Electronics,2022,69(7):7428-7440. doi: 10.1109/TIE.2021.3095802
[20]	赵文,薛涛,凡成华,等. 矿井提升机钢丝绳损伤在线检测方法研究[J]. 矿山机械,2022,50(6):22-26. doi: 10.3969/j.issn.1001-3954.2022.06.006 ZHAO Wen,XUE Tao,FAN Chenghua,et al. Research on online detection method for damage of wire rope of mine hoist[J]. Mining & Processing Equipment,2022,50(6):22-26. doi: 10.3969/j.issn.1001-3954.2022.06.006
[21]	LIU Shiwei,CHEN Muchao. Wire rope defect recognition method based on MFL signal analysis and 1D-CNNs[J]. Sensors,2023,23(7). DOI: 10.3390/s23073366.
[22]	CHANG X D,PENG Y X,ZHU Z C,et al. Tribological behavior and mechanical properties of transmission wire rope bending over sheaves under different sliding conditions[J]. Wear,2023(514/515). DOI: 10.1016/j.wear.2022.204582.