To achieve better anti-vibration performance in a low frequency region and expand the range of vibration isolation,a bilateral supported bio-inspired anti-vibration(BBAV)structure composed of purely linear elements is...To achieve better anti-vibration performance in a low frequency region and expand the range of vibration isolation,a bilateral supported bio-inspired anti-vibration(BBAV)structure composed of purely linear elements is proposed,inspired by the motion form of bird legs and the nonlinear extension and compression of muscles and tendons.The kinematic relations and nonlinear dynamic model considering vertical and rotational vibrations are established.The loading capacity and equivalent stiffness are investigated with key parameters.The amplitude-frequency characteristics and force transmissibility are used to evaluate the stability and anti-vibration performance with the effects of the excitation amplitude,rod length,installation angle,and spring stiffness.The results show that the loading requirements and resonant characteristics of the BBAV structure are adjustable,and superior vibration isolation performance can be achieved readily by tuning the parameters.The X-shaped vibration structure is sensitive to the spring stiffness,which exhibits a wider vibration isolation bandwidth with smaller spring stiffness.Besides,depending on the parameters,the nonlinear behavior of the BBAV system can be interconverted between the softening type and the hardening type.The theoretical analysis in this study demonstrates the advantages and effectiveness of the vibration isolation structure.展开更多
In the harsh environment,the structural health of the anti-vibration hammer,which suffers from the coupled effects of corrosion and fatigue damage,is significantly reduced.As part of the conductor structure,the anti-v...In the harsh environment,the structural health of the anti-vibration hammer,which suffers from the coupled effects of corrosion and fatigue damage,is significantly reduced.As part of the conductor structure,the anti-vibration hammer is rigidly attached to the conductor,effectively suppressing conductor vibration.The conductor’s breeze vibration law and natural modal frequency are altered damage to the anti-vibration hammer structure.Through built a vibration experiment platform to simulate multiple faults such as anti-vibration hammer head drop off and position slippage,which to obtained the vibration acceleration signal of the conductor.The acceleration vibration signal is processed and analyzed in the time and frequency domains.The results are used to derive the breeze vibration law of the conductor under multiple faults and propose an anti-vibration hammer damage online monitoring technology.The results show that the vibration acceleration value and vibration intensity of the conductor are significantly increased after the anti-vibration hammer damage.The natural frequency increases for each order,with an absolute change ranging from 0.15 to 6.49 Hz.The anti-vibration hammer slipped due to a loose connection,the 1st natural frequency increases from 8.18 to 16.62 Hz.Therefore,in engineering applications,there can be no contact to determine the anti-vibration hammer damage situation by monitoring the modal natural frequency of the conductor.This is even a tiny damage that cannot be seen.This method will prevent the further expansion of the damage that can cause accidents.展开更多
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 inspect...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.展开更多
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 r...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.展开更多
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 condu...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.展开更多
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 ...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.展开更多
无砟轨道扣件实现了钢轨与混凝土道床、下部基础的连接。扣件受力除轮轨作用下的正常服役工作状态外,还包括桥梁地段梁端变形诱发扣件受力状态变化。为保障桥梁地段无砟轨道扣件正常服役状态,本文以32 m T型梁为载体,通过分析梁端转角...无砟轨道扣件实现了钢轨与混凝土道床、下部基础的连接。扣件受力除轮轨作用下的正常服役工作状态外,还包括桥梁地段梁端变形诱发扣件受力状态变化。为保障桥梁地段无砟轨道扣件正常服役状态,本文以32 m T型梁为载体,通过分析梁端转角、垂向位移、预留梁缝及扣件安装位置等变量,研究了T型梁梁端变形及接口构造对无砟轨道扣件受力的影响。结果表明:梁端变形引起的扣件上拔力与梁端转角、垂向位移及转角位移组合变形线性相关,提出了不同构造变量与扣件上拔力的线性表达式。研究成果可为客货共线桥梁铺设无砟轨道梁端变形控制及构造设计提供参考。展开更多
针对常规深度学习网络规模大、对现场设备硬件要求高且人工标注位置数据复杂费时等问题,提出了一种语义数据标注的轻量化轨道扣件故障检测方法。该方法仅对训练数据做语义标注,改进轻量化Transformer模型,嵌入梯度加权类激活映射(gradie...针对常规深度学习网络规模大、对现场设备硬件要求高且人工标注位置数据复杂费时等问题,提出了一种语义数据标注的轻量化轨道扣件故障检测方法。该方法仅对训练数据做语义标注,改进轻量化Transformer模型,嵌入梯度加权类激活映射(gradient-weighted class activation mapping,简称Grad-CAM)模块对模型输出的特征图权重作映射处理,可将模型对轨道扣件检测效果可视化。将获得的激活图进行二值化定位检测目标位置,实验结果表明,在真实铁路环境下,改进的轻量化轨道扣件模型的准确率为94.31%。展开更多
基金Project supported by the National Natural Science Foundation of China(No.52275091)the Fundamental Research Funds for the Central Universities of China(No.N2103008)the Natural Science Foundation of Liaoning Province of China(No.2020-MS-125)。
文摘To achieve better anti-vibration performance in a low frequency region and expand the range of vibration isolation,a bilateral supported bio-inspired anti-vibration(BBAV)structure composed of purely linear elements is proposed,inspired by the motion form of bird legs and the nonlinear extension and compression of muscles and tendons.The kinematic relations and nonlinear dynamic model considering vertical and rotational vibrations are established.The loading capacity and equivalent stiffness are investigated with key parameters.The amplitude-frequency characteristics and force transmissibility are used to evaluate the stability and anti-vibration performance with the effects of the excitation amplitude,rod length,installation angle,and spring stiffness.The results show that the loading requirements and resonant characteristics of the BBAV structure are adjustable,and superior vibration isolation performance can be achieved readily by tuning the parameters.The X-shaped vibration structure is sensitive to the spring stiffness,which exhibits a wider vibration isolation bandwidth with smaller spring stiffness.Besides,depending on the parameters,the nonlinear behavior of the BBAV system can be interconverted between the softening type and the hardening type.The theoretical analysis in this study demonstrates the advantages and effectiveness of the vibration isolation structure.
基金supported by the National Natural Science Foundation of China(No.52007138)the Natural Science Basis Research Plan in Shaanxi Province of China(No.2022JQ-568)the Key Research and Development Program of Shaanxi Province(No.2023-YBGY-069).
文摘In the harsh environment,the structural health of the anti-vibration hammer,which suffers from the coupled effects of corrosion and fatigue damage,is significantly reduced.As part of the conductor structure,the anti-vibration hammer is rigidly attached to the conductor,effectively suppressing conductor vibration.The conductor’s breeze vibration law and natural modal frequency are altered damage to the anti-vibration hammer structure.Through built a vibration experiment platform to simulate multiple faults such as anti-vibration hammer head drop off and position slippage,which to obtained the vibration acceleration signal of the conductor.The acceleration vibration signal is processed and analyzed in the time and frequency domains.The results are used to derive the breeze vibration law of the conductor under multiple faults and propose an anti-vibration hammer damage online monitoring technology.The results show that the vibration acceleration value and vibration intensity of the conductor are significantly increased after the anti-vibration hammer damage.The natural frequency increases for each order,with an absolute change ranging from 0.15 to 6.49 Hz.The anti-vibration hammer slipped due to a loose connection,the 1st natural frequency increases from 8.18 to 16.62 Hz.Therefore,in engineering applications,there can be no contact to determine the anti-vibration hammer damage situation by monitoring the modal natural frequency of the conductor.This is even a tiny damage that cannot be seen.This method will prevent the further expansion of the damage that can cause accidents.
基金supported in part by the National Natural Science Foundation of China (Grant Nos.51975347 and 51907117)in part by the Shanghai Science and Technology Program (Grant No.22010501600).
文摘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.
文摘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.
文摘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.
基金supported by the National Key R&D Program of China(Grant 2021YFF0501102)National Natural Science Foundation of China(Grant U1934219)+1 种基金National Science Fund for Excellent Young Scholars(Grant 52022010)National Natural Science Foundation of China(Grant 52202392,Grant 62120106011).
文摘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.
文摘无砟轨道扣件实现了钢轨与混凝土道床、下部基础的连接。扣件受力除轮轨作用下的正常服役工作状态外,还包括桥梁地段梁端变形诱发扣件受力状态变化。为保障桥梁地段无砟轨道扣件正常服役状态,本文以32 m T型梁为载体,通过分析梁端转角、垂向位移、预留梁缝及扣件安装位置等变量,研究了T型梁梁端变形及接口构造对无砟轨道扣件受力的影响。结果表明:梁端变形引起的扣件上拔力与梁端转角、垂向位移及转角位移组合变形线性相关,提出了不同构造变量与扣件上拔力的线性表达式。研究成果可为客货共线桥梁铺设无砟轨道梁端变形控制及构造设计提供参考。
文摘针对常规深度学习网络规模大、对现场设备硬件要求高且人工标注位置数据复杂费时等问题,提出了一种语义数据标注的轻量化轨道扣件故障检测方法。该方法仅对训练数据做语义标注,改进轻量化Transformer模型,嵌入梯度加权类激活映射(gradient-weighted class activation mapping,简称Grad-CAM)模块对模型输出的特征图权重作映射处理,可将模型对轨道扣件检测效果可视化。将获得的激活图进行二值化定位检测目标位置,实验结果表明,在真实铁路环境下,改进的轻量化轨道扣件模型的准确率为94.31%。