基于PSO-SVM的无砟轨道路基沉降病害识别

任娟娟; 张亦弛; 刘伟; 章恺尧; 杜威

doi:10.19713/j.cnki.43-1423/u.T20222451

您当前的位置：

首页 >

文章列表页 >

基于PSO-SVM的无砟轨道路基沉降病害识别

工程诊所•工匠之家 | 更新时间：2023-12-05

- 基于PSO-SVM的无砟轨道路基沉降病害识别
- Identification of subgrade settlement diseases in ballastless track based on PSO-SVM
- 铁道科学与工程学报 2023年20卷第11期页码：4400-4411
- 作者机构：
  
  1.西南交通大学高速铁路线路工程教育部重点实验室，四川成都 610031
  2.西南交通大学土木工程学院，四川成都 610031
  3.长沙理工大学交通运输工程学院，湖南长沙 410114
  4.同济大学交通运输工程学院，上海 201804
- 作者简介：
  
  任娟娟(1983—)，女，山西霍州人，教授，博士，从事高速铁路无砟轨道结构设计理论与损伤机理研究；E-mail：jj.ren@swjtu.cn
- 基金信息：
  
  国家重点研发计划资助项目(2021YFF0502100);国家自然科学基金资助项目(52022085;52278461);四川省青年科技创新研究团队项目(2022JDTD0015)
- DOI：10.19713/j.cnki.43-1423/u.T20222451
  中图分类号：
扫描看全文
任娟娟,张亦弛,刘伟等.基于PSO-SVM的无砟轨道路基沉降病害识别[J].铁道科学与工程学报,2023,20(11):4400-4411.

REN Juanjuan,ZHANG Yichi,LIU Wei,et al.Identification of subgrade settlement diseases in ballastless track based on PSO-SVM[J].Journal of Railway Science and Engineering,2023,20(11):4400-4411.
任娟娟,张亦弛,刘伟等.基于PSO-SVM的无砟轨道路基沉降病害识别[J].铁道科学与工程学报,2023,20(11):4400-4411. DOI： 10.19713/j.cnki.43-1423/u.T20222451.

REN Juanjuan,ZHANG Yichi,LIU Wei,et al.Identification of subgrade settlement diseases in ballastless track based on PSO-SVM[J].Journal of Railway Science and Engineering,2023,20(11):4400-4411. DOI： 10.19713/j.cnki.43-1423/u.T20222451.

摘要

为了探究实时、准确的路基不均匀沉降识别方法，以CRTS Ⅰ双块式无砟轨道路基沉降病害为研究对象，建立车辆-轨道-路基垂向耦合动力学模型，讨论不同路基沉降状态下的车辆-轨道系统振动规律，选取路基沉降识别敏感特征，并基于粒子群优化支持向量机算法实现对无砟轨道路基沉降病害的有效识别。研究结果表明：钢轨及道床的垂向位移对路基沉降变化较为敏感，随沉降幅值的增大而增大，随沉降波长的增大而减小，而路基沉降对钢轨及道床垂向振动加速度影响较小，利用轨道结构振动响应判断路基沉降状态可行性较低。车体、转向架及轮对垂向振动加速度随着路基沉降幅值增大而增大，其中车体、转向架对路基沉降幅值变化相对敏感，而轮对相对不敏感。随着沉降波长增加，车体与转向架垂向振动加速度先增大后减小，车体对沉降波长的敏感程度远高于转向架，故可将车体垂向振动加速度作为识别敏感特征。车体垂向加速度振动信号对无砟轨道路基沉降波长的识别准确率高于对沉降幅值的识别准确率，其中沉降幅值为20 mm时的识别准确率为84.78%，表明算法在该工况下的识别性能相对较低，但仍能保证一定的准确率，而对于无沉降和不同沉降波长工况，算法识别准确率接近100%。研究成果证明了粒子群优化支持向量机算法可实现对无砟轨道路基沉降的有效识别。

Abstract

In order to investigate the real-time and accurate uneven settlement identification method, a vehicle-track-subgrade vertical coupling dynamics model was established for subgrade settlement problem of CRTS Ⅰ twin-block ballastless track. The vibration law of vehicle-track system under different subgrade settlements was discussed, and the sensitive features of subgrade settlement identification were selected. The particle swarm optimization support vector machine algorithm was used to realize the effective identification of ballastless track subgrade settlement. The results are drawn as follows. The vertical displacement of the rail and track bed is sensitive to the change of subgrade settlement, increases with the increase of settlement amplitude and decreases with the increase of settlement wavelength. The subgrade settlement has less influence on the vertical vibration acceleration of the rail and track bed, and the feasibility of using the vibration response of the rail structure to judge the subgrade settlement state is low. The car body, bogie and wheel pair vertical vibration acceleration increases with the increase of subgrade settlement amplitude, among which the car body and bogie are more sensitive to the change of subgrade settlement amplitude, and the wheel pair is relatively insensitive. As the settlement wavelength increases, the car body and bogie vertical vibration acceleration first increases and then decreases. The car body is much more sensitive to the settlement wavelength than the bogie, so the car body vertical vibration acceleration can be used as a sensitive feature for recognition. The recognition accuracy of the car body vertical acceleration vibration signal to the ballastless track subgrade settlement wavelength is higher than the recognition accuracy of the settlement amplitude. The recognition accuracy of the algorithm is 84.78% when the settlement amplitude is 20 mm. It shows that the algorithm has a relatively low recognition performance in this working condition, but can still guarantee a certain accuracy rate. The algorithm recognition accuracy is close to 100% for non-settlement and different settlement wavelength conditions. The research results demonstrate that the particle swarm optimization support vector machine algorithm can achieve effective recognition of ballastless track subgrade settlement.

关键词

无砟轨道路基沉降动力响应粒子群优化算法支持向量机

Keywords

ballastless tracksubgrade settlementdynamic responseparticle swarm optimization algorithmsupport vector machine

references

REN Juanjuan, ZHANG Kaiyao, ZHENG Jianlong, et al. Railway subgrade thermal-hydro-mechanical behavior and track irregularity under the sunny-shady slopes effect in seasonal frozen regions[J]. Journal of Central South University, 2022, 29(11): 3793-3810.

ZHANG Keping, ZHANG Xiaohui, ZHOU Shunhua. Effect of lateral differential settlement of high-speed railway subgrade on dynamic response of vehicle-track coupling systems[J]. Structural Engineering and Mechanics, 2021, 80(5): 491-501.

GUO Yu, ZHAI Wanming, SUN Yu. A mechanical model of vehicle-slab track coupled system with differential subgrade settlement[J]. Structural Engineering and Mechanics, 2018, 66: 15-25.

石熊. 高速铁路路基动力累积变形模型试验研究[J]. 铁道科学与工程学报, 2020, 17(6): 1346-1355.

SHI Xiong. The model test of dynamic accumulative deformation of high-speed railway track foundation[J]. Journal of Railway Science and Engineering, 2020, 17(6): 1346-1355.

缪炳荣, 刘俊利, 张盈, 等. 轨道车辆结构振动损伤识别技术综述[J]. 交通运输工程学报, 2021, 21(1): 338-357.

MIAO Bingrong, LIU Junli, ZHANG Ying, et al. Review on structural vibration damage identification technology for railway vehicles[J]. Journal of Traffic and Transportation Engineering, 2021, 21(1): 338-357.

AZAMI M, SALEHI M. Response-based multiple structural damage localization through multi-channel empirical mode decomposition[J]. Journal of Structural Integrity and Maintenance, 2019, 4(4): 195-206.

AVCI O, ABDELJABER O, KIRANYAZ S, et al. A review of vibration-based damage detection in civil structures: from traditional methods to Machine Learning and Deep Learning applications[J]. Mechanical Systems and Signal Processing, 2021, 147: 107077.

TANG Liqun, LUO Xu, LIU Zejia, et al. Octonion structural response vector and potential structural damage identification method[J]. International Journal of Damage Mechanics, 2013, 22(4): 572-589.

ABDELJABER O, AVCI O, KIRANYAZ S, et al. Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks[J]. Journal of Sound and Vibration, 2017, 388: 154-170.

刘泽佳, 金梦茹, 周立成, 等. 基于结构响应向量与支持向量机的桥梁损伤识别方法[J]. 济南大学学报(自然科学版), 2020, 34(2): 106-112.

LIU Zejia, JIN Mengru, ZHOU Licheng, et al. Bridge damage identification method based on structural response vectors and support vector machine algorithms[J]. Journal of University of Jinan (Science and Technology), 2020, 34(2): 106-112.

任娟娟, 杜威, 叶文龙, 等. 基于PSO-SVM的板式无砟轨道CA砂浆脱空损伤识别[J]. 中南大学学报(自然科学版), 2021, 52(11): 4021-4031.

REN Juanjuan, DU Wei, YE Wenlong, et al. Contact loss identification of CA mortar in prefabricated slab track based on PSO-SVM[J]. Journal of Central South University (Science and Technology), 2021, 52(11): 4021-4031.

张龙, 彭小明, 熊国良, 等. 基于MSE与PSO-SVM的机车轮对轴承智能诊断方法[J]. 铁道科学与工程学报, 2021, 18(9): 2408-2417.

ZHANG Long, PENG Xiaoming, XIONG Guoliang, et al. Intelligent diagnosis of locomotive wheelset bearings using MSE and PSO-SVM[J]. Journal of Railway Science and Engineering, 2021, 18(9): 2408-2417.

REN Juanjuan, YANG Rongshan, WANG Ping, et al. Influence of contact loss underneath concrete underlayer on dynamic performance of prefabricated concrete slab track[J]. Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, 2017, 231(3): 345-358.

THOMPSON D J. Experimental analysis of wave propagation in railway tracks[J]. Journal of Sound and Vibration, 1997, 203(5): 867-888.

肖威, 郭宇, 高建敏, 等. 高速铁路路基不均匀沉降对CRTS Ⅲ板式轨道受力变形的影响[J]. 铁道科学与工程学报, 2015, 12(4): 724-730.

XIAO Wei, GUO Yu, GAO Jianmin, et al. Effect of uneven subgrade settlement on the CRTS III slab track stress and deformation of high-speed railway[J]. Journal of Railway Science and Engineering, 2015, 12(4): 724-730.

吴斌, 张勇, 徐庆元, 等. 路基上双块式无砟轨道道床板空间力学特性研究[J]. 铁道科学与工程学报, 2010, 7(6): 24-29.

WU Bin, ZHANG Yong, XU Qingyuan, et al. Study on spatial mechanical characteristic of track concrete layer of twin-block ballastless track on subgrade[J]. Journal of Railway Science and Engineering, 2010, 7(6): 24-29.

蔡成标, 翟婉明, 王开云. 遂渝线路基上板式轨道动力性能计算及评估分析[J]. 中国铁道科学, 2006, 27(4): 17-21.

CAI Chengbiao, ZHAI Wanming, WANG Kaiyun. Calculation and assessment analysis of the dynamic performance for slab track on Sui-yu railway[J]. China Railway Science, 2006, 27(4): 17-21.

郭宇. 高速铁路路基不均匀沉降及其演化对车辆-轨道耦合系统力学性能的影响[D]. 成都: 西南交通大学, 2018.

GUO Yu. Influence of uneven settlement of high-speed railway subgrade and its evolution on mechanical properties of vehicle-track coupling system[D].Chengdu: Southwest Jiaotong University, 2018.

单文娣. 基于车辆轨道动力特性的轨道基础沉降智能识别方法[D]. 北京: 北京交通大学, 2012.

SHAN Wendi. Intelligent identification method of track foundation settlement based on vehicle track dynamic characteristics[D].Beijing: Beijing Jiaotong University, 2012.

姚菲, 陆幸奇, 陈光宇. 基于冲击回波法的混凝土-围岩缺陷检测与信号处理研究[J]. 铁道科学与工程学报, 2021, 18(9): 2316-2323.

YAO Fei, LU Xingqi, CHEN Guangyu. Experimental and signal processing research on concrete-rock structural defects by impact-echo method[J]. Journal of Railway Science and Engineering, 2021, 18(9): 2316-2323.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

高速列车荷载作用下无砟轨道-路基-地基精细化有限元模型与验证

考虑CA砂浆黏弹特征的板式无砟轨道动力响应分析

基于支持向量机的CRTS Ⅱ型板式无砟轨道板正温度梯度预警方法

墩顶水平冲击作用下承插式桥墩动力响应研究

高速铁路拱塔斜拉桥上无砟轨道隔离层变形协调性能研究