铁路隧道衬砌病害智能检测技术研究

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (2633 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要 研究目的： 铁路隧道方面的缺陷病害已成为影响我国铁路运营安全的重要隐患,其中隧道衬砌质量问题尤为突出。针对隧道衬砌病害检测智能化程度低、人工判释标准不统一、地质雷达检测图像判释过程受设备型号、判释人员技术能力因素干扰等问题,本文以隧道衬砌病害地质雷达检测图像为样本库,训练图像识别算法,实现铁路隧道衬砌各类病害的智能识别,为铁路隧道质量检测提供自动化、智能化工具。
研究结论：（1）研究了隧道衬砌地质雷达缺陷图像特征,发现隧道衬砌质量病害的判定对检测人员的经验要求较高,具有一定主观性;（2）构建了不同种类病害的地质雷达影像样本库,通过训练智能识别模型,实现对不同病害的快速、自动识别;（3）针对某铁路项目中的隧道衬砌质量检测图像,进行了衬砌质量智能检测,识别结果显示,该技术可以有效避免检测过程中的人为因素干扰,提高了检测的工作效率和分析结论的准确性。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	张晨
	石峰
	孟超
	刘桂卫
	李炳秀
	田阳

关键词 ：铁路工程, 隧道衬砌, 深度学习, 图像识别

Abstract：Research purposes: Railway tunnel defects in terms of disease have become an important hidden danger affecting the safety of China's railway operations, in which the quality of tunnel lining is particularly prominent. In response to the low level of intelligence in tunnel lining disease detection, inconsistent manual interpretation standards, and interference from equipment models and technical capabilities of interpretation personnel during the interpretation process of geological radar detection images, research is being conducted on intelligent detection technology suitable for railway tunnel lining quality. Using geological radar detection images of tunnel lining diseases as a sample library, image recognition algorithms are trained to achieve intelligent recognition of various diseases of railway tunnel lining, providing automated and intelligent tools for railway tunnel quality detection.
Research conclusions: (1) The characteristics of geological radar defect images in tunnel lining were studied, and it was found that the determination of tunnel lining quality defects requires a high level of experience from the inspection personnel, and this work has a certain degree of subjectivity. (2) Build a geological radar image sample library for different types of diseases, train intelligent recognition models, and achieve fast and automatic recognition of different diseases. (3) Based on the quality inspection images of tunnel lining in a certain railway project, this technology is used for intelligent detection of lining quality. The recognition results show that this technology can effectively avoid human interference during the detection process, and improve the efficiency of detection and the accuracy of analysis conclusions.

Key words： railway engineering tunnel lining deep learning image recognition

收稿日期: 2024-07-26

PACS:

U212.3

基金资助:中国铁路设计集团有限公司课题项目（2024B0103000001）

作者简介: ** 张晨,1983年出生,男,高级工程师,注册土木工程师（岩土）。

引用本文:

张晨, 石峰, 孟超, 刘桂卫, 李炳秀, 田阳. 铁路隧道衬砌病害智能检测技术研究[J]. 铁道工程学报, 2024, 41(10): 57-62.
ZHANG Chen, SHI Feng, MENG Chao, LIU Guiwei, LI Bingxiu, TIAN Yang. Research of Intelligent Detection Technology for Railway Tunnel Engineering Quality. Journal of Railway Engineering Society, 2024, 41(10): 57-62.

链接本文:

https://tdgcxb.crec.cn/CN/ 或 https://tdgcxb.crec.cn/CN/Y2024/V41/I10/57

[1]	张湘桂, 邹明,刘宗辉,等.隧道衬砌后方隐伏洞穴地质雷达响应特征分析[J].地下空间与工程学报, 2021(5):1598-1605.Zhang Xianggui, Zou Ming, Liu Zonghui, etc. Technology of Detection and Identification of Hidden Caves behind Tunnel Lining by GPR[J].Chinese Journal of Underground Space and Engineering,2021(5):1598-1605.
[2]	毛宁, 吕擎峰, 郑立晟,等.基于神经网络的探地雷达隧道衬砌检测方法[J]. 地下空间与工程学报, 2023(S1):323-328.Mao Ning, Lv Qingfeng, Zheng Lisheng, etc.Detection Method of Tunnel Lining by Ground Penetrating Radar Based on Neural Network[J].Chinese Journal of Underground Space and Engineering,2023(S1): 323-328.
[3]	柴子贺. 基于数理统计方法的铁路隧道衬砌质量缺陷分析[J].铁道勘察, 2021(2): 56-62.Chai Zihe.Quality Defects Analysis of Railway Tunnel Lining Based on Mathematical Statistics[J].Railway Investigation and Surveying, 2021(2): 56-62.
[4]	贾志清,张顶立.隧道病害地质雷达检测结果自动处理系统[J].隧道建设, 2005(1):80-84.Jia Zhiqing, Zhang Dingli.Automatic Processing System for Geological Radar Detection Results of Tunnel Diseases[J].Tunnel Construction, 2005(1):80-84.
[5]	许磊,王长进. 隧道断面自动提取方法研究[J]. 铁道工程学报,2016(8):94-99.Xu Lei, Wang Changjin.Research on the Automated Extraction of Tunnel Cross Section[J]. Journal of Railway Engineering Society,2016(8):94-99.
[6]	许磊,王长进. 基于激光点云的隧道超欠挖检测方法研究[J]. 铁道工程学报,2016(12):77-81.Xu Lei, Wang Changjin.Research on the Detection Method of Tunnel Under–over Break Using Point Cloud[J]. Journal of Railway Engineering Society,2016(12):77-81.
[7]	王华,路耀邦,冯国峰,等. 隧道结构健康管理大数据平台研发及应用[J]. 隧道建设,2023(8):1425-1437.Wang Hua, Lu Yaobang, Feng Guofeng, etc.Development and Application of Big Data Platform for Tunnel Structure Health Management[J]. Tunnel Construction,2023(8):1425-1437.
[8]	成枢,桑泽磊,吴勇,等.基于移动三维激光扫描技术的隧道管片错台检测[J].测绘与空间地理信息,2022(4):5-9.Cheng Shu,Sang Zelei,Wu Yong,etc. Tunnel Segment Dislocation Detection Based on Mobile 3D Laser Scanning[J]. Geomatics & Spatial Information Technology,2022(4):5-9.