YOLO-O2E:A Variant YOLO Model for Anomalous Rail Fastening Detection

Rail fasteners are a crucial component of the railway transportation safety system.These fasteners,distinguished by their high length-to-width ratio,frequently encounter elevated failure rates,necessitating manual inspection and maintenance.Manual inspection not only consumes time but also poses the risk of potential oversights.With the advancement of deep learning technology in rail fasteners,challenges such as the complex background of rail fasteners and the similarity in their states are addressed.We have proposed an efficient and high-precision rail fastener detection algorithm,named YOLO-O2E(you only look once-O2E).Firstly,we propose the EFOV(Enhanced Field of View)structure,aiming to adjust the effective receptive field size of convolutional kernels to enhance insensitivity to small spatial variations.Additionally,The OD_MP(ODConv and MP_2)and EMA(EfficientMulti-Scale Attention)modules mentioned in the algorithm can acquire a wider spectrum of contextual information,enhancing the model’s ability to recognize and locate objectives.Additionally,we collected and prepared the GKA dataset,sourced from real train tracks.Through testing on the GKA dataset and the publicly available NUE-DET dataset,our method outperforms general-purpose object detection algorithms.On the GKA dataset,our model achieved a mAP 0.5 value of 97.6%and a mAP 0.5:0.95 value of 83.9%,demonstrating excellent inference speed.YOLO-O2E is an algorithm for detecting anomalies in railway fasteners that is applicable in practical industrial settings,addressing the industry gap in rail fastener detection.

A Railway Fastener Inspection Method Based on Abnormal Sample Generation

Regular fastener detection is necessary to ensure the safety of railways.However,the number of abnormal fasteners is significantly lower than the number of normal fasteners in real railways.Existing supervised inspectionmethods have insufficient detection ability in cases of imbalanced samples.To solve this problem,we propose an approach based on deep convolutional neural networks(DCNNs),which consists of three stages:fastener localization,abnormal fastener sample generation based on saliency detection,and fastener state inspection.First,a lightweight YOLOv5s is designed to achieve fast and precise localization of fastener regions.Then,the foreground clip region of a fastener image is extracted by the designed fastener saliency detection network(F-SDNet),combined with data augmentation to generate a large number of abnormal fastener samples and balance the number of abnormal and normal samples.Finally,a fastener inspection model called Fastener ResNet-8 is constructed by being trained with the augmented fastener dataset.Results show the effectiveness of our proposed method in solving the problem of sample imbalance in fastener detection.Qualitative and quantitative comparisons show that the proposed F-SDNet outperforms other state-of-the-art methods in clip region extraction,reaching MAE and max F-measure of 0.0215 and 0.9635,respectively.In addition,the FPS of the fastener state inspection model reached 86.2,and the average accuracy reached 98.7%on 614 augmented fastener test sets and 99.9%on 7505 real fastener datasets.

Regression Method for Rail Fastener Tightness Based on Center-Line Projection Distance Feature and Neural Network

In the railway system,fasteners have the functions of damping,maintaining the track distance,and adjusting the track level.Therefore,routine maintenance and inspection of fasteners are important to ensure the safe operation of track lines.Currently,assessment methods for fastener tightness include manual observation,acoustic wave detection,and image detection.There are limitations such as low accuracy and efficiency,easy interference and misjudgment,and a lack of accurate,stable,and fast detection methods.Aiming at the small deformation characteristics and large elastic change of fasteners from full loosening to full tightening,this study proposes high-precision surface-structured light technology for fastener detection and fastener deformation feature extraction based on the center-line projection distance and a fastener tightness regression method based on neural networks.First,the method uses a 3D camera to obtain a fastener point cloud and then segments the elastic rod area based on the iterative closest point algorithm registration.Principal component analysis is used to calculate the normal vector of the segmented elastic rod surface and extract the point on the centerline of the elastic rod.The point is projected onto the upper surface of the bolt to calculate the projection distance.Subsequently,the mapping relationship between the projection distance sequence and fastener tightness is established,and the influence of each parameter on the fastener tightness prediction is analyzed.Finally,by setting up a fastener detection scene in the track experimental base,collecting data,and completing the algorithm verification,the results showed that the deviation between the fastener tightness regression value obtained after the algorithm processing and the actual measured value RMSE was 0.2196 mm,which significantly improved the effect compared with other tightness detection methods,and realized an effective fastener tightness regression.

Research on Track Fastener Service Status Detection Based on Improved Yolov4 Model

As an important part of railway lines, the healthy service status of track fasteners was very important to ensure the safety of trains. The application of deep learning algorithms was becoming an important method to realize its state detection. However, there was often a deficiency that the detection accuracy and calculation speed of model were difficult to balance, when the traditional deep learning model is used to detect the service state of track fasteners. Targeting this issue, an improved Yolov4 model for detecting the service status of track fasteners was proposed. Firstly, the Mixup data augmentation technology was introduced into Yolov4 model to enhance the generalization ability of model. Secondly, the MobileNet-V2 lightweight network was employed in lieu of the CSPDarknet53 network as the backbone, thereby reducing the number of algorithm parameters and improving the model’s computational efficiency. Finally, the SE attention mechanism was incorporated to boost the importance of rail fastener identification by emphasizing relevant image features, ensuring that the network’s focus was primarily on the fasteners being inspected. The algorithm achieved both high precision and high speed operation of the rail fastener service state detection, while realizing the lightweight of model. The experimental results revealed that, the MAP value of the rail fastener service state detection algorithm based on the improved Yolov4 model reaches 83.2%, which is 2.83% higher than that of the traditional Yolov4 model, and the calculation speed was improved by 67.39%. Compared with the traditional Yolov4 model, the proposed method achieved the collaborative optimization of detection accuracy and calculation speed.

Corrosion Test of the Steel Plate in a WJ-8 Fastener for High Speed Rail

It was found that the steel plate in the composite plate in the WJ-8 fastener used in high speed rail is rusty. The objective of this study is to test the zinc coating of the steel plate. A literature review was conducted to identify the zinc coating techniques, and the companies that can provide different coating service was identified. A salt fog chamber was built that was in compliance with the ANSI B117 code, and the steel plates that were coated by the identified companies were tested using the salt fog chamber. The results indicated that the coating technique that had the best performance in preventing corrosion was the Greenkote plates with passivation. The galvanized option had the roughest coating layer, and it was the most reactive in the salt water solution. This makes it non-ideal for the dynamic rail environment because the increased friction of the plate could damage the supports, especially during extreme temperatures that would cause the rail to expand or contract. Greenkote with Phosphate and ArmorGalv also provided increased corrosion prevention with a smooth, strong finish, but it had more rust on the surface area than the Greenkote with ELU passivation. The ArmorGalv sample had more rust on the surface area than the Greenkote samples. This may not be a weakness in the ArmorGalv process;rather, it likely was the result of this particular sample not having the added protection of a colored coating.

Parametric investigation of railway fastenings into the formation and mitigation of short pitch corrugation

Short pitch corrugation has been a problem for railways worldwide over one century.In this paper,a parametric investigation of fastenings is conducted to understand the corrugation formation mechanism and gain insights into corrugation mitigation.A three-dimensional finite element vehicle-track dynamic interaction model is employed,which considers the coupling between the structural dynamics and the contact mechanics,while the damage mechanism is assumed to be differential wear.Various fastening models with different configurations,boundary conditions,and parameters of stiffness and damping are built up and analysed.These models may represent different service stages of fastenings in the field.Besides,the effect of train speeds on corrugation features is studied.The results indicate:(1)Fastening parameters and modelling play an important role in corrugation formation.(2)The fastening longitudinal constraint to the rail is the major factor that determines the corrugation formation.The fastening vertical and lateral constraints influence corrugation features in terms of spatial distribution and wavelength components.(3)The strengthening of fastening constraints in the longitudinal dimension helps to mitigate corrugation.Meanwhile,the inner fastening constraint in the lateral direction is necessary for corrugation alleviation.(4)The increase in fastening longitudinal stiffness and damping can reduce the vibration amplitudes of longitudinal compression modes and thus reduce the track corrugation propensity.The simulation in this work can well explain the field corrugation in terms of the occurrence possibility and major wavelength components.It can also explain the field data with respect to the small variation between the corrugation wavelength and train speed,which is caused by frequency selection and jump between rail longitudinal compression modes.

Derailment risk and dynamics of railway vehicles in curved tracks： Analysis of the effect of failed fasteners

The effect of the fastener＇s failure in a railway track on the dynamic forces produced in the wheel-rail contact is studied using the simulation software VAMPIRE to assess the derailment risk of two different vehicles in two curves with distinct characteristics. First, a 3D-FEM model of a real track is constructed, paying special attention to fasteners, and calibrated with displacement data obtained experimentally during a train passage. This numerical model is subsequently used to determine the track vertical and lateral stiffness. This study evidences that although the track can practically lose its lateral stiffness as a consequence of the failure of 7 consecutive fasteners, the vehicle stability would not be necessarily compromised in the flawed zone. Moreover, the results reveal that the uncompensated acceleration and the distance along which the fasteners are failed play an important role in the dynamic behavior of the vehicle-track system, influencing strongly the risk of derailment.

A study on joining magnesium alloy high pressure die casting components with thread forming fasteners

With tremendous weight saving potential,magnesium alloy high pressure die casting components have been widely used for automotive applications.Magnesium fastening technology is thus becoming increasingly important to design engineers.Joining as-cast holes of magnesium high pressure die casting components with thread forming fasteners provides significant advantages for the assembly process,overall cost benefit and joint integrity.This type of joint is thus preferred for structural applications.Designing the thread forming fasteners with as-cast holes follows the general rules for designing for the machined holes,including carefully designing the variables such as the assembly torque range,the length of thread engagement,the hole diameter and required failure mode.In addition,special attention needs to be paid to the draft angles of the magnesium cast components that are required for the die casting process.In this paper,the effects of above key factors,individually and combined,on the joint performance of thread forming fasteners with as-cast blind holes of AM60B magnesium components are studied.A joint design philosophy was proposed to optimize both joint performance(the prevailing torque,the failure torque and the failure mode)and manufacturing easiness(the hole diameter and corresponding draft angle).The detailed design considerations for as-cast holes of magnesium HPDC are discussed and explained through a hypothetical example.

Rail fastener detection of heavy railway based on deep learning

Image detection based on machine learning and deep learning currently has a good application prospect for railway fault diagnosis,with good performance in feature extraction and the accuracy of image localization and good classification results.To improve the speed of locating small target objects of fasteners,the YOLOv5 framework model with faster algorithm speed is selected.To improve the classification accuracy of fasteners,YOLOv5-based heavy-duty railway rail fastener detection is proposed.The anchor size is modified on the original basis to improve the attention to small targets of fasteners.The CBAM(Convolutional Block Attention Module)module and TPH(Transformer Prediction Head)module are introduced to improve the speed and accuracy issues.The rail fasteners are divided into 6 categories.Experiment comparisons show that before the improvement,the MAP@0.5 value of all categories are close to the peak of 0.989 after the epoch of 150,and the F1 score approaches 1 with confidence in the interval(0.2,0.95).The improved mAP@0.5 value approached the highest value of 0.991 after the epoch of 75,and the F1 score approached 1 with confidence in the interval(0.01,0.95).The experiment results indicate that the improved YOLOv5 model proposed in this paper is more suitable for the task of detecting rail fasteners.

Experimental and Numerical Investigation on the Mechanical Behaviors of the Velcro^(■) and Dual-Lock Fasteners

In order to characterize the mechanical behaviors of the Velcro~? and Dual-lock fasteners, a series of tests including the butt-joint(BJ) monotonic tensile and shear, mixed tensile-shear with various loading angles, the loading rates effects, the double cantilever beam(DCB) fracture and 180° peel experiments were performed. The tensile and shear tests results showed that the mechanical behaviors of Velcro~? fastener separation are analogous to ductile materials, and those of Dual-lock fasteners are more like brittle ones. The mixed tensile-shear with various loading angles tests results demonstrated that magnitudes of the peak stresses in 30°, 45°, and 60° have no significant differences, which are lower than those in the monotonic tensile or shear tests for the two fasteners. The effects of the loading rate tests show that the peak stresses of the Velcro~? fastener manifested good performance at the loading rate of 10 to 20 mm/min in the tensile and shear conditions, and the Dual-lock did it well around the loading rates of 10 to 20 mm/min in the tensile condition. The cohesive zone model(CZM) is employed to numerical predict the DCB fracture and the 180° peel tests. The CZM predictions results are proven to commendably capture the two tests separation processes, of the tow fasteners, and the numerical results agreed well with the peeling tests data of the Dual lock fasteners. The results and discussions in this study are expected to bring more understanding to engineers and designers about the performance of Velcro~? and Dual lock fasteners.

Self-Sensing TDR for Bearing Failure Detection of CFRP Laminate Fastener Hole with Particular Reference to the Effect of Fasteners

Carbon fiber reinforced polymer composites (CFRP) have been applied to aerospace and automobile structures. For many CFRP structures, mechanical metallic fasteners are usually adopted. For the fasteners used in internal structures such as a wing box, the damage to the CFRP structures around fastener holes is visually quite difficult to find. A simple method to find the damage around fastener holes is required. In this study a self-sensing time domain reflectometry (TDR) method is newly applied to detect bearing failure around the fastener holes of CFRP structures. A microstrip-line method is generally used to create a transmission line. When the transmission line is mounted near the metallic fasteners, they may affect the impedance of the transmission line. In this study, the effect of distance between the fasteners and the transmission line was numerically investigated using a finite difference time domain analysis method. After finding the appropriate distance, experiments were performed to detect the bearing failure around a fastener hole. The experiments showed the performance of the self-sensing TDR for detecting bearing failure.

某线震后轨道设计方案研究

2022年1月8日门源发生6.9级地震,造成某线隧道二衬坍塌、桥梁梁体移位、接触网脱落等基础、设备严重损坏,并伴随钢轨折断、扣件脱落、道床倾斜等轨道震害,是我国高速铁路首次遭遇的强震作用下严重受灾事件。结合震后预测余滑变形量水平150~300 mm,垂向100 mm条件下,在设防区段提出一种“三孔连体套管承轨台可调式WJ-8型扣件+长枕埋入式单层道床预留切割孔”新型大调整量无砟轨道结构方案。其中,三孔连体套管承轨台可调式WJ-8型扣件在既有WJ-8型扣件基础上通过增设铁承轨台、调距块和三联套管等部件可满足单股钢轨左右位置调整量达152 mm,长枕埋入式单层道床中预留切割孔,便于在基础变形超出扣件调整范围后切割纠偏;通过对余滑变形后线路拟合获得拨距包络图,确定了大调整量无砟轨道铺设范围。

高速铁路无砟轨道车辆荷载传递特性研究

为进一步探讨无砟轨道荷载设计参数取值,为我国时速400 km及以上高速铁路建设提供重要理论支撑,通过开展无砟轨道静力分析以及钢轨焊缝、车轮扁疤激励下车辆-轨道动力相互作用,揭示列车静、动力载荷特征及传递规律;同时根据轨道承载特点和无砟轨道设计原理,提出车辆竖向设计荷载取值建议。研究结果表明:(1)高速列车车轮扁疤和钢轨焊缝不平顺将造成显著的轮轨冲击,其轮载动载系数最大可达5.79,远高于现行规范取值;扣件系统对扁疤和焊缝不平顺引起的轮轨力高频成分衰减作用明显,扣件支反力的动载系数最大为2.64;(2)钢轨焊缝不平顺引起的轮轨冲击力随着车速提高以及焊缝不平顺幅值的增大而迅速增大,建议500 km/h及以下高速铁路的钢轨焊缝平直度标准取0.2 mm/m;(3)列车轮载通过扣件支反力作用于轨道板,从传递特性的角度来说,以扣件支反力作为轨道板竖向设计荷载更为合理,其动载系数可取3.0,且能满足400 km/h及以上高速铁路的行车需求。

杭州世纪中心西天窗索夹抗滑移试验研究

索夹是预应力索结构中拉索与其他构件连接的重要节点构件,若索夹和拉索之间发生滑移,会产生较大的预应力损失,甚至改变结构的受力性能。杭州世纪中心西天窗采用箱形桁架+钢索结构,其结构特殊性决定了拉索在施工张拉和施工完成后的状态不同,即施工张拉时要求拉索对索夹是可相对滑移的,而施工完成后需要锁定拉索和索夹的相对位置,不允许发生滑移。通过拉索施工张拉、高强螺栓紧固力衰减、索夹顶推三个阶段的全过程模拟试验确定了索夹与拉索之间的摩擦力情况。试验结果表明,在拉索施工张拉阶段,索夹对索体产生的摩擦力较小,通过锌板垫层能够有效降低铸钢索夹节点处的摩擦损失,保证拉索张拉过程中索力的有效传递;高强螺栓终拧后12h内会产生不可忽视的紧固力损失;使用阶段索夹的抗滑移承载能力能够确保索夹节点安全。

扣件系统横向刚度及抗倾翻试验研究:以WJ-8型扣件为例

为了对扣件系统横向刚度及抗倾翻性能进行研究,首先借鉴等间隔弹性点支承梁模型分析钢轨横向力学行为;然后,考虑钢轨自身截面特性,推导扣件系统横向刚度及抗倾翻系数计算公式;最后,设计并进行1组及3组WJ-8型扣件横向加载试验,验证分析及推导的合理性。试验结果表明:当钢轨整体扭转中心为轨底中心点时,钢轨自身扭转不可忽略,钢轨横向力及扭矩的传递过程都是非线性过程;1组及3组扣件试验得到的位移相对误差不超过6.8%,转角相对误差不超过9.5%;3组扣件试验允许加载范围更大,且更适用于非对称式扣件;WJ-8型扣件钢轨横移过程可按扣件横向反力的组成划分为钢轨滑移、线性变形和非线性变形3个阶段,钢轨倾翻过程可按钢轨轨底与轨下垫板接触形式划分为钢轨脱离和整体倾翻2个阶段。扣件横向刚度及抗倾翻系数可通过横向力-位移曲线、扭矩-转角曲线确定。

基于3D线激光传感器的轨道弹条扣件结构缺陷检测方法

轨道扣件缺陷是铁路安全运行的重大安全隐患,基于二维图像处理方法能检测扣件外观缺陷,但难以检测扣件结构缺陷,提出了一种3D线激光传感器的轨道扣件结构缺陷检测方法。首先,利用3D线激光传感器获取轨道点云,根据扣件高度快速定位扣件区域点云,利用PointNet++网络对扣件区域点云分割获得弹条点云;其次,将弹条点云映射至二维图像,在二维图像中提取弹条骨架,二维骨架融合至三维点云获得初始骨架,对每个初始骨架点云拟合截面圆,以各截面圆心作为弹条骨架精确表示弹条轮廓及空间结构;最后,提取弹条三维骨架的特征点,根据特征点构造扣压平面和计算弹条缝隙,基于弹条缝隙检测扣件结构缺陷。为了验证文中方法的有效性,以WJ-7、WJ-8、WJ-2型弹条扣件为对象测量弹条缝隙,并将文中方法测量的弹条缝隙与人工使用缝隙尺测量的真实值进行比较,单个扣件的测量误差在0.1 mm内,且文中方法对轨道油污、锈斑及环境有较好的鲁棒性;对批量扣件的结构缺陷检测,当测量误差允许在±0.1 mm时,扣件结构缺陷检测的准确率不低于95%。

重型燃气轮机拉杆组合转子热瞬变振动特性研究

重型燃气轮机冷态启动的稳速升负荷阶段,常出现轮盘间径向变形不协调导致的相对滑移,拉杆组合转子会发生振动随燃气轮机转子温度的增加而显著变化的热瞬变振动现象。建立了周向拉杆组合转子平面摩擦带止口配合结构轮盘相对滑移的力学模型,包括平面摩擦、周向拉杆和止口与轮盘相对滑移的力学模型。以某型重型燃气轮机拉杆组合转子为研究对象,对其冷态启动过程中出现的热瞬变振动进行了分析,根据轮盘相对滑移的规律,通过减小轮盘间的变形不协调,并适当增加止口配合的过盈量的方法,成功优化了该重型燃气轮机拉杆组合转子冷态启动过程中的热瞬变振动。

铆接夹层间隙对单搭接铆接头振动疲劳性能的影响研究

为了考虑装配间隙对单搭接结构振动疲劳性能的影响,利用ABAQUS软件建立了飞机发动机进气道局部结构金属铆接头的弹塑性有限元模型,采用Johnson-Cook失效模型模拟铆钉与被连接件的渐进失效行为,得到了不同装配间隙量的铆接头应力场分布与振动疲劳寿命。对单搭接铆接头进行振动疲劳试验,试验结果与仿真结果在疲劳寿命方面吻合较好,验证了数值模型的准确性。与无间隙模型相比,含间隙铆接试件各铆钉疲劳寿命降低了4.7%~18.0%,间隙的存在也导致了压铆过程中相邻铆钉间的挤压载荷传递,整体表现为铆钉与被连接板接触界面应力激增,导致有间隙铆接头残余应力场与无间隙铆接头残余应力场之间的差异,从而影响铆接头的振动疲劳性能。

地铁扣件刚度对车致轨道-隧道振动影响分析

为揭示减振扣件动力性能参数对轨道-隧道振动特性的影响规律,对减振扣件轨道与一般非减振轨道的动力响应进行现场实测,对比分析两种类型轨道的钢轨、道床与隧道壁的振动特性,同时建立车辆-轨道耦合动力学仿真模型与轨道-隧道-大地有限元模型,进行不同扣件刚度下钢轨加速度导纳、轨道振动衰减率与隧道动力响应的影响分析。对比线上实测结果发现,减振扣件轨道中钢轨、道床与隧道壁的振动均明显减弱;减小扣件刚度对车体的振动响应无显著影响,钢轨的位移与加速度会随之增大;减小扣件刚度会增强钢轨的低频振动,但会减弱钢轨的高频振动;沿钢轨纵向1~50Hz频段的振动耗散更为迅速,500~1000Hz频段的振动更容易沿着钢轨纵向传播;减振扣件对50~200Hz频段隧道动力响应有明显影响,能有效降低隧道壁测点处振动响应6~8dB。

地铁车底松动紧固件视觉检测算法研究

提高列车车底检测机器人中检测算法的适应性与准确性,对提升列车的智能化运维能力具有重要意义。针对现有紧固件检测算法中拍摄条件苛刻和防松线标记要求高等问题,提出应用于地铁车底检测机器人的紧固件松动检测算法。首先,在图像内紧固件完整且防松线无明显遮挡的情况下,采用改进的YOLOv5目标检测网络获取图片内每个紧固件目标,并将其分为3类;其次,使用DeepLabv3_plus图分割网络提取防松线图形轮廓,并将其转为二值图片;然后分别计算螺栓、螺母螺杆及金属管道接头这3类紧固件去噪后图片内两防松线的角度差、三防松线质心所接三角形内角极值以及多防松线与其最小外接矩形的面积占比,将对应计算值与拟定阈值对比,进行松动判定。最后,统计检测图片内实际松动情况,制定每类紧固件的二分类混淆矩阵,计算分析评价指标,与其他算法进行对比,并对实际地铁车底紧固件应用算法进行验证。结果显示改进YOLOv5目标检测网络模型MAP@0.5值由0.877提升至0.911,DeepLabv3_plus图像分割网络模型MIoU值高达0.950,松动判定检测所得658个紧固件MCA值分别为0.907、0.959以及0.888,表明算法可有效避免松动紧固件漏检。应用算法检测实际地铁车底各类紧固件,准确率均达90%以上。实验证明了目标检测算法改进的可行性、图像分割网络的适用性和松动判定算法的可靠性,为地铁车底紧固件松动检测智能运维工作提供重要技术支撑。