Mechanical Systems and Signal Processing最新文献

{"title":"Flow-HBM: A generative likelihood-free hierarchical Bayesian model updating framework with dual normalizing flow-based inference networks","authors":"Jice Zeng ,&nbsp;Hui Chen ,&nbsp;Zhao Zhao ,&nbsp;Zi-Jun Cao","doi":"10.1016/j.ymssp.2025.113398","DOIUrl":"10.1016/j.ymssp.2025.113398","url":null,"abstract":"<div><div>Hierarchical Bayesian modeling (HBM) has emerged as a powerful framework for quantifying uncertainties in structural dynamics by introducing hyperparameters that govern the distributions of model parameters. However, practical application of HBM is hindered by several challenges. Approximations such as Laplace and variational inference often impose restrictive assumptions, the sampling methods are computationally expensive. Most critically, the likelihood function in complex hierarchical models is typically intractable, limiting the feasibility of standard Bayesian inference. To address these challenges, this study proposes Flow-HBM, a novel data-driven, likelihood-free HBM framework based on normalizing flow generative model. First, synthetic datasets are generated by sampling hyperparameters from the prior and simulating responses using a finite element model. A normalizing flow model is then trained to learn the complex posterior distributions by minimizing the Kullback–Leibler divergence between the true and model-estimated posteriors via maximum likelihood training on the synthetic data. To efficiently estimate model parameters, hyperparameters, and prediction error, the joint posterior is factorized into two components: (1) the posterior of hyperparameters and prediction error given all data, and (2) the posterior of model parameters given the hyperparameters, prediction error, and individual dataset. This leads to two flow-based inference networks: a model inference network (MIN) for estimating the posterior distribution of model parameters conditioned on dataset-specific observations, and a hyper inference network (HIN) for inferring the posterior of hyperparameters and prediction error parameters conditioned on the aggregated data across all datasets. Both MIN and HIN are implemented using interleaved affine coupling and neural spline flow layers, and trained jointly in an offline phase. Once trained, the framework enables near-instant inference of all unknowns by sampling from a base Gaussian and applying the learned invertible mappings, bypassing the need for likelihood evaluation. The proposed method is validated on a four-story shear building and a reinforced concrete slab, demonstrating accurate parameter estimation and significant computational gains, paving the way for real-time hierarchical Bayesian model updating.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"240 ","pages":"Article 113398"},"PeriodicalIF":8.9,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159962","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":"Structural dynamics-driven digital twin framework for real-time vibration modelling of machine tools","authors":"Yue Qiu ,&nbsp;Xinyong Mao ,&nbsp;Chaoyang Gao ,&nbsp;Yanyan Xu ,&nbsp;Hongqi Liu","doi":"10.1016/j.ymssp.2025.113406","DOIUrl":"10.1016/j.ymssp.2025.113406","url":null,"abstract":"<div><div>Structural vibration under dynamic machining conditions poses a significant challenge to the performance and reliability of machine tools. Traditional static and offline modeling approaches fail to adapt to real-time changes in structural dynamics. To address this, this paper proposed a structure dynamics-driven digital twin framework, enabling real-time monitoring, modeling, and feedback control of machine tool vibrations. Distinct from conventional digital twins that focus solely on geometric and kinematic synchronization, the proposed approach integrates physical dynamic characteristics as the core driver of digital twin evolution. To realize this concept, a digital twin platform for machine tool dynamics has been developed, featuring real-time feedback and interaction with the physical machine via bidirectional communication. Built upon this platform, a self-excitation strategy using servo feedforward injection is introduced to identify modal parameters during operation. Furthermore, variational mode decomposition (VMD) integrated with real-time modal information and modal superposition enables accurate full-field vibration twinning. Experiments conducted on a CHX-5240i vertical lathe verify the system’s accuracy, with modal frequency identification errors of less than 2% and a vibration twinning relative error of under 9%. Moreover, vibration thresholds derived from the digital twin are shown to correlate with surface roughness limits and machining stability boundaries, validating its utility in performance-oriented process control. This work achieves system-level advances in dynamic model integration, real-time modal updating, and vibration–performance mapping. By enabling real-time bidirectional interaction with the physical system, the proposed structure vibration digital twin framework offers a novel paradigm for intelligent manufacturing, with strong potential for deployment in closed-loop control and adaptive optimization of machining processes.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"240 ","pages":"Article 113406"},"PeriodicalIF":8.9,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145160034","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":"Hysteresis characteristics of functional gradient metal rubber: constitutive modeling and experimental validation","authors":"Yuhan Wei ,&nbsp;Fang Yang ,&nbsp;Yanpeng Yang ,&nbsp;Xin Xue","doi":"10.1016/j.ymssp.2025.113388","DOIUrl":"10.1016/j.ymssp.2025.113388","url":null,"abstract":"<div><div>As a novel material that integrates structural properties and functions, functional gradient metal rubber (FGMR) has extensive application prospects due to its lightweight, thermal protection, and vibration-damping properties. Of particular interest and complexity is the unclear relationship between the gradient configuration and hysteresis characteristics, owing to the unique internal structure of FGMR. In this work, the relationship between pore characteristics and the internal structure of FGMR was investigated using a gradient equation derived from image processing, which incorporates CT scanning and threshold segmentation techniques. Based on this, a constitutive model was constructed and validated through quasi-static and dynamic compression experiments. The results demonstrate that a gradient equation can effectively represent the relationship between the density and structure. The gradient configuration has a significant influence on the compressive performance of FGMR. High prediction accuracy of the constitutive modeling for the FGMR is exhibited.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"240 ","pages":"Article 113388"},"PeriodicalIF":8.9,"publicationDate":"2025-09-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145160030","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 approach to surrogate modelling of modal properties: Mode-shape-adapted input parameter domain cutting","authors":"Blaž Kurent ,&nbsp;Bence Popovics ,&nbsp;Boštjan Brank ,&nbsp;Noémi Friedman","doi":"10.1016/j.ymssp.2025.113381","DOIUrl":"10.1016/j.ymssp.2025.113381","url":null,"abstract":"<div><div>Surrogate models, also known as meta-models or proxy models, have become invaluable in structural engineering. They are a great addition to the finite element models, providing a fast computational alternative for approximating the quantity of interest (QOI). By the quick evaluation of the surrogate model, they can accelerate stochastic analyses of the structural response under the uncertainties of its input parameters (such as uncertainty quantification and sensitivity analysis) as well as the processes of optimisation and probabilistic model updating. They also offer an offline computation of the QOI which is particularly beneficial in scenarios of structural health monitoring where access to licenced software is limited. Surrogate modelling of modal properties is particularly challenging due to the mode degeneration phenomena, such as mode crossing, veering, and coalescence. The paper introduces a novel approach to surrogate modelling of modal properties that is accurate and reduces the required number of training points. The here-introduced mode-shape-adapted input parameter domain cutting (MOSAIC) surrogate modelling technique is a form of piecewise approximation. The novelty of this approach lies in the intelligent cutting of the parameter domain into subdomains, which identifies regions where the mode shapes smoothly change. As with all black-box surrogate modelling techniques, the method requires only a set of parameter samples and the computation of the corresponding QOIs (here the modal properties) by the finite element model. The paper presents the method in detail and provides three examples with two, six, and seven input parameters, respectively. In all of the examples, mode degeneration phenomena are present. The MOSIAC surrogate model achieves significantly better accuracy than the benchmark surrogate model, which is trained over the whole parameter domain without cutting it. The accuracy of the MOSAIC surrogate model outperforms even the benchmark model that is trained on ten times as many training points. This indicates a large time-saving potential in building surrogate models of modal properties. The accuracy and efficiency of the MOSAIC method are further enhanced by the proposed active learning approach.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"240 ","pages":"Article 113381"},"PeriodicalIF":8.9,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159966","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":"PET-AE: Physics-informed enhanced temporal autoencoder for incipient fault detection of shafting systems","authors":"Zhan Gao ,&nbsp;Kaiwei Yu ,&nbsp;Jun Wu ,&nbsp;Weixiong Jiang ,&nbsp;Bo Yang","doi":"10.1016/j.ymssp.2025.113345","DOIUrl":"10.1016/j.ymssp.2025.113345","url":null,"abstract":"<div><div>Incipient fault detection is crucial for improving the stable operation of shafting systems. Autoencoders (AEs) have gained popularity in the field of incipient fault detection. However, AE-based methods are weak in capturing temporal and periodic dependencies hidden in monitoring signals. This hinders the timely detection of incipient faults. To tackle these challenges, a physics-informed enhanced temporal autoencoder (PET-AE) is proposed for incipient fault detection of shafting systems. In this method, a Transformer autoencoder is constructed to reconstruct signals, where the differential Transformer encoder is used to mine temporal features from input signals. Moreover, a spectrum module is designed to capture global and local frequency information to enhance the periodic representations. Then, an enhanced memory module is employed to enlarge the distribution gap between normal samples and degradation samples. To verify the effectiveness of the proposed method, experimental studies are implemented on IMS bearing dataset and a self-built propulsive shafting system. Experimental results demonstrate that the proposed PET-AE has outstanding fault detection performance compared to other advanced detection methods.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"240 ","pages":"Article 113345"},"PeriodicalIF":8.9,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145120135","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":"Cross-fidelity nonlinear dynamic response predictions of steel frame buildings using CNN-LSTM deep learning models with transformer and attention mechanisms","authors":"Lei Liao ,&nbsp;Yazhou Xie ,&nbsp;Chunxiao Ning ,&nbsp;Suiwen Wu","doi":"10.1016/j.ymssp.2025.113305","DOIUrl":"10.1016/j.ymssp.2025.113305","url":null,"abstract":"<div><div>Seismic responses of building frames can be predicted using simplistic low fidelity (e.g., equivalent single-degree-of-freedom mass–spring–dashpot systems) or material mechanics-based high fidelity (e.g., fiber-section beam column or solid element finite element models) numerical models with a trade-off between prediction accuracy and computational efficiency. While low fidelity models have inherent limitations, their embedded computational efficiency and physics mechanism can be leveraged to couple with data-driven approaches to achieve high-fidelity seismic response predictions. This paper develops a novel cross-fidelity deep learning (DL) framework, which combines seismic ground motions (GM) and low fidelity structural responses as complementary inputs, to improve the accuracy and robustness in predicting high-fidelity nonlinear seismic responses of different steel frame buildings. The proposed models utilize hybrid architectures that integrate convolutional neural networks (CNN), long short-term memory (LSTM), transformer, and self-attention mechanisms to effectively capture time–frequency–magnitude dependencies inherent in seismic response data. Performance of these models is evaluated on three representative steel frame buildings in California and compared against six GM single-input DL models, as well as three dual-input models without having the CNN module. The proposed DL models with hybrid architectures and the cross-fidelity input mechanism consistently outperform other models, demonstrating significantly improved effectiveness in predicting the entire dynamic response history. Results indicate that integrating low-fidelity model responses as physics-guided inputs reduces prediction variance and enhances the reliability of time-series inference. This study highlights the potential of the proposed cross-fidelity DL approaches for improving seismic response predictions, which could be utilized to support downstream applications such as seismic risk assessment, rapid post-earthquake evaluation, and performance-based seismic design.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"240 ","pages":"Article 113305"},"PeriodicalIF":8.9,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145120728","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":"Dynamic response prediction of air-floating large-size sheet with mode superposition","authors":"Deyi Kong ,&nbsp;Yu Fu ,&nbsp;Ming Xie ,&nbsp;Wei Tang ,&nbsp;Kun Bai ,&nbsp;Jiankui Chen ,&nbsp;Zhouping Yin","doi":"10.1016/j.ymssp.2025.113402","DOIUrl":"10.1016/j.ymssp.2025.113402","url":null,"abstract":"<div><div>Non-contact air support and high-precision feeding for large-size sheet are crucial for inkjet printing (IJP) of displays. Modeling and predicting the dynamic response of the flexible sheet is vital in system design, as the continuous deformations and vibrations during the handling of the large sheet lead to reduced precision and quality in the final products. This paper presents a computationally efficient algorithm for analyzing and predicting the dynamic behavior of an air-floating large-size sheet. A modal expression of the coupled system consisting of air film and flexible sheet is developed based on a distributed-parameter model. The proposed predictive algorithm allows for fast computation and analysis of the sheet’s dynamic behavior by superimposing finite-order air-sheet modes. The solutions have been numerically verified against those obtained from a fluid–structure interaction (FSI) model using finite element analysis (FEA). In the experimental section, a parameter identification method based on the dynamic model is introduced. Subsequently, experiments were conducted on IJP equipment for thin film encapsulation (TFE) with a cantilevered glass sheet supported by four porous air-bearing guides. The theoretical calculations were validated by comparing them with the experimental findings. This model provides a quick and effective approach for the dynamic modeling of air-floating systems and can assist in their design and parameter optimization.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"240 ","pages":"Article 113402"},"PeriodicalIF":8.9,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159864","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":"Tension–path co-optimization of cable-driven manipulators based on a digital twin framework","authors":"Dongyang Hu ,&nbsp;Dawei Xu ,&nbsp;Hamid Reza Karimi ,&nbsp;Yanfeng Wang ,&nbsp;Huanlong Zhang ,&nbsp;Yongjie Zhai","doi":"10.1016/j.ymssp.2025.113369","DOIUrl":"10.1016/j.ymssp.2025.113369","url":null,"abstract":"<div><div>Cable-driven manipulators, characterized by their hyper-redundant degrees of freedom and exceptional flexibility, present significant potential for performing complex tasks. However, traditional approaches face challenges in concurrently addressing path planning and tension distribution, often leading to reduced control accuracy and compromised system stability. This study introduces a Digital Twin-based framework for tension–path co-optimization. A kinematic model of the manipulator and a cable tension model are developed in a virtual environment. By integrating physical modeling with data-driven techniques, the framework enables accurate simulation of the manipulator’s motion and cable tension distribution. A gradient descent optimization method is employed to simultaneously optimize tension distribution and the motion path. To ensure high-precision closed-loop control between the virtual and physical spaces, a self-compensating optical fiber angle sensor feedback mechanism is implemented, effectively minimizing joint angle errors. The proposed methodology is comprehensively validated through Digital Twin simulations and experimental testing, with a focus on tension prediction accuracy, path optimization efficacy, and control precision. The results demonstrate that the proposed approach outperforms traditional models in terms of tension prediction accuracy, path optimization, and joint angle error reduction, exhibiting superior precision and stability under various stiffness settings.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"240 ","pages":"Article 113369"},"PeriodicalIF":8.9,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145120727","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":"Sensor placement strategy and evaluation for underwater active electrosense arrays","authors":"Guangyu Jiang ,&nbsp;Qiao Hu ,&nbsp;Haoran Peng ,&nbsp;Tongqiang Fu ,&nbsp;Yi Rong ,&nbsp;Qian Yang ,&nbsp;Pengtao Wang ,&nbsp;Liqing Huang","doi":"10.1016/j.ymssp.2025.113400","DOIUrl":"10.1016/j.ymssp.2025.113400","url":null,"abstract":"<div><div>The active electrosense mechanism of weakly electric fish provides a novel approach for underwater robot localization and navigation. However, unsuitable sensor placement can easily lead to data redundancy or the loss of crucial information. Herein, we proposed a novel sensor placement strategy for localizing interested-region targets with active electrosense arrays. This paper formulates an optimization problem for sensor placement based on the array signal model and Cramer-Rao lower bound (CRLB). To solve such a nonlinear and nonconvex problem, an enhanced grey wolf optimizer (EGWO) with dynamic weight and segment search mechanisms is developed. Two indexes, relative Euclidean error and effective localization area, are introduced to evaluate the array performance. The results in both simulation and experiment demonstrate the effectiveness of the proposed sensor placement strategy, as well as its superiority over uniform placement. The proposed sensor placement strategy can be used to design active electrosense systems on bionic robotic fish and other underwater vehicles.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"240 ","pages":"Article 113400"},"PeriodicalIF":8.9,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145109838","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 Conditional Adaption Alignment Dynamic Graph Neural Network model for unsupervised fault diagnosis of rotating machinery","authors":"Ran Wang ,&nbsp;Zhengtai Lyu ,&nbsp;Fucheng Yan ,&nbsp;Liang Yu ,&nbsp;Xiong Hu","doi":"10.1016/j.ymssp.2025.113361","DOIUrl":"10.1016/j.ymssp.2025.113361","url":null,"abstract":"<div><div>In rotating machinery, the diverse operating conditions and cross-machine discrepancies lead to significant distribution shifts in monitoring data, with label scarcity persisting under new or specific conditions. These challenges motivate the application of domain adaptation techniques in cross-domain fault diagnosis. In recent years, unsupervised fault diagnosis methods based on graph neural networks still face several limitations. The geometric structure of data distributions is often neglected by existing conditional distribution adaptation methods. To address this problem, a Conditional Adaptive Alignment Dynamic Graph Neural Network (CA-DGNN) cross-domain unsupervised model is proposed. First, a Dynamic Graph Neural Network integrated with Gaussian edge features is constructed, where inter-sample correlations are learned through dynamic topological structures and embedded into graph-level fault feature representations. Subsequently, the relationship between graph-level fault features and labels is explicitly established through the Conditional Maximum Mean Discrepancy (CMMD), which is formulated within the Reproducing Kernel Hilbert Space (RKHS) using the conditional covariance operator. The CMMD is then used to measure the domain discrepancy of feature-conditional distributions, and a conditional adaptive loss is designed to realize the domain alignment. The intra-class knowledge transfer is enhanced compared to traditional marginal alignment. Additionally, mutual information between fault features and predicted labels is utilized to extract discriminative information and improve the reliability of pseudo-labels. The proposed method is evaluated through experiments on two varying operational condition cases and one cross-machine case, with the results demonstrating that the model is more effective than other models in cross-domain fault diagnosis tasks. The codes of CA-DGNN model are released at: <span><span>https://github.com/Pear-so/CA-DGNN</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"240 ","pages":"Article 113361"},"PeriodicalIF":8.9,"publicationDate":"2025-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145120136","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}