铁路GPR检测数据智能管理分析系统的研究与设计
|
摘要
为提高铁路基础设施检测探地雷达数据处理效率,提出了一种铁路探地雷达(GPR)检测数据智能管理分析系统方案。结合铁路基础设施检测工程需求与探地雷达数据特点,开展系统需求分析和设计。优化数据管理、工作流处理、交互解释和统计分析等步骤,克服目前软件的不足;利用Hadoop大数据技术进行多通道探地雷达数据的快速处理,实现海量探地雷达文件的并行计算;利用深度神经网络技术对探地雷达图像进行智能识别。对高铁隧道衬砌检测车采集的数据进行系统应用测试,测试效果显示,该系统可以提高探地雷达检测数据的管理和处理效率,推动铁路基础设施探地雷达检测向信息化、智能化和标准化发展。
In order to improve the data processing efficiency of ground penetrating radar(GPR) for railway infrastructure detection, this article proposed an intelligent management analysis system for railway GPR detection data.According to the requirement of the railway infrastructure inspection project and the characteristics of ground penetrating radar data, the system requirement analysis and design were carried out. By optimizing data management,workflow processing, interactive interpretation, statistical analysis and other steps, the article overcame the shortcomings of current software, used Hadoop big data technology to process multi-channel ground penetrating radar data quickly, and implemented parallel computation of massive ground penetrating radar files. The data collected by the high-speed railway tunnel lining inspection vehicle were tested systematically. The test results show that the system can improve the management and processing efficiency of ground penetrating radar detection data, and promote the development of ground penetrating radar detection in railway infrastructure towards informatization, intelligence and standardization.
In order to improve the data processing efficiency of ground penetrating radar(GPR) for railway infrastructure detection, this article proposed an intelligent management analysis system for railway GPR detection data.According to the requirement of the railway infrastructure inspection project and the characteristics of ground penetrating radar data, the system requirement analysis and design were carried out. By optimizing data management,workflow processing, interactive interpretation, statistical analysis and other steps, the article overcame the shortcomings of current software, used Hadoop big data technology to process multi-channel ground penetrating radar data quickly, and implemented parallel computation of massive ground penetrating radar files. The data collected by the high-speed railway tunnel lining inspection vehicle were tested systematically. The test results show that the system can improve the management and processing efficiency of ground penetrating radar detection data, and promote the development of ground penetrating radar detection in railway infrastructure towards informatization, intelligence and standardization.
引文
[1]王同军.智能铁路总体架构与发展展望[J].铁路计算机应用,2018,27(7):1-8.
[2]马小宁,史天运.智能铁路的内涵及关键技术研究[C]//中国智能交通协会.第八届中国智能交通年会论文集.北京:电子工业出版社,2013.
[3]白明洲,谢晋水,张爱军,等.高速铁路路基工程岩溶注浆效果无损检测评估方法研究[J].铁道学报,2012,34(7):89-95.
[4]许学良,马伟斌,蔡德钩,等.铁路隧道检测与监测技术的现状及发展趋势[J].铁道建筑,2018,58(1):14-19.
[5]刘宗辉,吴恒,周东,等.频谱反演法在探地雷达隧道衬砌检测中的应用研究[J].岩土工程学报,2015,37(4):711-717.
[6]杨艳青,高永涛,贺少辉,等.隧道衬砌背后积水地质雷达检测模型试验研究[J].中国铁道科学,2014,35(6):63-69.
[7]梁胤程,杜翠,刘杰,等.铁路路基状态检测中探地雷达数据并行处理[J].中国铁道科学,2017,38(2):11-18.
[8]杜翠,张千里,刘杰.基于支持向量机的铁路路基病害智能识别方法[C]//中国土木工程学会.2017年学术年会论文集,上海,2017.
[9]杜彦良:侯哲哲,赵维刚.一种基于稀疏表示的重载铁路路基病害快速识别方法[J].土木工程学报,2013,46(11):138-144.
[10]周辉林,姜玉玲,徐立红,等.基于SVM的高速公路路基病害自动检测算法[J].中国公路学报,2013,26(2):42-47.
[11]杜攀峰,廖立坚,杨新安.铁路路基病害的智能识别[J].铁道学报,2010,32(3):142-146.
[2]马小宁,史天运.智能铁路的内涵及关键技术研究[C]//中国智能交通协会.第八届中国智能交通年会论文集.北京:电子工业出版社,2013.
[3]白明洲,谢晋水,张爱军,等.高速铁路路基工程岩溶注浆效果无损检测评估方法研究[J].铁道学报,2012,34(7):89-95.
[4]许学良,马伟斌,蔡德钩,等.铁路隧道检测与监测技术的现状及发展趋势[J].铁道建筑,2018,58(1):14-19.
[5]刘宗辉,吴恒,周东,等.频谱反演法在探地雷达隧道衬砌检测中的应用研究[J].岩土工程学报,2015,37(4):711-717.
[6]杨艳青,高永涛,贺少辉,等.隧道衬砌背后积水地质雷达检测模型试验研究[J].中国铁道科学,2014,35(6):63-69.
[7]梁胤程,杜翠,刘杰,等.铁路路基状态检测中探地雷达数据并行处理[J].中国铁道科学,2017,38(2):11-18.
[8]杜翠,张千里,刘杰.基于支持向量机的铁路路基病害智能识别方法[C]//中国土木工程学会.2017年学术年会论文集,上海,2017.
[9]杜彦良:侯哲哲,赵维刚.一种基于稀疏表示的重载铁路路基病害快速识别方法[J].土木工程学报,2013,46(11):138-144.
[10]周辉林,姜玉玲,徐立红,等.基于SVM的高速公路路基病害自动检测算法[J].中国公路学报,2013,26(2):42-47.
[11]杜攀峰,廖立坚,杨新安.铁路路基病害的智能识别[J].铁道学报,2010,32(3):142-146.