高速铁路大跨桥梁端钢轨伸缩调节器监测技术
Monitoring Technology for Rail Expansion Joints of Large-span Bridges in HSR
王伟华 1高梓航 1王一 2蔡小培2
作者信息
- 1. 中国铁路设计集团有限公司线路站场枢纽设计研究院,天津 300308
- 2. 北京交通大学土木建筑工程学院,北京 100044
- 折叠
摘要
钢轨伸缩调节器被广泛应用于高速铁路大跨桥上无缝线路中,但因相对位移量大、零部件多等问题成为高铁薄弱环节之一,实时监测其服役状态对于高铁运营安全至关重要.为保证监测数据的实时性与准确性,结合计算机视觉技术与实际工程现状,建立调节器服役状态在线监测系统.具体包括:完成监测系统设计与布点方案优化研究,开发现代化的信息管理系统;针对现场传感器的实时采集数据设计一整套数据处理流程,结合随机森林、长短期神经网络等算法实现调节器服役状态的超前预测,结合YOLO系列目标检测算法实现钢轨位移量的非接触式测量;基于现场需求设计综合试验平台,对重点关注的钢轨位移、应力等重要指标进行模拟试验.试验结果表明,研究的调节器监测数据处理方法可以适应调节器区域监测需求,具有重要的工程意义.
Abstract
Rail expansion joints are widely used in the CWR tracks of large-span bridges in HSRs.However,due to issues such as significant relative displacement and the presence of multiple components,they have become one of the vulnerable points in HSR operations.Real-time monitoring of their operational status is crucial for the safety of HSR operations.To ensure the real-time and accuracy of the monitoring data,this paper establishes an online monitoring system for the operational status of rail expansion joints in combination with computer vision technology and the current engineering conditions.It specifically includes the design of monitoring system and optimization study on layout scheme and development of a modern information management system;a complete set of data processing flow is designed according to the real-time collected data from field sensors.The advanced prediction of the service state of rail expansion joint is realized by combining random forest,LSTM and other algorithms,and the non-contact measurement of rail displacement is realized by combining YOLO series target detection algorithms;a comprehensive test platform is designed based on field requirements to carry out simulation tests on important indicators such as rail displacement and stress.The test results indicate that the developed data processing method for monitoring rail expansion joints can meet the area monitoring needs,highlighting its significant engineering importance.
关键词
高速铁路/钢轨伸缩调节器/数据处理/目标检测/试验平台Key words
high speed railway/rail expansion joint/data processing/target detection/test platform引用本文复制引用
出版年
2024
NETL
NSTL
万方数据