YOLO-VSI: An Improved YOLOv8 Model for Detecting Railway Turnouts Defects in Complex Environments

Railway turnouts often develop defects such as chipping,cracks,and wear during use.If not detected and addressed promptly,these defects can pose significant risks to train operation safety and passenger security.Despite advances in defect detection technologies,research specifically targeting railway turnout defects remains limited.To address this gap,we collected images from railway inspectors and constructed a dataset of railway turnout defects in complex environments.To enhance detection accuracy,we propose an improved YOLOv8 model named YOLO-VSS-SOUP-Inner-CIoU(YOLO-VSI).The model employs a state-space model(SSM)to enhance the C2f module in the YOLOv8 backbone,proposed the C2f-VSS module to better capture long-range dependencies and contextual features,thus improving feature extraction in complex environments.In the network’s neck layer,we integrate SPDConv and Omni-Kernel Network(OKM)modules to improve the original PAFPN(Path Aggregation Feature Pyramid Network)structure,and proposed the Small Object Upgrade Pyramid(SOUP)structure to enhance small object detection capabilities.Additionally,the Inner-CIoU loss function with a scale factor is applied to further enhance the model’s detection capabilities.Compared to the baseline model,YOLO-VSI demonstrates a 3.5%improvement in average precision on our railway turnout dataset,showcasing increased accuracy and robustness.Experiments on the public NEU-DET dataset reveal a 2.3%increase in average precision over the baseline,indicating that YOLO-VSI has good generalization capabilities.

An improved YOLOv7 for the state identification of sliding chairs in railway turnout

The sliding chairs are important components that support the switch rail conversion in the railway turnout.Due to the harsh environmental erosion and the attack from the wheel vibration,the failure rate of the sliding chairs accounts for up to 10%of the total failure number in turnout.However,there is little research carried out in the existing literature to diagnose the deterioration states of the sliding chairs.To fill out this gap,by utilizing the images containing the sliding chairs,we propose an improved You Only Look Once version 7(YOLOv7)to identify the state of the sliding chairs.Specifically,to meet the challenge brought by the small inter-class differences among the sliding chair states,we first integrate the Convolutional Block Attention Module(CBAM)into the YOLOv7 backbone to screen the information conducive to state identification.Then,an extra detector for a small object is customized into the YOLOv7 network in order to detect the small-scale sliding chairs in images.Meanwhile,we revise the localization loss in the objective function as the Efficient Intersection over Union(EIoU)to optimize the design of the aspect ratio,which helps the localization of the sliding chairs.Next,to address the issue caused by the varying scales of the sliding chairs,we employ K-means++to optimize the priori selection of the initial anchor boxes.Finally,based on the images collected from real-world turnouts,the proposed method is verified and the results show that our method outperforms the basic YOLOv7 in the state identification of the sliding chairs with 4%improvements in terms of both mean Average Precision@0.5(mAP@0.5)and F1-score.

Influence of span-to-depth ratio on dynamic response of vehicle-turnoutbridge system in high-speed railway

For high-speed railways,the smoothness of the railway line significantly affects the operational speed of trains.When the train passes through the turnout on a long-span bridge,the wheel-rail impacts caused by the turnout structure irregularities,and the instability arising from the bridge's flexural deformation lead to a strong coupling effect in the vehicle-turnout-bridge system.This significantly affects both ride comfort and operational safety.For addressing this issue,the present study considered a long-span continuous rigid-frame bridge as an example and established a train-turnout-bridge coupled dynamic model of high-speed railway.Utilizing a selfdeveloped dynamic simulation program,the study analysed the dynamic response characteristics when the train passes through the turnouts on the bridge.It also investigated the influence of different span-to-depth ratios of the bridge on the vehicle dynamic response when the train passes through the main line and branch line of turnouts and then proposed a span-to-depth ratio limit value for a long-span continuous rigid-frame bridge.The research findings suggest that the changes in the span-to-depth ratio have a relatively minor impact on the train’s operational performance but significantly affect the dynamic characteristics of the bridge structure.Based on the findings and a comprehensive assessment of safety indicators,it is advisable to establish a span-to-depth ratio limit of 1/4500 for a long-span continuous rigid-frame bridge.

Seasonal Behavior of Pavement in Geothermal Snow-Melting System with Solar Energy Storage

A two-dimensional unsteady heat transfer model of pavement of geothermal road snow-melting system (GRSS) with solar energy storage is established and numerical simulation is carried out based on annual hourly meteorological data and boundary conditions. Simulated results show that ground surface temperature and heating flux decrease with the increase of buried depth, but increase with the increase of fluid temperature in winter. Heat-extracted amount and efficiency drop with the increase of fluid temperature in summer.Compared with ambient temperature, solar radiation has more direct influence on the heat-extracted flux of pipe walls of GRSS in summer. The relationships among maximum and idling snow-melting load, the rate of snowfall, ambient temperature and wind speed are made clear, which provides necessary references for the design and optimization of a practical road snow-melting system.

A Convolutional Autoencoder Based Fault Detection Method for Metro Railway Turnout

Railway turnout is one of the critical equipment of Switch&Crossing(S&C)Systems in railway,related to the train’s safety and operation efficiency.With the advancement of intelligent sensors,data-driven fault detection technology for railway turnout has become an important research topic.However,little research in the literature has investigated the capability of data-driven fault detection technology for metro railway turnout.This paper presents a convolutional autoencoder-based fault detection method for the metro railway turnout considering human field inspection scenarios.First,the one-dimensional original time-series signal is converted into a twodimensional image by data pre-processing and 2D representation.Next,a binary classification model based on the convolutional autoencoder is developed to implement fault detection.The profile and structure information can be captured by processing data as images.The performance of our method is evaluated and tested on real-world operational current data in themetro stations.Experimental results show that the proposedmethod achieves better performance,especially in terms of error rate and specificity,and is robust in practical engineering applications.

Study on wear characteristics and authorized limits of switch rails of high-speed turnout

Purpose-It is quite universal for high-speed turnouts to be exposed to the wear of the stock rail of the switch rail during the service process.The wear will cause the change of railhead profile and the relative positions of the switch rail and the stock rail,which will directly affect the wheel-rail contact state and wheel load transition when a train passes the turnout and will further impose serious impacts on the safety and stability of train operation.The purpose of this paper is to provide suggestions for wear management of high-speed turnout.Design/methodology/approach-The actual wear characteristics of switch rails of high-speed turnouts in different guiding directions were studied based on the monitoring results on site;the authorized wear limits for the switch rails of high-speed turnout were studied through derailment risk analysis and switch rail strength analysis.Findings-The results show that:the major factor for the service life of a curved switch rail is the lateral wear.The wear characteristics of the curved switch rail of a facing turnout are significantly different from those of a trailing turnout.To be specific,the lateral wear of the curved switch rail mainly occurs in the narrower section at its front end for a trailing turnout,but in the wider section at its rear end when for a facing turnout.The maximum lateral wear of a dismounted switch rail from a trailing turnout is found on the 15-mm wide section and is 3.9 mm,which does not reach the specified limit of 6 mm.For comparison,the lateral wear of a dismounted switch rail from a facing turnout is found from the 35-mm wide section to the full-width section and is greater than 7.5 mm,which exceeds the specified limit.Based on this,in addition to meeting the requirements of maintenance rules,the allowed wear of switch rails of high-speed turnout shall be so that the dangerous area with a tangent angle of wheel profile smaller than 43.68 will not contact the switch rail when the wheel is lifted by 2 mm.Accordingly,the lateral wear limit at the 5-mm wide section of the curved switch rail shall be reduced from 6 mm(as specified)to 3.5 mm.Originality/value-The work in this paper is of reference significance to the research on the development law of rail wear in high-speed turnout area and the formulation of relevant standards.

基于参数优化VMD和改进LSSVM的道岔故障诊断方法

为了解决道岔设备智能故障诊断中特征指标难以提取以及模型训练时间较长的问题,以ZDJ9型转辙机带动的道岔设备为研究对象,以转辙机功率曲线为数据基础,提出一种基于参数优化变分模态分解(Variational Mode Decomposition,VMD)和改进最小二乘支持向量机(Least Squares Support Vector Machines,LSSVM)的道岔故障诊断方法。首先,采用鲸鱼优化算法(Whale Optimization Algorithm,WOA)优化VMD参数,得到模态(Intrinsic Mode Functions,IMF)分量个数和惩罚因子的最优参数组合。其次,计算IMF分量与功率曲线的相关系数,优选相关性较大的前3阶IMF分量,并计算功率谱熵、模糊熵及包络熵值,建立多特征融合样本数据库。最后,针对麻雀搜索算法(Sparrow Search Algorithm,SSA)易陷入局部最优的问题,通过改进Tent混沌映射初始化策略随机生成种群,正余弦算法(Sine Cosine Algorithm,SCA)更新追随者的位置,并采用改进SSA优化LSSVM算法的惩罚因子和核函数方差,构建基于TSSSA-LSSVM的道岔故障诊断模型。实验结果表明:所提道岔故障诊断方法是可行的,采用多特征融合能够更加全面地提取道岔典型故障特征,反映道岔的真实运行状态,提高了故障诊断准确率,而且较TSSSA-SVM,PSO-LSSVM,GWO-LSSVM以及SSA-LSSVM等方法具有较高的故障诊断准确率、召回率以及较低的漏报率,减少了模型训练时间,完全满足现场道岔故障导向安全的原则,具有更好的故障诊断性能,对现场道岔设备的故障维修具有一定的指导意义。

9号道岔区铺设减振垫浮置板的行车安全性及舒适性研究

地铁道岔区在基本轨、尖轨处以及道岔钢轨接头处存在大量钢轨不连续地段,轮轨相互作用强烈,轨道系统振动明显,减振垫浮置板轨道作为一种高等减振措施已应用于地铁线路道岔区。为探究减振垫浮置板应用于9号道岔的适用性,并作为减振地段与普通无砟轨道的过渡段的可行性,以某地铁高架9号道岔为研究对象,建立桥上道岔三维有限元模型,仿真分析地铁B型车在120 km/h直向过岔和30 km/h侧向过岔的情况下,列车的行车安全性、舒适性以及轨下结构形变。对比在9号单开道岔区域设置的2种刚度过渡方式的优劣及其减振效果,较为系统地论证了在9号道岔区域铺设减振垫浮置板的可行性。研究结果表明:1)列车通过铺设减振垫浮置板的9号道岔区域时,浮置板位移极值出现在转辙区域和心轨区域,最大位移出现在尖轨位置,可增大尖轨处单板橡胶垫支承刚度,以保证轨道结构变形均匀;2)列车逆向侧向通过设置过渡段的减振垫浮置板道岔区时,2级过渡方式的车体最大垂向加速度、最大横向加速度、最大脱轨系数、最大轮重减载率、轨道刚度变化率以及减振效果等指标均优于1级过渡方式;3)在道岔区铺设减振垫浮置板,列车直向和侧向通过时的行车安全性和舒适性均能满足相关评价指标,且2级过渡方式更优。研究结果可为后续道岔区减振设计研究提供一定的理论参考。

矿用单轨吊气动道岔的结构设计与研究

我国煤炭现代化高速发展,生产规模也日益增大,所需设备的自动化、机械化程度也越来越高,但使用的辅助运输系统却十分落后。为了克服上述困扰,提出了一种带有气动道岔的单轨吊辅助运输设备,能够克服卡轨车运输条件的限制,使得多条巷道能够连通使用。这种设备在多弯道、多起伏的巷道中具有较高的灵活性和适应性,对于优化运输环境、提升运输效率、确保运输安全具有重大的实际价值。其结构设计包含道岔部分,其中道岔部分主要由道岔直轨、弯轨和气缸构成。首先通过SOLIDWORKS进行三维建模,再通过ANSYS WORKBENCH对轨道部分进行有限元静力学分析,可以直观的呈现出轨道受力变形情况。

新型高速道岔结构研发及列车过岔安全性分析

为了减小尖轨对列车高速运行安全性的影响,研发了一种新型高速道岔转辙器结构。新型转辙器去掉了既有结构中的尖轨,全部采用标准60 kg/m钢轨的轨头断面;转辙轨一端固定,另一端可左右转动,从而实现转向。基于车辆-道岔耦合动力学理论,分别建立标准18号道岔、新型道岔模型以及CRH_(2)型车的单车模型;通过与标准道岔对比,分析列车侧逆向通过新型道岔的安全性。结果表明:与标准道岔相比,列车侧逆向通过新型道岔时,在转辙器区域脱轨系数由0.49降为0.32,降低34.69%;轮轨横向力峰值由40.24 kN降为26.51 kN,降低34.12%;轮轨垂向力峰值由90.71 kN降到84.66 kN,降低6.67%。列车通过新型道岔时轮重减载率在跟端轨缝位置出现突变,在转辙器区域优于标准道岔。

高原地区无缝道岔健康监测研究

研究目的:无缝道岔是无缝线路的关键技术之一,高原地区的气候复杂使其受力更为复杂,为保证列车的安全运营,有必要对其进行长期实时健康监测。本文针对高原地区无缝道岔构建实时健康监测系统,并对一组道岔进行为期2个月的在线监测,对监测期内的轨温、尖轨尖端位移及尖轨断轨进行相应分析。研究结论:(1)钢轨温度变化趋势与气温变化趋势相同,最高轨温始终高于日最高气温,平均差值为11.75℃,其出现时间总体来讲早于气温出现3 h左右;(2)最低轨温与气温值基本相同,其出现时间也基本相近,大致于早上7:00—8:00出现;(3)钢轨温度和大气温度之间存在较好的一次线性关系,当气温变化1℃时轨温变化约为1.5℃;(4)尖轨尖端的纵向位移变化趋势与气温和轨温变化趋势相同,与气温间呈一次线性关系;(5)本文研究可为高原地区无缝道岔的养护维修提供数据支撑。

基于GADF与2D CNN-改进SVM的道岔故障诊断方法研究

针对道岔故障特征不易提取以及道岔故障诊断准确率较低的问题,提出一种格拉姆角差场(Gramian Angular Difference Fields, GADF)与二维卷积神经网络(Two Dimensional Convolutional Neural Network, 2D CNN)-改进支持向量机(Support Vector Machine, SVM)的道岔故障诊断组合方法。首先,结合现场实际应用情况,选取道岔设备正常转换与典型故障的转辙机功率曲线,建立转辙机功率曲线样本数据库;采用GADF编码将一维转辙机功率曲线信号转换为具有时间相关性的二维特征图,分别选择16×16、32×32以及64×64大小的特征图并提取图像数据。其次,在LeNet-5模型的基础上设计2D CNN网络结构,并将图像数据输入至基于2D CNN的道岔故障特征提取模型中,经多层的卷积层、池化层以及全连接层提取特征指标,建立道岔故障诊断样本数据库。最后,通过北方苍鹰优化(Northern Goshawk Optimization, NGO)算法优化SVM算法的惩罚因子与核函数方差,构建基于NGO-SVM的道岔故障诊断模型。实验结果分析表明,将转辙机功率曲线数据经GADF编码为64×64大小的特征图,并通过2D CNN模型提取道岔典型特征数据,较其他数据处理方法具有较高的故障诊断准确率,同时提高了故障诊断实时性;将建立的道岔故障诊断样本数据库输入至NGO-SVM道岔故障诊断模型,其故障诊断准确率高达97.5%,较其他故障诊断模型具有更好的故障诊断性能,为道岔故障诊断提供了一种新方法,对现场道岔设备的日常维修具有一定的指导意义。

复杂条件下道岔主要病害的整治措施

高海拔、强紫外线、冻土等复杂环境及线路条件对道岔现有结构、扣件系统、跨区间无缝线路等关键技术的适应性提出了严峻挑战。为提高道岔产品的安全性能,减少缺氧环境下的维护作业量,本文调研了复杂条件下道岔的主要病害,并提出针对性整治措施。主要措施包括:通过优化道岔平面线形和尖轨结构、材质、制造工艺,延长曲线尖轨的服役寿命;通过优化辙叉结构、材料成分及制造工艺,解决心轨裂纹及剥落掉块问题;通过采用岔枕预埋铁座式扣件、加强型Ⅱ型弹条及弹性夹,提升扣件系统稳定性能;通过采用防护电务整体式岔枕,顶铁和间隔铁预留融雪加热条安装空间,避免冰雪及异物干扰电务转换系统,改善了工电接口;通过采用60AT2钢轨、辊轮装置,对台板进行减摩处理,减小了道岔扳动力;通过设置螺栓防松机构,减少了道岔维护作业量。

面向高速铁路道岔运维文本的知识抽取方法

为实现高速铁路道岔运维知识图谱的自动构建,从而为道岔智能运维提供决策支持,须利用知识抽取技术从高速铁路道岔运维文本中提取关键知识。同时,为进一步解决道岔运维文本中的实体嵌套与知识三元组重叠问题,提出一种基于多模块联合学习的高速铁路道岔运维知识抽取模型RTOM-KE。首先,依据定义的实体与关系类型,采用基于BIOES的2阶段知识标注策略,分别标注头实体以及相对应关系下的尾实体;其次,通过轻量级预训练模型BERT-base与双向长短期记忆神经网络BiLSTM构成的编码器模块获取文本的多维共享编码表示,将编码器模块的隐藏状态和全局上下文特征组合之后作为头实体抽取模块的输入;最后,通过头实体抽取模块提取文本中的所有候选头实体。同时,将候选头实体标签和来自编码器模块的多维共享词表征作为尾实体抽取模块的联合输入,通过特定的关系门机制筛选与头实体相关联的尾实体,从而获得高速铁路道岔运维知识三元组。通过充分的对比实验以及消融实验,结果表明:RTOM-KE模型能够较为精准全面地抽取不同复杂程度的三元组,有效解决实体嵌套与三元组重叠问题,基于道岔运维数据集的模型精确率、召回率与F1值分别可达88.3%、86.9%和87.6%。研究结果可为进一步构建高速铁路道岔运维知识图谱以及道岔智能运维提供支持。

多域特征提取结合AdaBoost的含未知故障提速道岔故障诊断方法

针对提速道岔未知新故障误判影响列车安全运行及道岔检修效率的问题,提出一种基于多域特征提取和自适应提升算法(Adaptive boosting, adaboost)的信号分析及故障诊断模型。首先,为了深入挖掘道岔的故障特征,分别从时域、频域及时频域提取故障特征,构造原始特征集;然后根据AdaBoost模型获得的特征重要度排序构造不同特征数量的分类模型,并利用模型分类精度进一步获得最佳特征子集;最后将最佳特征子集输入含判定机制的AdaBoost故障诊断模型,完成对提速道岔含未知故障类型的诊断,同时,通过对模型的再训练,实现了对现有故障诊断模型的自适应更新。结果表明:本文方法在有效提取故障特征,提高道岔已知类故障诊断精度的同时,可以有效地识别出道岔之前未出现的新故障。

面向高速铁路道岔设备故障处置的知识图谱构建与应用

为了解决道岔设备故障处置过程中产生的大量知识难以存储应用,半结构化故障文本数据不能直接检索和推理等问题,提出一种高速铁路道岔设备故障处置知识图谱的构建方法。以少量人工标注故障文本数据为基础,采用基于强化学习的方法搭建BERT-BiLSTM-CRF*深度学习模型抽取实体;根据知识领域特点和数据结构设计关系匹配模板抽取各实体间的关系;针对实体冗余问题,采用文本相似度与结构相似度结合的方法进行故障实体融合;以我国各铁路局近3年的高速铁路道岔设备故障文本为研究对象,累计抽取实体807个且标注准确率在90%以上,构建实体关系2232条。基于以上知识图谱构建的高速铁路道岔设备故障处置知识图谱直观展示道岔设备故障关系,实现不同实体对象间的数据关联查询,挖掘隐含关系,辅助道岔维护人员进行故障处置决策,实现道岔设备预防性维护,延长设备寿命。

悬挂式单轨交通车辆基地站场设计关键技术研究

[目的]悬挂式单轨交通车辆基地的总平面布局形式直接影响列车作业及工程造价。道岔选型对车辆基地站场总平面布局影响较大,运用库单线列位数的选择也将直接影响道岔数量及段场规模,因此需对悬挂式单轨交通车辆基地站场设计的关键技术进行深入研究。[方法]依托武汉光谷生态大走廊悬挂式单轨交通工程的龙泉山车辆段,阐述了该车辆段的总平面布局形式及创新点,分析了该车辆段道岔选型及场坪标高的制定方法,建立了理想条件下车辆段运用库单线列位数模型。[结果及结论]悬挂式单轨交通车辆基地总平面布置宜采用贯通式洗车方案,将洗车库放在出入线上;咽喉设计时,需同时考虑桥墩布置的美观性;需联合道岔厂家加快对三开道岔及五开道岔的研发,并合理选用不同道岔组合以更进一步节省投资;在不考虑其他作业干扰的理想条件下,悬挂式单轨交通车辆基地运用库最多可采用1线5列位,实际设计过程中需结合车辆基地条件、综合考虑列车出库的便捷性、扣除其他作业干扰等因素选择合适的运用库单线列位数。

市域(郊)铁路桥上无缝道岔至梁缝距离研究

市域(郊)铁路桥梁占比高,结合车站配线设计,道岔不可避免需要设置在高架站前后的道岔连续梁上。考虑市域(郊)铁路采用的道岔型号、桥梁结构及设计荷载与高速铁路明显不同,桥上无缝道岔至梁缝距离研究相对缺乏。由于市域(郊)铁路设计速度比地铁高,轮轨作用力大,轮轨之间的振动和冲击对周围环境影响大,因此,桥上道岔采用跨区间无缝线路技术。基于“道岔-桥梁”相互作用原理,建立“岔-板-梁-墩”一体化12号无砟轨道、无缝道岔有限元模型,对常用的单开道岔、4.3 m线间距渡线道岔至梁缝距离开展计算分析并研究其影响规律,以道岔钢轨强度、尖轨尖端位移值、心轨尖端位移值、转辙机处基本轨与桥梁相对位移值及断缝值为限值条件,提出市域(郊)铁路桥上无缝道岔至梁缝合理距离限值。研究结果表明:桥上无缝道岔始终端距梁缝距离越大,尖轨尖端位移、心轨尖端位移和转辙机处基本轨与桥梁相对位移明显减小,道岔钢轨伸缩力和制动力计算结果差异不大;温暖地区道岔始端、终端至梁缝距离不小于10 m,寒冷地区道岔始端、终端至梁缝距离不小于18 m,相关研究成果纳入《市域(郊)铁路设计规范》。

基于MDS和改进SSA-SVM的高速铁路道岔故障诊断方法研究

针对高速铁路道岔设备故障频繁,现场维修工作量大等问题,提出基于多维尺度缩放法(MDS)和改进麻雀搜索算法(SSA)优化支持向量机(SVM)的高速铁路道岔故障诊断模型。首先以ZDJ9道岔转换功率曲线为研究对象,总结现场典型道岔故障类型及故障原因,分别提取道岔功率曲线的时域、频域特征指标以及小波包能量熵,组成特征指标向量;其次采用MDS方法进行多维特征指标的降维优化,建立道岔故障特征指标样本数据库;最后利用改进Circle混沌映射初始化种群,并通过自适应t分布增强麻雀种群的多样性,再以改进SSA算法优化SVM模型中的惩罚因子和核函数方差2个关键参数,构建改进SSA-SVM的道岔故障诊断模型。故障诊断结果表明,本模型的故障诊断正确率高达96.25%,诊断效果优于其他方法,可以为道岔设备的故障维修提供理论依据。

市域铁路无砟道岔扣件系统设计及性能试验

市域铁路一般采用无砟轨道,为此设计无砟道岔扣件系统。道岔扣件系统板下弹性垫层采用三元乙丙橡胶,实现了无砟轨道道床低刚度要求;岔枕螺栓安装处设置具有防转功能的复合偏心套和防转盖板,增大道岔轨距调整范围,同时改善岔枕螺栓受力,增强扣件系统稳定性。扣件系统性能试验研究表明,扣件系统各项指标均满足规范要求,同时具有良好的减振性能。