Pub Date : 2026-01-16DOI: 10.1016/j.ymssp.2026.113881
Sheng-Wang Zhang , Said Quqa , Antonio Palermo , Alessandro Marzani , Zhao-Hui Lu
Damage identification methods based on traffic-induced vibration data have gained significant attention in structural health monitoring of bridges, driven by the need for cost-effective sensing solutions. Recent studies have demonstrated that bridge curvature profiles can be identified from sparse acceleration measurements collected during vehicle passages using standard accelerometers. However, existing approaches for estimating curvature from acceleration data often struggle to suppress dynamic effects induced by moving vehicles. These methods typically rely on low-pass filters with a rigid cutoff threshold, which can compromise accuracy, especially during high-speed vehicle passages. To overcome this limitation, this study introduces a novel approach based on the continuous wavelet transform to isolate the quasi-static curvature profile and effectively remove dynamic components. The method is tested on a model that incorporates vehicle-bridge interaction effects and road roughness. Sensitivity analyses show that the proposed method outperforms standard filtering techniques across various sensor configurations, damage locations, severities, and multiple damage scenarios, even at relatively high vehicle speeds. Validation using field data further confirms the effectiveness and generality of the proposed approach.
{"title":"Damage localization in bridges using curvature profiles identified from acceleration data via continuous wavelet transform","authors":"Sheng-Wang Zhang , Said Quqa , Antonio Palermo , Alessandro Marzani , Zhao-Hui Lu","doi":"10.1016/j.ymssp.2026.113881","DOIUrl":"10.1016/j.ymssp.2026.113881","url":null,"abstract":"<div><div>Damage identification methods based on traffic-induced vibration data have gained significant attention in structural health monitoring of bridges, driven by the need for cost-effective sensing solutions. Recent studies have demonstrated that bridge curvature profiles can be identified from sparse acceleration measurements collected during vehicle passages using standard accelerometers. However, existing approaches for estimating curvature from acceleration data often struggle to suppress dynamic effects induced by moving vehicles. These methods typically rely on low-pass filters with a rigid cutoff threshold, which can compromise accuracy, especially during high-speed vehicle passages. To overcome this limitation, this study introduces a novel approach based on the continuous wavelet transform to isolate the quasi-static curvature profile and effectively remove dynamic components. The method is tested on a model that incorporates vehicle-bridge interaction effects and road roughness. Sensitivity analyses show that the proposed method outperforms standard filtering techniques across various sensor configurations, damage locations, severities, and multiple damage scenarios, even at relatively high vehicle speeds. Validation using field data further confirms the effectiveness and generality of the proposed approach.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"246 ","pages":"Article 113881"},"PeriodicalIF":8.9,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145969432","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}
Pub Date : 2026-01-14DOI: 10.1016/j.ymssp.2026.113874
Ketson R.M. dos Santos, João G.C.S. Duarte
Mechanical and structural systems subject to parametric excitations—fluctuations in mass, damping, or stiffness caused by phenomena such as fluid property variations or particle adhesion—are common in engineering applications. These excitations, whether deterministic or stochastic, can induce chaotic motion, instabilities, and stochastic resonance, compromising system reliability. Analyzing such systems is particularly challenging because external and parametric excitations must be addressed simultaneously, while fractional derivative terms modeling viscoelastic effects add further complexity to uncertainty propagation in nonlinear oscillators. This paper introduces an analytical score-matching methodology to evaluate the non-stationary probability density function (PDF) of the response amplitude of nonlinear oscillators equipped with a parametric fractional damper and subjected to white noise excitation. The method employs stochastic averaging to derive the stochastic differential equation governing the amplitude dynamics and reformulates the associated Fokker–Planck equation as a continuity equation. This formulation enables tracing amplitude evolution along equiprobability trajectories, thereby recovering the time-dependent PDF of the response amplitude. Numerical studies are performed for both linear and Duffing oscillators. The results reveal that the fractional derivative order significantly influences system dynamics by contributing simultaneously to damping and stiffness, which in turn shapes the response distribution. Comparisons with Monte Carlo simulations confirm the accuracy and computational efficiency of the proposed approach, demonstrating its potential as a robust tool for analyzing stochastic dynamical systems with combined parametric and fractional effects.
{"title":"Analytical score matching for efficient stochastic response determination of nonlinear oscillators with parametric fractional dampers","authors":"Ketson R.M. dos Santos, João G.C.S. Duarte","doi":"10.1016/j.ymssp.2026.113874","DOIUrl":"10.1016/j.ymssp.2026.113874","url":null,"abstract":"<div><div>Mechanical and structural systems subject to parametric excitations—fluctuations in mass, damping, or stiffness caused by phenomena such as fluid property variations or particle adhesion—are common in engineering applications. These excitations, whether deterministic or stochastic, can induce chaotic motion, instabilities, and stochastic resonance, compromising system reliability. Analyzing such systems is particularly challenging because external and parametric excitations must be addressed simultaneously, while fractional derivative terms modeling viscoelastic effects add further complexity to uncertainty propagation in nonlinear oscillators. This paper introduces an analytical score-matching methodology to evaluate the non-stationary probability density function (PDF) of the response amplitude of nonlinear oscillators equipped with a parametric fractional damper and subjected to white noise excitation. The method employs stochastic averaging to derive the stochastic differential equation governing the amplitude dynamics and reformulates the associated Fokker–Planck equation as a continuity equation. This formulation enables tracing amplitude evolution along equiprobability trajectories, thereby recovering the time-dependent PDF of the response amplitude. Numerical studies are performed for both linear and Duffing oscillators. The results reveal that the fractional derivative order significantly influences system dynamics by contributing simultaneously to damping and stiffness, which in turn shapes the response distribution. Comparisons with Monte Carlo simulations confirm the accuracy and computational efficiency of the proposed approach, demonstrating its potential as a robust tool for analyzing stochastic dynamical systems with combined parametric and fractional effects.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"245 ","pages":"Article 113874"},"PeriodicalIF":8.9,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145961924","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}
Pub Date : 2026-01-14DOI: 10.1016/j.ymssp.2026.113846
Rutong Chen, Jinhui Jiang, Yiyuan Guan
As the second inverse problem in structural dynamics, dynamic load identification is highly dependent on the system’s intrinsic properties. For distributed loads, establishing an accurate mapping between structural responses and the underlying dynamic excitations remains particularly challenging, and the ill-posed nature of the problem further amplifies measurement noise, leading to significant identification errors. To overcome these difficulties, this study proposes a novel distributed dynamic load identification framework based on a Transformer architecture that directly learns the inverse dynamic relationship without requiring explicit system parameter estimation. Specifically, Legendre orthogonal polynomial decomposition is first employed to transform the load identification task into the estimation of a finite set of orthogonal polynomial coefficients. Building upon this framework, innovative architectural optimizations are introduced by embedding physical constraints into attention computation and linear prediction, leveraging the temporal causality of dynamic responses. These enhancements improve model interpretability and substantially reduce training difficulty. Numerical simulations demonstrate that the proposed method can accurately identify sinusoidal, impact, and random loads under various noise levels. Furthermore, a distributed load identification experiment on a cantilever beam is carried out, validating the practical applicability of the approach. Finally, the selection of model hyperparameters is discussed based on fitting and generalization performance, and a comparative study with traditional dynamic calibration methods was conducted in an experimental setting, further demonstrating the superior accuracy, noise robustness, and practical reliability of the proposed framework.
{"title":"A data-driven framework for distributed dynamic load identification incorporating physics-based temporal causality constraints","authors":"Rutong Chen, Jinhui Jiang, Yiyuan Guan","doi":"10.1016/j.ymssp.2026.113846","DOIUrl":"10.1016/j.ymssp.2026.113846","url":null,"abstract":"<div><div>As the second inverse problem in structural dynamics, dynamic load identification is highly dependent on the system’s intrinsic properties. For distributed loads, establishing an accurate mapping between structural responses and the underlying dynamic excitations remains particularly challenging, and the ill-posed nature of the problem further amplifies measurement noise, leading to significant identification errors. To overcome these difficulties, this study proposes a novel distributed dynamic load identification framework based on a Transformer architecture that directly learns the inverse dynamic relationship without requiring explicit system parameter estimation. Specifically, Legendre orthogonal polynomial decomposition is first employed to transform the load identification task into the estimation of a finite set of orthogonal polynomial coefficients. Building upon this framework, innovative architectural optimizations are introduced by embedding physical constraints into attention computation and linear prediction, leveraging the temporal causality of dynamic responses. These enhancements improve model interpretability and substantially reduce training difficulty. Numerical simulations demonstrate that the proposed method can accurately identify sinusoidal, impact, and random loads under various noise levels. Furthermore, a distributed load identification experiment on a cantilever beam is carried out, validating the practical applicability of the approach. Finally, the selection of model hyperparameters is discussed based on fitting and generalization performance, and a comparative study with traditional dynamic calibration methods was conducted in an experimental setting, further demonstrating the superior accuracy, noise robustness, and practical reliability of the proposed framework.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"245 ","pages":"Article 113846"},"PeriodicalIF":8.9,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978954","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}
Pub Date : 2026-01-14DOI: 10.1016/j.ymssp.2026.113868
Kristopher Campbell , Maria Pregnolato , Raj Kamal Arora , Su Taylor , Remco Nieuwland , Piet van Andel , Myra Lydon
Bridges are a vital part of the infrastructure that shapes our society. The management of these assets against ever increasing climatic changes is providing unprecedented challenges for bridge asset owners worldwide. The effects of more frequent and severe rainfall flood events in the UK have exacerbated concerns around the management of bridge scour events. This research presents the development and field deployment of a Fiber Optic Scour Sensor (FOSS), designed to remotely monitor scour and infilling in real-time. This sensor consists of three sensing elements (fins) buried at different depths. As scouring occurs, the fins are exposed and free to move and register a response. Following the flood event, as the scour hole begins to infill, these fins are buried, and this process can be picked up on the data trace. A prototype FOSS was installed at Regent bridge in Northern Ireland; a site selected for its accessibility and suitability for monitoring. This paper outlines the installation, and the initial findings, following two storms in October 2023, demonstrating the sensor’s potential for real-time scour detection in operational environments.
{"title":"Preliminary results of a fiber optic scour sensor (FOSS) for bridges","authors":"Kristopher Campbell , Maria Pregnolato , Raj Kamal Arora , Su Taylor , Remco Nieuwland , Piet van Andel , Myra Lydon","doi":"10.1016/j.ymssp.2026.113868","DOIUrl":"10.1016/j.ymssp.2026.113868","url":null,"abstract":"<div><div>Bridges are a vital part of the infrastructure that shapes our society. The management of these assets against ever increasing climatic changes is providing unprecedented challenges for bridge asset owners worldwide. The effects of more frequent and severe rainfall flood events in the UK have exacerbated concerns around the management of bridge scour events. This research presents the development and field deployment of a Fiber Optic Scour Sensor (FOSS), designed to remotely monitor scour and infilling in real-time. This sensor consists of three sensing elements (fins) buried at different depths. As scouring occurs, the fins are exposed and free to move and register a response. Following the flood event, as the scour hole begins to infill, these fins are buried, and this process can be picked up on the data trace. A prototype FOSS was installed at Regent bridge in Northern Ireland; a site selected for its accessibility and suitability for monitoring. This paper outlines the installation, and the initial findings, following two storms in October 2023, demonstrating the sensor’s potential for real-time scour detection in operational environments.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"245 ","pages":"Article 113868"},"PeriodicalIF":8.9,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145961919","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}
Electrical equipment in substations subjected to earthquakes typically exhibits brittle damage at multiple vulnerable sections, but the exact positions on the sections are unpredictable. Relevant standards and research raise the importance of the stress response levels in seismic assessment. However, monitoring all strains at the vulnerable sections necessitates lots of strain sensors for each equipment, which is impractical because of the extensive quantity of equipment in a substation, and the strong electromagnetic interference induced by the equipment. Therefore, this paper proposes a simulation and learning co-driven prediction framework to identify multi-objective monitoring schemes. It develops multiple machine learning (ML) models to predict peak stress at multiple vulnerable sections by inputting easily-monitored responses (MRs). In which, the simulation model is cooperated to acquire precise response data, addressing the scarcity of actual samples due to the absence of monitoring systems and the limited number of earthquakes. Then, it ranks the importance of MRs for each ML model using the Shapley additive explanation method, and combines the important MRs of various ML models through the proposed Intersection, Union, or Stack strategies. The combined MRs facilitate the reconstruction of ML models, which are subsequently implemented at the site to monitor responses for post-earthquake efficient predictions. A case study on a high-voltage transformer bushing is performed. Shaking table tests validate the efficacy of the obtained monitoring schemes in both intact and damaged scenarios, revealing the practicality of applying the proposed framework to efficiently identify damage to substation equipment after earthquakes.
{"title":"Simulation and interpretable learning co-driven framework for multi-objective seismic monitoring of substation equipment","authors":"Wang Zhu , Fabrizio Paolacci , Gianluca Quinci , Qiang Xie","doi":"10.1016/j.ymssp.2026.113876","DOIUrl":"10.1016/j.ymssp.2026.113876","url":null,"abstract":"<div><div>Electrical equipment in substations subjected to earthquakes typically exhibits brittle damage at multiple vulnerable sections, but the exact positions on the sections are unpredictable. Relevant standards and research raise the importance of the stress response levels in seismic assessment. However, monitoring all strains at the vulnerable sections necessitates lots of strain sensors for each equipment, which is impractical because of the extensive quantity of equipment in a substation, and the strong electromagnetic interference induced by the equipment. Therefore, this paper proposes a simulation and learning co-driven prediction framework to identify multi-objective monitoring schemes. It develops multiple machine learning (ML) models to predict peak stress at multiple vulnerable sections by inputting easily-monitored responses (MRs). In which, the simulation model is cooperated to acquire precise response data, addressing the scarcity of actual samples due to the absence of monitoring systems and the limited number of earthquakes. Then, it ranks the importance of MRs for each ML model using the Shapley additive explanation method, and combines the important MRs of various ML models through the proposed Intersection, Union, or Stack strategies. The combined MRs facilitate the reconstruction of ML models, which are subsequently implemented at the site to monitor responses for post-earthquake efficient predictions. A case study on a high-voltage transformer bushing is performed. Shaking table tests validate the efficacy of the obtained monitoring schemes in both intact and damaged scenarios, revealing the practicality of applying the proposed framework to efficiently identify damage to substation equipment after earthquakes.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"245 ","pages":"Article 113876"},"PeriodicalIF":8.9,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978955","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}
Pub Date : 2026-01-14DOI: 10.1016/j.ymssp.2026.113885
Zhao Wang , Junkai Tong , Xiao Ying , He Sun , Lei Qi , Haoran Jin , Mengying Xie , Yang Liu
Guided wave tomography is a promising technique for quantitative evaluation of pipe defects, but its in-service application has long been constrained by the inherent limited-view imposed by typical transducer layouts, leading to low resolution and severe artifacts. This study proposes a novel imaging method based on multi-mode flexural guided waves. By exploiting the helical propagation characteristics of flexural modes with different circumferential orders, a larger synthetic angular coverage of virtual rays is achieved, overcoming the detection limitations of conventional axial paths. A multi-mode physics-informed neural network is developed, which decouples and reconstructs mixed-mode guided wave signals in parallel branches and embeds dispersion equations as a physical interpreter to supervise inversion consistency and realize multi-source information fusion. Datasets containing randomized defects are generated by a finite difference forward operator. Numerical simulations demonstrate that the proposed method can accurately reconstruct various defect types, achieving an average Pearson correlation coefficient of 0.9244 on an independent test set. Comparative analyses against single mode imaging are conducted, and imaging performance is further evaluated under different defect sizes, eccentricity, and noise levels. In a real pipe experiment, the reconstructed result achieves a correlation of 0.9068 with the ground truth, and the relative error in maximum wall loss prediction is only 4.5%. The proposed method deeply integrates physical mechanisms with data driven framework to address the limited-view imaging challenge in pipes, demonstrating strong potential for engineering applications.
{"title":"Multi-mode flexural guided waves imaging in pipes","authors":"Zhao Wang , Junkai Tong , Xiao Ying , He Sun , Lei Qi , Haoran Jin , Mengying Xie , Yang Liu","doi":"10.1016/j.ymssp.2026.113885","DOIUrl":"10.1016/j.ymssp.2026.113885","url":null,"abstract":"<div><div>Guided wave tomography is a promising technique for quantitative evaluation of pipe defects, but its in-service application has long been constrained by the inherent limited-view imposed by typical transducer layouts, leading to low resolution and severe artifacts. This study proposes a novel imaging method based on multi-mode flexural guided waves. By exploiting the helical propagation characteristics of flexural modes with different circumferential orders, a larger synthetic angular coverage of virtual rays is achieved, overcoming the detection limitations of conventional axial paths. A multi-mode physics-informed neural network is developed, which decouples and reconstructs mixed-mode guided wave signals in parallel branches and embeds dispersion equations as a physical interpreter to supervise inversion consistency and realize multi-source information fusion. Datasets containing randomized defects are generated by a finite difference forward operator. Numerical simulations demonstrate that the proposed method can accurately reconstruct various defect types, achieving an average Pearson correlation coefficient of 0.9244 on an independent test set. Comparative analyses against single mode imaging are conducted, and imaging performance is further evaluated under different defect sizes, eccentricity, and noise levels. In a real pipe experiment, the reconstructed result achieves a correlation of 0.9068 with the ground truth, and the relative error in maximum wall loss prediction is only 4.5%. The proposed method deeply integrates physical mechanisms with data driven framework to address the limited-view imaging challenge in pipes, demonstrating strong potential for engineering applications.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"245 ","pages":"Article 113885"},"PeriodicalIF":8.9,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145962642","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}
Pub Date : 2026-01-13DOI: 10.1016/j.ymssp.2026.113873
Jiahou Zhao , Wanqiang Liu , Hongwu Li , Xinhua Long
The multi-harmonic vibration response, with the blade-passing frequency as the fundamental harmonic, is the primary target in helicopter vibration control. Algorithms based on the filtered-x least mean square (FxLMS) framework typically struggle to balance control performance and computational efficiency in multi-harmonic scenarios. To enhance control effectiveness while reducing computational cost, a decoupling-based multi-harmonic hybrid control (DMHHC) algorithm is proposed. A novel decoupling compensator is designed to achieve secondary-path decoupling and amplitude equalization of the filtered reference signals across frequency bands, allowing a single FxLMS to efficiently control multiple harmonics and thus significantly reduce computational cost. Furthermore, a multi-harmonic hybrid control framework is established by integrating repetitive control (RC) with FxLMS. The secondary-path decoupling effectively eliminates the interference between RC and FxLMS during integration, enabling the algorithm to combine the fast convergence of RC with the steady-state accuracy of FxLMS. Both simulations and experiments verify the effectiveness of the proposed DMHHC algorithm, demonstrating its potential as a practical engineering solution for helicopter vibration suppression.
{"title":"An efficient decoupling-based multi-harmonic hybrid control for helicopter active vibration suppression","authors":"Jiahou Zhao , Wanqiang Liu , Hongwu Li , Xinhua Long","doi":"10.1016/j.ymssp.2026.113873","DOIUrl":"10.1016/j.ymssp.2026.113873","url":null,"abstract":"<div><div>The multi-harmonic vibration response, with the blade-passing frequency as the fundamental harmonic, is the primary target in helicopter vibration control. Algorithms based on the filtered-x least mean square (FxLMS) framework typically struggle to balance control performance and computational efficiency in multi-harmonic scenarios. To enhance control effectiveness while reducing computational cost, a decoupling-based multi-harmonic hybrid control (DMHHC) algorithm is proposed. A novel decoupling compensator is designed to achieve secondary-path decoupling and amplitude equalization of the filtered reference signals across frequency bands, allowing a single FxLMS to efficiently control multiple harmonics and thus significantly reduce computational cost. Furthermore, a multi-harmonic hybrid control framework is established by integrating repetitive control (RC) with FxLMS. The secondary-path decoupling effectively eliminates the interference between RC and FxLMS during integration, enabling the algorithm to combine the fast convergence of RC with the steady-state accuracy of FxLMS. Both simulations and experiments verify the effectiveness of the proposed DMHHC algorithm, demonstrating its potential as a practical engineering solution for helicopter vibration suppression.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"245 ","pages":"Article 113873"},"PeriodicalIF":8.9,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145961927","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}
Pub Date : 2026-01-13DOI: 10.1016/j.ymssp.2026.113883
Mingguang Shan , Jiakun Huang , Jianfeng Wang , Mengmeng Dang , Zhi Zhong , Bin Liu , Lei Liu
Out-of-plane vibration modal analysis is essential in structural health monitoring (SHM) for identifying dynamic anomalies associated with structural damage. However, vision-based approaches using depth cameras often suffer from missing depth data and noise interference, limiting their applicability in modal identification. To address these challenges specifically for slender one-dimensional (1D) beam-like structures, this study proposes a noise-robust depth reconstruction-based modal identification method (DRMI). The method integrates directional depth completion to restore missing regions, adaptive moving average filtering for noise suppression and computational efficiency, and least-squares polynomial fitting to mitigate quantization noise induced by limited depth resolution. Differential amplification is further applied to enhance high-order modal components, followed by Hankel dynamic modal decomposition for modal parameter extraction. Validated through both numerical simulations and laboratory experiments on slender beam-type structures, the proposed DRMI method enables accurate and noise-resilient identification of high-order and multi-band vibration modes directly from depth sequences. Compared with the state-of-the-art pyramid reconstruction method, DRMI improves the average modal assurance criterion by 8.6% and reduces the root mean square error by 64.8%. These results demonstrate that DRMI provides a physically consistent and noise-robust framework for out-of-plane modal analysis of slender 1D structures in practical SHM scenarios.
{"title":"Noise-robust modal identification via depth reconstruction for out-of-plane vibration in slender beam-like structures","authors":"Mingguang Shan , Jiakun Huang , Jianfeng Wang , Mengmeng Dang , Zhi Zhong , Bin Liu , Lei Liu","doi":"10.1016/j.ymssp.2026.113883","DOIUrl":"10.1016/j.ymssp.2026.113883","url":null,"abstract":"<div><div>Out-of-plane vibration modal analysis is essential in structural health monitoring (SHM) for identifying dynamic anomalies associated with structural damage. However, vision-based approaches using depth cameras often suffer from missing depth data and noise interference, limiting their applicability in modal identification. To address these challenges specifically for slender one-dimensional (1D) beam-like structures, this study proposes a noise-robust depth reconstruction-based modal identification method (DRMI). The method integrates directional depth completion to restore missing regions, adaptive moving average filtering for noise suppression and computational efficiency, and least-squares polynomial fitting to mitigate quantization noise induced by limited depth resolution. Differential amplification is further applied to enhance high-order modal components, followed by Hankel dynamic modal decomposition for modal parameter extraction. Validated through both numerical simulations and laboratory experiments on slender beam-type structures, the proposed DRMI method enables accurate and noise-resilient identification of high-order and multi-band vibration modes directly from depth sequences. Compared with the state-of-the-art pyramid reconstruction method, DRMI improves the average modal assurance criterion by 8.6% and reduces the root mean square error by 64.8%. These results demonstrate that DRMI provides a physically consistent and noise-robust framework for out-of-plane modal analysis of slender 1D structures in practical SHM scenarios.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"245 ","pages":"Article 113883"},"PeriodicalIF":8.9,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145962595","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}
Pub Date : 2026-01-13DOI: 10.1016/j.ymssp.2026.113878
Yonghang Sun , Yapeng Li , Hui Zheng , Heow Pueh Lee , Xueguan Song
As the main engineering structure, large-scale thin-walled stiffened cylindrical shells have been widely used as basic structural units in the key components of high-end equipment. Excessive structural vibration of them degrades the performance and service life of the equipment. The emergence of acoustic metastructures has provided an innovative way for vibration and noise reduction in stiffened cylindrical shells in low frequency range. In this study, an inertial amplification metastructure shell (IA meta-shell) is presented to achieve low-frequency vibration control of stiffened shell structures. An energy method based on the variational principle is developed for the analysis of the band structure and vibration transmission. A collocation-based Lagrange multiplier method is presented to realize the imposition of periodic boundary conditions to shell structures. The proposed method and the solution procedures are sufficient to deal with metastructure systems according to the numerical validation results. Using this energy method, the formation mechanism of IA bandgaps and their interaction with the stiffened cylindrical shells were investigated. Under the same added mass, the proposed structures exhibit significantly lower bandgap frequencies than conventional local resonance (LR) metastructures, offering an effective solution to the mass penalty encountered in low-frequency applications. The effects of four key parameters on wave propagation characteristics of IA meta-shells are revealed in parametric studies, and finally, the application potential of the proposed meta-shell on low-frequency vibration control are evaluated by vibration experiments of configurations with different bandgap frequencies.
{"title":"Inertial amplification metastructure shells for low frequency vibration suppression","authors":"Yonghang Sun , Yapeng Li , Hui Zheng , Heow Pueh Lee , Xueguan Song","doi":"10.1016/j.ymssp.2026.113878","DOIUrl":"10.1016/j.ymssp.2026.113878","url":null,"abstract":"<div><div>As the main engineering structure, large-scale thin-walled stiffened cylindrical shells have been widely used as basic structural units in the key components of high-end equipment. Excessive structural vibration of them degrades the performance and service life of the equipment. The emergence of acoustic metastructures has provided an innovative way for vibration and noise reduction in stiffened cylindrical shells in low frequency range. In this study, an inertial amplification metastructure shell (IA <em>meta</em>-shell) is presented to achieve low-frequency vibration control of stiffened shell structures. An energy method based on the variational principle is developed for the analysis of the band structure and vibration transmission. A collocation-based Lagrange multiplier method is presented to realize the imposition of periodic boundary conditions to shell structures. The proposed method and the solution procedures are sufficient to deal with metastructure systems according to the numerical validation results. Using this energy method, the formation mechanism of IA bandgaps and their interaction with the stiffened cylindrical shells were investigated. Under the same added mass, the proposed structures exhibit significantly lower bandgap frequencies than conventional local resonance (LR) metastructures, offering an effective solution to the mass penalty encountered in low-frequency applications. The effects of four key parameters on wave propagation characteristics of IA <em>meta</em>-shells are revealed in parametric studies, and finally, the application potential of the proposed <em>meta</em>-shell on low-frequency vibration control are evaluated by vibration experiments of configurations with different bandgap frequencies.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"245 ","pages":"Article 113878"},"PeriodicalIF":8.9,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145961926","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}
Pub Date : 2026-01-13DOI: 10.1016/j.ymssp.2026.113853
Jing Yuan , Yifeng Lu , Hanlu Qian , Huiming Jiang , Qian Zhao , Yaguo Lei
Timely detection of early gear failures is a significant challenge in gear health monitoring, particularly challenging in novel gear drive systems such as planetary gears and harmonic gears. Frequency-based health indicators (HIs) are relatively less sensitive to early faults compared to time-domain approaches and inherently exhibit a delay in gear early fault warning. Meanwhile, multichannel signals inherently contain richer machine condition information compared to single channel signals. Thus, temporal sparse weight based gear health monitoring tool by multichannel phase synchronized fusion dual-lifting tree model is proposed for gear health monitoring. Multichannel phase synchronized fusion dual-lifting tree model is constructed, decomposing the raw data into several components through three levels. First, high order singular value decomposition (HOSVD) is employed, to separate noise and feature components from raw data as the 1st level of tree model. Second, multichannel phase synchronized fusion (MPSF) is proposed as the 2nd level of tree model to address phase desynchronization in full-lifecycle multichannel vibration signals, enabling linear multichannel feature fusion. It introduces multi-IMF mean phase coherence (MIMPC) for phase synchronization and compensation, producing multichannel phase synchronized feature components. Additionally, MPSF employs an estimated noise-assisted random matrix model for feature fusion, generating fused feature that integrate multichannel gear vibration signals effectively. Third, a dual-lifting transform (DLT) is proposed as the 3rd level, aimed at obtaining a dual-lifting enhanced signal to extract and quantitatively amplify early weak fault features related to faults in the fused time-domain signal. Adaptive blind deconvolution is employed as a first lifting processing to extract the gear fault features from the fused features after MPSF. Subsequently, a neighboring coefficient operator is applied to quantitatively amplify gear fault features and suppress other irrelevant residual signals. Finally, the dual-lifting enhanced signal is introduced into unified sparsity measurement framework, and the optimized temporal sparse weights are calculated by solving convex optimization for constructing temporal sparse weight based gear health indicator (TSWGHI). An experimental case of robotic harmonic reducer and an engineering case of finishing mill gearbox show that the proposed tool demonstrates remarkable performance in gear health monitoring by comparing with traditional and popular HIs.
{"title":"Temporal sparse weight based gear health monitoring tool by multichannel phase synchronized fusion dual-lifting tree model","authors":"Jing Yuan , Yifeng Lu , Hanlu Qian , Huiming Jiang , Qian Zhao , Yaguo Lei","doi":"10.1016/j.ymssp.2026.113853","DOIUrl":"10.1016/j.ymssp.2026.113853","url":null,"abstract":"<div><div>Timely detection of early gear failures is a significant challenge in gear health monitoring, particularly challenging in novel gear drive systems such as planetary gears and harmonic gears. Frequency-based health indicators (HIs) are relatively less sensitive to early faults compared to time-domain approaches and inherently exhibit a delay in gear early fault warning. Meanwhile, multichannel signals inherently contain richer machine condition information compared to single channel signals. Thus, temporal sparse weight based gear health monitoring tool by multichannel phase synchronized fusion dual-lifting tree model is proposed for gear health monitoring. Multichannel phase synchronized fusion dual-lifting tree model is constructed, decomposing the raw data into several components through three levels. First, high order singular value decomposition (HOSVD) is employed, to separate noise and feature components from raw data as the 1st level of tree model. Second, multichannel phase synchronized fusion (MPSF) is proposed as the 2nd level of tree model to address phase desynchronization in full-lifecycle multichannel vibration signals, enabling linear multichannel feature fusion. It introduces multi-IMF mean phase coherence (MIMPC) for phase synchronization and compensation, producing multichannel phase synchronized feature components. Additionally, MPSF employs an estimated noise-assisted random matrix model for feature fusion, generating fused feature that integrate multichannel gear vibration signals effectively. Third, a dual-lifting transform (DLT) is proposed as the 3rd level, aimed at obtaining a dual-lifting enhanced signal to extract and quantitatively amplify early weak fault features related to faults in the fused time-domain signal. Adaptive blind deconvolution is employed as a first lifting processing to extract the gear fault features from the fused features after MPSF. Subsequently, a neighboring coefficient operator is applied to quantitatively amplify gear fault features and suppress other irrelevant residual signals. Finally, the dual-lifting enhanced signal is introduced into unified sparsity measurement framework, and the optimized temporal sparse weights are calculated by solving convex optimization for constructing temporal sparse weight based gear health indicator (TSWGHI). An experimental case of robotic harmonic reducer and an engineering case of finishing mill gearbox show that the proposed tool demonstrates remarkable performance in gear health monitoring by comparing with traditional and popular HIs.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"245 ","pages":"Article 113853"},"PeriodicalIF":8.9,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145961925","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号