Mechanical Systems and Signal Processing最新文献

{"title":"Identification of non-conforming temporal patterns in mixed shock and vibration data with the Information Impulse Function","authors":"Angela Montoya ,&nbsp;Daniel Roettgen ,&nbsp;Samuel Parker ,&nbsp;Benjamin Moldenhauer ,&nbsp;Fernando Moreu","doi":"10.1016/j.ymssp.2025.112946","DOIUrl":"10.1016/j.ymssp.2025.112946","url":null,"abstract":"<div><div>The Information Impulse Function (IIF) is an analysis technique that operates on time<em>–</em>frequency representations of a signal to detect non-conforming, transient patterns. Transient patterns can signify events of scientific and engineering interest but may be difficult to confidently identify against non-Gaussian backgrounds. IIF analysis highlights non-conforming patterns and diminishes conforming patterns, providing a high confidence discriminator between potential events of interest and the background. This is accomplished by first calculating the relative participation of the temporal principal shape vectors as a function of signal compression. The integral of the resulting surface represents both signal complexity and uniqueness per instance in time. To demonstrate the IIF operator, acceleration response data was collected on a complex aluminum structure excited by stationary random vibration while simultaneously impacted by three different modal hammers. The force output of each hammer was recorded to corroborate the impacts in time with IIF results. Detection efficacy of the IIF is benchmarked using the scale-averaged wavelet power and the local Hölder exponent representing changes in signal energy and continuity, respectively. Results show that IIF analysis is highly effective at detecting impacts. General behavior of the IIF is discussed with specific examples.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 112946"},"PeriodicalIF":7.9,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144255022","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}

{"title":"Physics-preserved graph learning of differential equations for structural dynamics","authors":"Yu Wang ,&nbsp;Jinzhao Li ,&nbsp;Xuan Kong ,&nbsp;Weiwei He ,&nbsp;Lu Deng ,&nbsp;Liangrui Pan ,&nbsp;Jiaqiang Peng","doi":"10.1016/j.ymssp.2025.112956","DOIUrl":"10.1016/j.ymssp.2025.112956","url":null,"abstract":"<div><div>Spatio-temporal differential equations are fundamental to understanding the world, describing the dynamic behavior of a structure/system under external stimuli. The equations are typically solved with numerical methods, such as the finite element method, which is computationally inefficient for complex structures, especially under multi-load case analysis requiring repeated calls to slow numerical solvers. Meanwhile, the emerging data-driven deep learning approaches heavily rely on extensive labeled datasets. Here we propose a physics-preserved neural network that seamlessly integrates physical knowledge towards accurate and rapid computation of dynamic characteristics for complex systems without relying on labeled data. A graph convolutional network is created for modal computation in space domain, where physical laws and constraints are inherently encoded within the network architecture (termed ‘hard-embedding’). A physics-informed neural network is then adopted for the dynamic response computation in time domain. This hard-embedding approach remarkably improves computational accuracy compared to the state-of-the-art soft-constraint methods based on loss functions. The proposed model also realizes end-to-end generalization computations under different loading and initial conditions, thereby improving computational speed by hundreds of times compared to the finite element method. This characteristic renders our approach a promising alternative for realizing real-time structural dynamic computations.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 112956"},"PeriodicalIF":7.9,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144255023","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}

{"title":"Complex dynamics and targeted energy transfer of double-beam nonlinear energy sink with magnetic interactions","authors":"Hongxiang Hu ,&nbsp;Ruining Huang ,&nbsp;Haoran Qin ,&nbsp;Zhongwen Zhang ,&nbsp;Zhao-dong Xu","doi":"10.1016/j.ymssp.2025.112942","DOIUrl":"10.1016/j.ymssp.2025.112942","url":null,"abstract":"<div><div>Nonlinear energy sinks (NESs) have garnered sustained attention in recent years as highly efficient broadband vibration suppression devices. However, achieving optimal vibration control within limited space through the design of compact NESs configurations remains a key issue hindering its engineering applications. To address this, the present paper proposes a novel compact double-beam nonlinear energy sink device with magnetic interactions. The device constructs a two-degree-of-freedom vibration unit by cutting an internal sub-beam within the main beam, and integrates magnet mass blocks at the ends of the inner and outer beams. It utilizes the magnetic repulsion effect to develop an adjustable nonlinear stiffness system. The dynamic model of the 3DOF coupled system is formulated using Lagrangian method, with approximate analytical solutions derived through the harmonic balance method. The study conducts a parameter sensitivity analysis to explore the impact of NES parameters on the frequency response characteristics and vibration suppression performance of the primary system. Finally, the transient vibration suppression performance of various damping devices under various initial energy excitations is assessed based on critical performance indicators, such as the energy dissipation ratio. The findings show that the proposed NES achieves remarkable energy dissipation efficiency and robust performance across a wide frequency excitation range. This research presents a novel solution and theoretical foundation for broadband vibration control in engineering structures.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 112942"},"PeriodicalIF":7.9,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144255057","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}

{"title":"Data imbalance fault diagnosis method based on an ensemble multi-scale convolutional attention network","authors":"Jialong He ,&nbsp;Wentao Huang ,&nbsp;Yan Liu ,&nbsp;Chenhui Qian ,&nbsp;Chi Ma ,&nbsp;Wanfu Gao ,&nbsp;Xingze Jin","doi":"10.1016/j.ymssp.2025.112934","DOIUrl":"10.1016/j.ymssp.2025.112934","url":null,"abstract":"<div><div>In recent years, mechanical intelligence fault diagnosis methods based on deep learning are in the ascendant. However, the problem of noise interference and data imbalance is often faced in practical applications, so it is still a challenge to achieve high precision and reliable fault diagnosis. To solve the problem that traditional convolutional neural networks have poor anti-noise performance and are easy to ignore the minority class samples, this paper proposes a mechanical intelligence fault diagnosis method based on an ensemble multi-scale convolutional attention network (EMCAN). First, a multi-scale convolutional attention network is constructed as the base classifier, which is mainly composed of the multi-scale convolutional denoising module (MCDM) and the cooperative attention module (CAM). MCDM suppresses high-frequency noise and extracts multi-scale discriminant features. Differentiated CAMs adaptively focus on important features and increase the diversity of base classifiers. Second, an ensemble strategy based on improved weighted voting is proposed, and balanced training subsets are constructed for each base classifier by sampling with replacement to improve the robustness and generalization of the ensemble model. The proposed EMCAN is validated on a bearing open dataset and a gearbox experimental dataset. Compared with the state-of-the-art comparison method, the Gmean of the proposed EMCAN is 4.60% and 12.11% higher under the most imbalanced conditions, respectively, which proves the validity and superiority of EMCAN.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 112934"},"PeriodicalIF":7.9,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144255025","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}

{"title":"Bayesian FFT modal identification for multi-setup experimental modal analysis","authors":"Peixiang Wang ,&nbsp;Binbin Li","doi":"10.1016/j.ymssp.2025.112931","DOIUrl":"10.1016/j.ymssp.2025.112931","url":null,"abstract":"<div><div>In full-scale forced vibration tests, the demand often arises to capture high-spatial-resolution and high-precision mode shapes with a limited number of sensors and shakers. Such detailed mode shapes are valuable for applications such as damage localization and high-dimensional finite element model updating. Multi-setup experimental modal analysis (EMA) addresses this challenge by roving sensors and shakers across multiple setups. To enable fast and accurate multi-setup EMA, this paper develops a Bayesian modal identification strategy by extending the existing single-setup algorithm. Specifically, a frequency-domain probabilistic model is first formulated using multiple sets of structural multiple-input, multiple-output (MIMO) vibration data. A constrained Laplace method is then employed for Bayesian posterior approximation, providing the maximum a posteriori estimate of modal parameters along with a posterior covariance matrix (PCM) for uncertainty quantification. Utilizing complex matrix calculus, analytical expressions are derived for parameter updates in the coordinate descent optimization, as well as for PCM computation, enhancing both coding simplicity and computational efficiency. The proposed algorithm is intensively validated with synthetic and field data from bridge and building structures. It demonstrates that the proposed method yields highly consistent results compared to scenarios with adequate test equipment. The resulting high-fidelity MIMO model enables structural response prediction under future loading conditions and supports condition assessment.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 112931"},"PeriodicalIF":7.9,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144255024","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}

{"title":"Transient mixed elastohydrodynamic performance of journal bearings in wind turbine gearbox during start-up considering multi-component elastic deformations","authors":"Hao Li ,&nbsp;Caichao Zhu ,&nbsp;Jianjun Tan ,&nbsp;Wenjun Fei ,&nbsp;Zhangdong Sun ,&nbsp;Wei Ye","doi":"10.1016/j.ymssp.2025.112975","DOIUrl":"10.1016/j.ymssp.2025.112975","url":null,"abstract":"<div><div>To improve the torque density of wind turbine gearboxes (WTG), planet gear roller bearings are being replaced by planet gear journal bearings (PGJBs). However, the low rotational speeds, wide load variations, and varying operating conditions during startup significant challenges to the performance of PGJB. In this paper, a transient mixed elastohydrodynamic (EHD) model of PGJBs during start-up is proposed, which comprehensively considers multi-component elastic deformations, misalignment, and intermediate oil return groove (IORG) characteristics. Specifically, the dynamic rotational speed, meshing force, and moment of the planet gear, extracted from the wind turbine drivetrain model, serve as inputs for the PGJB. The influence mechanism of the IORG, radial clearance, and modification parameters on the transient EHD performance of PGJB is investigated, followed by experimental validation. Results indicate that the IORG effectively mitigates misalignment in PGJBs while simultaneously inducing earlier contact and significantly increasing both film and contact pressures. This leads to a multi-regime lubrication distribution at the PGJB interface and further influences its contact characteristics. During startup, the primary load-bearing regions maintain hydrodynamic lubrication, whereas the IORG-affected zones transition from mixed lubrication during mid-startup to boundary lubrication under rated operating conditions. Contact occurs in the edge region of the PGJB when the modification depth is less than 30 μm, and shifts to the IORG region at greater depths.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 112975"},"PeriodicalIF":7.9,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144255021","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}

{"title":"Deep optical flow to identify structural vibration modal parameters","authors":"Rongliang Yang,&nbsp;Tao Liu,&nbsp;Sen Wang,&nbsp;Zhenya Wang","doi":"10.1016/j.ymssp.2025.112897","DOIUrl":"10.1016/j.ymssp.2025.112897","url":null,"abstract":"<div><div>Accurate measurement of vibration signals is an important prerequisite for accurate analysis of structural vibration modes. Optical flow estimation is suitable for non-contact full-field signal measurement. However, optical flow datasets are usually synthesized by computers and lack practical application in engineering structures. In addition, the algorithm has deficiencies in processing sparse textures and edge features, and the sub-pixel estimation accuracy needs to be improved. Therefore, this paper constructs real structural vibration optical flow datasets to provide basic data support for structural health monitoring. For the deep optical flow algorithm, an adaptive Gabor filter component is proposed. By dynamically adjusting the direction and scale, the edge and texture features in the optical flow field are enhanced, and the fusion with the convolutional neural network further models the relationship between global and local features. At the same time, the local similarity matching mechanism is combined to improve the accuracy of optical flow estimation and the ability to identify local details. In addition, the smoothing term and the loss function of the learnable parameters are added to improve the physical consistency and prediction accuracy of the model, and effectively suppress the generation of noise interference and discontinuous optical flow. The results of qualitative and quantitative scene and ablation experiments show that the proposed deep optical flow algorithm is superior to the existing methods in terms of optical flow estimation accuracy, structural edge preservation ability and low light adaptability, and shows excellent performance in sparse texture scenes and low signal-to-noise ratio environments. In addition, by performing modal frequency error evaluation and energy ratio analysis on the identification results, the effectiveness and stability of the proposed method in modal main frequency identification and high-order mode extraction are demonstrated, providing a high-precision solution for non-contact modal analysis.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 112897"},"PeriodicalIF":7.9,"publicationDate":"2025-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144255020","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}

{"title":"Online assessment of degradation status for metro wheel with robust unsupervised tensor domain adaptation","authors":"Hao Liu ,&nbsp;Wentao Mao ,&nbsp;Na Wang ,&nbsp;Linlin Kou","doi":"10.1016/j.ymssp.2025.112894","DOIUrl":"10.1016/j.ymssp.2025.112894","url":null,"abstract":"<div><div>With on-board collected vibration signals, machine learning-assisted metro wheel’s degradation status assessment has received sustained attention in most recent years. Despite promising advances, the acquired vibration signals, heavily interfered by irregular noise due to various factors like road condition, vehicle load and uneven tread, etc., will block the deployment of current assessment methods in open environment. Moreover, existing methods are merely capable of realizing status assessment using offline data within a turning repair cycle, but fail to achieve online assessment that is crucial to actual maintenance. This paper incorporates the concept of transfer learning into metro wheel health management. We identify two major challenges in evaluating wheel degradation status: (1) imprecise representation of degradation characteristics with real-world signals, and (2) lack of common degradation trajectory on the same metro line. To address these challenges, this paper proposes an online assessment approach of wheel degradation status based on unsupervised deep transfer learning, including a robust unsupervised tensor domain adaptation network (RUTDAN) for cross-wheel degradation feature extraction and an online early warning mechanism based on the common health indicator of different wheels. Extensive experiments are conducted with the real-world monitoring data collected from Metro Line 6 of Beijing Subway. The degradation status of target wheel is evaluated with online sequential data, while the results precisely match the actual maintenance records in terms of wheel diameter value.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 112894"},"PeriodicalIF":7.9,"publicationDate":"2025-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144242352","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}

{"title":"A novel sparse-aware contrastive learning network with adaptive gating neurons for extreme class imbalance diagnosis scenarios","authors":"Panpan Guo ,&nbsp;Weiguo Huang ,&nbsp;Chuancang Ding ,&nbsp;Juanjuan Shi ,&nbsp;Zhongkui Zhu","doi":"10.1016/j.ymssp.2025.112895","DOIUrl":"10.1016/j.ymssp.2025.112895","url":null,"abstract":"<div><div>The intelligent diagnosis model can exhibit excellent diagnostic performance with sufficient training samples and under ideal conditions. However, in engineering environments, it is difficult to obtain balanced fault samples from equipment, and the interference from noise components in signals makes it challenging for deep learning models to extract fault features, posing significant challenges for intelligent diagnosis. To address these issues, this paper proposes a novel sparse perception contrastive learning network with adaptive gating neurons for extreme class imbalance diagnosis scenarios. Specifically, we first propose an adaptive gating neurons residual network, derive and establish a mathematical relationship between the adaptive gating neuron and learnable weighted autocorrelation functions, demonstrating the model’s ability to extract relevant features from vibration signals and perform adaptive noise reduction. Building upon this, we propose a Sparse Perception Cross-entropy Loss (SPCL) function, which focuses on tail-class fault samples that are difficult to cluster based on class sparsity in the feature space. Furthermore, to further enhance the diagnostic performance of the contrastive learning model in class-imbalanced bearing faults, we propose a Center Contrastive Loss (CCL) function. CCL calculates the centers of each class using classifier weights optimized by SPCL function, ensuring that all fault classes are represented in each mini-batch during learning, thereby enabling effective contrastive learning. The efficacy of the introduced methodologies is confirmed through experimental outcomes obtained from both publicly accessible and proprietary datasets. Experimental results demonstrate that our proposed method significantly outperforms other methods in the intelligent diagnosis of bearing faults in scenarios with extreme class imbalance.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"235 ","pages":"Article 112895"},"PeriodicalIF":7.9,"publicationDate":"2025-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144241140","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}

{"title":"On the tyre–road friction coefficient fusion identification framework utilising computer vision-based and vehicle dynamics-based road surface information","authors":"Yuefeng Huang,&nbsp;Youpeng Zhang ,&nbsp;Xiangyu Wang,&nbsp;Heng Wei,&nbsp;Kai Deng,&nbsp;Liang Li,&nbsp;Jian Song","doi":"10.1016/j.ymssp.2025.112821","DOIUrl":"10.1016/j.ymssp.2025.112821","url":null,"abstract":"<div><div>The tyre–road friction coefficient (TFC) is a critical structural parameter in the vehicle–road closed-loop system for intelligent safety and automated driving functions. However, it is vulnerable to the variations of the environment conditions and road surface states, thereby difficult to be determined in advance. This article proposes the TFC fusion identification framework, which primarily comprises the computer vision-based road surface image classification module, the vehicle dynamics-based TFC observation module, and the reliability factor-based road surface information fusion module. First, full-frame images are transformed into block images of eight types of road surface and a special category of interfering objects. These block images are then fed into the road surface block image classification network under the GhostNetV2-based contrastive learning framework. The outputs of the proposed classification network are integrated using the evidence reasoning-based road surface block information fusion strategy to obtain the classification results for the left and right sides, which are subsequently converted into computer vision-based TFCs through mapping and spatio-temporal synchronisation. Meanwhile, the primary–secondary adaptive unscented Kalman filter (UKF) is constructed based on tyre force observation and the modified Dugoff model to acquire vehicle dynamics-based TFCs. The reliability factor-based TFC fusion strategy is then proposed to integrate the two types of TFCs. A series of offline and field tests demonstrate that the proposed classification network exhibits high precision in the block image classification task compared to those under GhostNetV2, MobileNetV3, and ShuffleNetV2 with augmentation techniques, and the proposed identification framework presents rapid responsiveness and strong robustness when compared with TFC observation algorithms based solely on the UKF and the primary–secondary adaptive UKF.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"235 ","pages":"Article 112821"},"PeriodicalIF":7.9,"publicationDate":"2025-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144241141","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}