Mechanical Systems and Signal Processing最新文献

Shrinkage mamba relation network with out-of-distribution data augmentation for rotating machinery fault detection and localization under zero-faulty data 采用分布外数据增强的收缩曼巴关系网络，用于零故障数据下的旋转机械故障检测和定位

IF 8.4 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2024-11-17 DOI: 10.1016/j.ymssp.2024.112145

Zuoyi Chen, Hong-Zhong Huang, Zhongwei Deng, Jun Wu

{"title":"Shrinkage mamba relation network with out-of-distribution data augmentation for rotating machinery fault detection and localization under zero-faulty data","authors":"Zuoyi Chen, Hong-Zhong Huang, Zhongwei Deng, Jun Wu","doi":"10.1016/j.ymssp.2024.112145","DOIUrl":"https://doi.org/10.1016/j.ymssp.2024.112145","url":null,"abstract":"Data-driven fault detection (FD) or diagnosis methods are key technologies to ensure safe operation of rotating machinery. These methods rely on a requisite volume of fault data. However, acquiring fault data from rotating machinery is typically problematic and can be entirely unattainable. The critical challenge is to accurately detect and localize the fault states of rotating machinery under the absence of any fault data. Therefore, a newly shrinkage Mamba relation network (SMRN) with out-of-distribution data (OODD) augmentation is proposed for FD and localization in rotating machinery with zero-faulty data. Firstly, the corresponding sensors are arranged for the key detection locations on the rotating machinery. The data generator is referenced to generate OODD for the health data at each detection locations, assisting in mining of intrinsic state information from health data. Then, feature pairs are built in health data and OODD to reveal inter-state attribute relationships. Finally, the location of faults in rotating machinery is determined by evaluating the similarity between feature pairs at each detection location. The SMRN method effectiveness is verified by using self-built propulsion shaft system experiments and rolling bearing cases. The experimental results show the SMRN method can effectively detect and localize fault state of rotating machinery in multiple fault modes, compound fault scenarios, and variable operating conditions.","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"99 1","pages":""},"PeriodicalIF":8.4,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142665633","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Loosening state monitoring and identification of multi-bolted flange joints based on nonlinear wave energy transmission 基于非线性波能传输的多螺栓法兰连接松动状态监测与识别

IF 8.4 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2024-11-17 DOI: 10.1016/j.ymssp.2024.112114

Xu Chen, Wen Han, Zhousuo Zhang

{"title":"Loosening state monitoring and identification of multi-bolted flange joints based on nonlinear wave energy transmission","authors":"Xu Chen, Wen Han, Zhousuo Zhang","doi":"10.1016/j.ymssp.2024.112114","DOIUrl":"https://doi.org/10.1016/j.ymssp.2024.112114","url":null,"abstract":"Looseness detection of complex multi-bolted flange joints has long been an important problem to be focused on, especially for the scene of unknown multi-bolt loosening at the same time. In this study, a stable, efficient and robust guided wave recognition method for multi-bolt loosening is proposed for the first time by taking long-term monitoring data. This method studies the nonlinear characteristics of transmitted wave energy with bolt preload by formula. Then, a novel probability indicator reflecting the loosening position is proposed and a prior prediction model of bolt loosening degree is established. The prediction model is based on prior data fitting in a small number of working conditions, which has obvious advantages over deep learning. The simulation and experimental results based on flange pipes show that the proposed indicator can effectively determine the loosening positions of multiple bolts, and the prediction model also performs well in degree recognition. The proposed detection method has great potential in real-time monitoring applications by virtue of its high sensitivity to the loosening of multi-bolted joint structures.","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"13 1","pages":""},"PeriodicalIF":8.4,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142665634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel microwave-based dynamic measurement method for blade tip clearance through nonlinear I/Q imbalance correction 通过非线性 I/Q 不平衡校正实现叶尖间隙的新型微波动态测量方法

IF 8.4 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2024-11-17 DOI: 10.1016/j.ymssp.2024.112138

Saisai Chen, Tong Zhou, Wei Fan, Yuyong Xiong

{"title":"A novel microwave-based dynamic measurement method for blade tip clearance through nonlinear I/Q imbalance correction","authors":"Saisai Chen, Tong Zhou, Wei Fan, Yuyong Xiong","doi":"10.1016/j.ymssp.2024.112138","DOIUrl":"https://doi.org/10.1016/j.ymssp.2024.112138","url":null,"abstract":"Blade is the core working components of an aero-engine, and its blade tip clearance (BTC) exerts a direct influence on the efficiency and safety of the engine. Aiming to extract the BTC region from the echo signal, a novel microwave-based dynamic measurement system through nonlinear I/Q imbalance correction is proposed. Firstly, to achieve accurately correction of the sensor under non-linear amplitude attenuation, a new correction method is proposed to map the amplitude decay attenuation into the imbalance parameters distribution. Secondly, to effectively localize the extraction BTC region, the Amplitude-phase Half Wave Extraction (APH) is proposed, which utilizes squared I/Q amplitude information to determine the BTC position and extract clearance information. Finally, in comparison to existing methods, the proposed algorithm exhibits excellent performance in signal correction under non-linear amplitude attenuation and achieves high-accurate BTC extraction. Experimental studies show a mean absolute error below 2 μm and a repeatability mean error of 0.154 μm for a BTC variation of 0.17 mm.","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"1 1","pages":""},"PeriodicalIF":8.4,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142665632","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Generalized synchroextracting transform: Algorithm and applications 广义同步提取变换：算法与应用

IF 8.4 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2024-11-17 DOI: 10.1016/j.ymssp.2024.112116

Wenjie Bao, Songyong Liu, Zhen Liu, Fucai Li

{"title":"Generalized synchroextracting transform: Algorithm and applications","authors":"Wenjie Bao, Songyong Liu, Zhen Liu, Fucai Li","doi":"10.1016/j.ymssp.2024.112116","DOIUrl":"https://doi.org/10.1016/j.ymssp.2024.112116","url":null,"abstract":"Time-frequency (TF) rearrangement methods represented by synchrosqueezing transform (SST) and synchroextracting transform (SET) have recently been considered efficient tools for obtaining time-varying features of nonstationary signals. However, so far improving concentration and accuracy is still an open problem, especially for the signal with strongly time-varying instantaneous frequency (IF), due to the fact that they cannot achieve an accurate and generalized IF estimation. In order to address this problem, we introduce a new TF analysis method termed as generalized synchroextracting transform (GSET) by constructing a general signal model. Our first contribution in this study is proposing a new computational framework to derive the generalized explicit formula of <mml:math altimg=\"si10.svg\"><mml:msup><mml:mrow><mml:mi>N</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant=\"italic\">th</mml:mi></mml:mrow></mml:msup></mml:math>-order IF estimation, which can realize the programming of any order IF. By extracting the energy of the TF representation (TFR) on the estimated IF, a more concentrated and accurate TFR can be obtained. Our second contribution is giving a more accurate signal reconstruction method of the TFR from a new perspective. It solves the problem that the reconstruction method of the synchroextracting transform cannot be extended to the <mml:math altimg=\"si10.svg\"><mml:msup><mml:mrow><mml:mi>N</mml:mi></mml:mrow><mml:mrow><mml:mi mathvariant=\"italic\">th</mml:mi></mml:mrow></mml:msup></mml:math>-order. Numerical analysis of multicomponent simulated signal demonstrates that the GSET can effectively improve the TF readability of strongly time-varying signal and accurately reconstruct the signal from the TFR. Moreover, experiment and application results verify that the proposed method can be used for fault diagnosis of rotating machinery.","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"43 1","pages":""},"PeriodicalIF":8.4,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142665642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A two-stage correction method for UAV movement-induced errors in non-target computer vision-based displacement measurement 基于计算机视觉的非目标位移测量中无人机运动诱发误差的两阶段修正方法

IF 8.4 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2024-11-17 DOI: 10.1016/j.ymssp.2024.112131

Chi Zhang, Ziyue Lu, Xingtian Li, Yifeng Zhang, Xiaoyu Guo

{"title":"A two-stage correction method for UAV movement-induced errors in non-target computer vision-based displacement measurement","authors":"Chi Zhang, Ziyue Lu, Xingtian Li, Yifeng Zhang, Xiaoyu Guo","doi":"10.1016/j.ymssp.2024.112131","DOIUrl":"https://doi.org/10.1016/j.ymssp.2024.112131","url":null,"abstract":"Displacement plays a pivotal role in bridge assessment, but accurate displacement monitoring remains a challenging task. Unmanned Aerial Vehicles (UAVs) provide a cost-effective, time-efficient, and high maneuverability alternative to infrastructure monitoring, as they overcome the spatial limitations of the fixed camera and acquire the high-resolution image sequence. However, the measurement accuracy is often affected by the movement of the UAV. To address these constraints, this study proposed a computer vision-based nontarget displacement measurement method and a two-stage UAV movement correction method using fixed point and variational mode decomposition (VMD). Initially, the adaptive fusion of deep features and shallow features can efficiently encode the informative representation of the natural texture on the structural surface. Subsequently, the movement of the UAV is eliminated by stationary fixed points (Step Ⅰ) and VMD techniques (Step Ⅱ). Finally, the performance of the proposed methodology is verified with the field tests on a concrete wall and an arch bridge. Through mode decomposition and reconstruction, the measurement accuracy is greatly improved compared to the correction method only using fixed points, which proves the reliability and effectiveness of the proposed non-target displacement measurement method.","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"22 1","pages":""},"PeriodicalIF":8.4,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142665629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multifaceted vibration absorption of a rotating magnetic nonlinear energy sink 旋转磁场非线性能量汇的多方面振动吸收

IF 8.4 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2024-11-17 DOI: 10.1016/j.ymssp.2024.112122

Collin Treacy, Dalton Stein, David Chelidze

{"title":"Multifaceted vibration absorption of a rotating magnetic nonlinear energy sink","authors":"Collin Treacy, Dalton Stein, David Chelidze","doi":"10.1016/j.ymssp.2024.112122","DOIUrl":"https://doi.org/10.1016/j.ymssp.2024.112122","url":null,"abstract":"Hazardous failures of engineering structures can be prevented by implementing passive vibration absorbers. Although tuned mass dampers (TMD) are used most frequently in practice, nonlinear energy sinks (NES) offer a broader frequency performance due to their targeted energy transfer (TET) mechanisms. However, this behavior occurs only in a limited amplitude range. In this work, the vibration suppression of novel monostable and bistable magnetic rotary nonlinear energy sinks (MRNESs) are studied numerically and experimentally over a range of excitation magnitudes for impulse and harmonic excitation. The MRNESs are tuned to achieve hybrid TMD and NES-like behavior. The medley of in-well, cross-well, and rotational TET mechanisms responsible for their performance are related to their underlying Hamiltonian systems and lower boundaries of chaos. For impulse excitation, rotational, subharmonic, and nonlinear beat responses lead to efficient energy dissipation. For harmonic excitation, the MRNEs’ frequency responses can resemble a TMDs’ or exhibit chaotic-like cross-well and rotational strongly modulated responses. Consequently, MRNESs can overcome the shortcomings of many NESs, which are inefficient at low excitation magnitudes, while also outperforming linear TMDs when the systems’ parameters are detuned at large excitation magnitudes, for both impulse and harmonic excitation. The MRNESs’ TET mechanisms and efficient performance over a broad range of excitation magnitudes were validated experimentally for both types of excitation. The MRNES may be more viable for practical use than other hybrid NESs since it is compact, highly customizable, and does not rely on impacts or complicated spring arrangements for its non-linearity.","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"106 1","pages":""},"PeriodicalIF":8.4,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142665631","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Damage identification in plate-like structures using frequency-coupled [formula omitted]-based sparse estimation 利用基于频率耦合[公式省略]的稀疏估计识别板状结构中的损伤

IF 8.4 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2024-11-17 DOI: 10.1016/j.ymssp.2024.112084

Nathan Dwek, Vasileios Dimopoulos, Dennis Janssens, Matteo Kirchner, Elke Deckers, Frank Naets

{"title":"Damage identification in plate-like structures using frequency-coupled [formula omitted]-based sparse estimation","authors":"Nathan Dwek, Vasileios Dimopoulos, Dennis Janssens, Matteo Kirchner, Elke Deckers, Frank Naets","doi":"10.1016/j.ymssp.2024.112084","DOIUrl":"https://doi.org/10.1016/j.ymssp.2024.112084","url":null,"abstract":"This article proposes a practical and effective damage identification approach for plate-like structures. This approach measures the back scattering caused by damage, and decomposes it into individual contributions from each defect, using the responses of the healthy structure as a dictionary. A data-driven model is used, which circumvents the challenge of numerically simulating the effect of damage, yet does not require training data from known-damaged structures. The decomposition itself is performed using sparsity-promoting optimization, which reduces the number of required measurements and streamlines the inspection procedure. A novel frequency-coupled method is proposed to obtain the desired spatial sparsity of the estimated damage, which results in improved accuracy compared to the previously proposed frequency-decoupled method. Damage identification is demonstrated on a <mml:math altimg=\"si3.svg\" display=\"inline\"><mml:mrow><mml:mtext>600</mml:mtext><mml:mspace width=\"0.16667em\"></mml:mspace><mml:mtext>mm</mml:mtext><mml:mo linebreak=\"goodbreak\" linebreakstyle=\"after\">×</mml:mo><mml:mtext>600</mml:mtext><mml:mspace width=\"0.16667em\"></mml:mspace><mml:mtext>mm</mml:mtext></mml:mrow></mml:math> composite plate, using a single accelerometer and 7 impact hammer hits. The performance is evaluated on 6 damage scenarios, for 7 accelerometer positions, and for SNRs ranging from 30 to 0<ce:hsp sp=\"0.16667\"></ce:hsp>dB. Detection and localization are shown to be excellent up to 5 defects and down to 15<ce:hsp sp=\"0.16667\"></ce:hsp>dB SNR, and to remain robust and predictable outside of that range. These results are compared to reference methods and a significant improvement is observed.","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"249 1","pages":""},"PeriodicalIF":8.4,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142665637","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Outlier-resistant guided wave dispersion curve recovery and measurement placement optimization base on multitask complex hierarchical sparse Bayesian learning 基于多任务复杂分层稀疏贝叶斯学习的抗离群导波频散曲线恢复与测量位置优化

IF 8.4 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2024-11-17 DOI: 10.1016/j.ymssp.2024.112137

Shicheng Xue, Wensong Zhou, Yong Huang, Lam Heung Fai, Hui Li

{"title":"Outlier-resistant guided wave dispersion curve recovery and measurement placement optimization base on multitask complex hierarchical sparse Bayesian learning","authors":"Shicheng Xue, Wensong Zhou, Yong Huang, Lam Heung Fai, Hui Li","doi":"10.1016/j.ymssp.2024.112137","DOIUrl":"https://doi.org/10.1016/j.ymssp.2024.112137","url":null,"abstract":"Due to extensive detection range and high sensitivity to defects, ultrasonic Lamb waves are extensively studied in the fields of Nondestructive Testing and Structural Health Monitoring. In scenarios where the material parameters or geometric parameters of the waveguide are unknown, the dispersion relation of the guided wave cannot be calculated by the forward model. Consequently, it becomes imperative to extract wave propagation characteristics of Lamb wave from the acquired Lamb wave data. This paper presents a multitask complex hierarchical sparse Bayesian learning (MuCHSBL) method which is aimed at enhancing the efficacy of the dispersion relation solution by considering the continuity of the recovered dispersion curve in the frequency-wavenumber domain. Furthermore, the posterior distributions quantified by MuCHSBL are employed to optimize the placement of measurement points. Numerical and experimental studies are conducted to verify the effectiveness of the proposed method. Comparison analysis with the conventional approach demonstrates the significant enhancement in accuracy of recovering dispersion curves by the proposed method.","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"128 1","pages":""},"PeriodicalIF":8.4,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142665630","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A new dynamic model for cylindrical roller bearings with flexible rollers and bearing sliding investigation 带柔性滚子的圆柱滚子轴承新动态模型及轴承滑动研究

IF 8.4 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2024-11-14 DOI: 10.1016/j.ymssp.2024.112133

Ke Yan, Shuaijun Ma, Bin Fang, Fei Chen, Jun Hong, Pan Zhang

{"title":"A new dynamic model for cylindrical roller bearings with flexible rollers and bearing sliding investigation","authors":"Ke Yan, Shuaijun Ma, Bin Fang, Fei Chen, Jun Hong, Pan Zhang","doi":"10.1016/j.ymssp.2024.112133","DOIUrl":"https://doi.org/10.1016/j.ymssp.2024.112133","url":null,"abstract":"Cylindrical roller bearings are inevitably impacted by external moments or mounting error, which leads to uneven load distribution on the rollers and triggers deformation. Current methods are insufficient to simulate the deformation while ensuring solution accuracy. To address this, a new slicing approach is innovatively proposed in the paper, where springs are added between the neighboring slices to simulate the elastic deformation of the roller under external loads. Compared with the existing methods, the method presented in this paper has the best agreement with the finite element model. On this basis, a dynamic model for cylindrical roller bearing with roller deformation is further developed and verified experimentally. Finally, the sliding behavior inside the bearing under three typical conditions is investigated. A rich spectrum of frequencies emerges in the bearing contact load between the roller and raceway because of the roller deformation. These are all integer multiples of the roller passage frequency. An interesting phenomenon is observed that the sliding velocity is strongly influenced by the orbital speed of the roller compared to its rotational speed. The modeling and analysis in this paper provide new directions to the future work.","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"46 1","pages":""},"PeriodicalIF":8.4,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142665636","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Wheel-rail force inversion via transfer learning-based residual LSTM neural network with temporal pattern attention mechanism 通过基于迁移学习的残差 LSTM 神经网络与时间模式关注机制实现轮轨力反转

IF 8.4 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2024-11-14 DOI: 10.1016/j.ymssp.2024.112135

Taoning Zhu, Yu Ren, Huailong Shi, Yunguang Ye, Piji Feng, Zhenhua Su, Chunxing Yao, Guangtong Ma

{"title":"Wheel-rail force inversion via transfer learning-based residual LSTM neural network with temporal pattern attention mechanism","authors":"Taoning Zhu, Yu Ren, Huailong Shi, Yunguang Ye, Piji Feng, Zhenhua Su, Chunxing Yao, Guangtong Ma","doi":"10.1016/j.ymssp.2024.112135","DOIUrl":"https://doi.org/10.1016/j.ymssp.2024.112135","url":null,"abstract":"As urbanization progresses, metropolitan transit vehicles are encountering a growing frequency of curved pathways, which presents challenges pertaining to both the safety of the vehicles and the comfort of the passengers. There is no doubt that reliable acquisition of wheel-rail force is critical, since it has great significance for the safety and stability of vehicle operation. However, conventional wheel-rail force measurement methods are costly and difficult to use on in-service vehicles. A data-driven approach to inverting the wheel-rail force will overcome the above problems. In this work, a transfer learning-based residual long short-term memory neural network with temporal pattern attention mechanism (TPA-ResLSTM) is proposed to realize real-time monitoring of wheel-rail force, even in scenarios where the dataset is deficient in sufficient features. Initially, a learnable wheel-rail force inversion neural network model is developed based on the physical relationship that exists between the wheel-rail force and acceleration. Subsequently, a dynamic model for a B-type metro vehicle is utilized to simulate various scenarios, serving as a virtual source to provide data for the neural network. Afterward, the performance of the model is synthetically validated by the ablation study and field experimental data. Finally, the deep learning model is further improved by the transfer learning network, whose performance is comprehensively evaluated using limited data. The results show that the inversion model still has remarkable accuracy, in which the coefficient of determination is more than 0.9, under the case of limited training data. The proposed methodology diminishes the data requirements for the network while facilitating real-time monitoring and feedback regarding wheel-rail forces, thereby enhancing the realism of operational safety assessments for trains.","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"50 1","pages":""},"PeriodicalIF":8.4,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142665635","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0