Mechanical Systems and Signal Processing最新文献

Extracting material-dependent attenuation from the PZT monitoring system based on FEM and transformer 基于有限元法和变压器的压电陶瓷监测系统材料衰减提取

IF 7.9 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2025-07-01 DOI: 10.1016/j.ymssp.2025.113047

Wenxuan Cao , Junjie Li

{"title":"Extracting material-dependent attenuation from the PZT monitoring system based on FEM and transformer","authors":"Wenxuan Cao , Junjie Li","doi":"10.1016/j.ymssp.2025.113047","DOIUrl":"10.1016/j.ymssp.2025.113047","url":null,"abstract":"<div><div>Piezoelectric (PZT) monitoring systems have been demonstrated to be both sensitive and efficient for structural health monitoring. However, the attenuation information collected by the PZT monitoring system often contains components that are unrelated to material properties, referred to as material-independent attenuation. Examples of this include wave diffusion attention, force-electric conversion loss, and so on. These material-independent attenuations can obscure valuable information within the signal and hinder the discrimination of structural damage. This study proposed a novel approach for extracting the attenuation components that are dependent on material internal properties, termed material-dependent attenuation. The operating mode of the PZT monitoring system was initially analyzed, leading to the derivation of the total attenuation equation. Subsequently, a surrogate model for predicting material-independent attenuation was introduced, which integrates the convolutional neural network (CNN) with the Transformer. The CNN was structured as a multi-residual channel module to improve feature resolution, while the Transformer served to encode and filter the positional attributes of these features. The training data for this model was generated through numerical simulations. Finally, based on the total attenuation equation, the material-independent attenuation was eliminated from the total attenuation to obtain the material-dependent attenuation. Numerical simulation tests demonstrated that the proposed approach can accurately extract the material-dependent attenuation from the PZT monitoring system. Additionally, a field experiment was also designed to apply the proposed approach to invert the concrete dynamic permeability coefficient. The inversion results indicated that the proposed approach can accurately extract the material-dependent attenuation with considerable practicality and generalization.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113047"},"PeriodicalIF":7.9,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517894","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DTFFNet: A dual-branch time–frequency feature fusion network for remaining useful life prediction of mechanical equipment DTFFNet：用于机械设备剩余使用寿命预测的双支路时频特征融合网络

IF 7.9 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2025-07-01 DOI: 10.1016/j.ymssp.2025.113006

Jiangyan Zhu , Jun Ma , Jiande Wu , Lekang Fan

{"title":"DTFFNet: A dual-branch time–frequency feature fusion network for remaining useful life prediction of mechanical equipment","authors":"Jiangyan Zhu , Jun Ma , Jiande Wu , Lekang Fan","doi":"10.1016/j.ymssp.2025.113006","DOIUrl":"10.1016/j.ymssp.2025.113006","url":null,"abstract":"<div><div>With its exceptional feature extraction and modeling capabilities, deep learning has emerged as a cornerstone technology in the field of remaining useful life (RUL) prediction for mechanical equipment. However, time-domain-based deep learning methods face significant limitations in capturing global features, hindering their ability to fully model the complexities of equipment degradation processes. To address these challenges, we propose DTFFNet, a dual-branch time–frequency interactive integration framework that combines time-domain and frequency-domain signal analysis for efficient feature interaction and fusion. Specifically, the time-domain branch employs a CNN-Transformer structure to extract local features and model long-term dependencies within sequences. The frequency-domain branch utilizes a frequency-domain Multilayer Perceptron (MLP) to extract global dependencies from the real and imaginary components of Fourier-transformed signals. Additionally, a multi-head cross-attention mechanism is introduced to construct a time–frequency cross-domain interactive fusion module, enabling information exchange and complementary representation learning across time and frequency domains, enhancing feature fusion depth and effectiveness. To meet the specific demands of RUL prediction, a joint learning strategy is proposed that combines weighted and constrained MSE losses, enabling targeted optimization during critical degradation phases and along specific error directions to enhance prediction accuracy. Finally, experimental analysis on widely used turbofan engine dataset and PHM 2012 bearing dataset demonstrates that DTFFNet not only significantly improves prediction accuracy compared to mainstream RUL prediction methods but also exhibits enhanced robustness in noisy environments. These results establish DTFFNet as a reliable and effective solution for equipment health management in complex industrial settings.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113006"},"PeriodicalIF":7.9,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517895","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Novel adaptive Bayesian scheme for real-time simultaneous anomaly detection and system identification 一种实时同步异常检测和系统识别的自适应贝叶斯算法

IF 7.9 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2025-07-01 DOI: 10.1016/j.ymssp.2025.113051

Dukang Huang , Ke Huang , Lei Xiao , Yafei Ma , Ka-Veng Yuen , Lei Wang

{"title":"Novel adaptive Bayesian scheme for real-time simultaneous anomaly detection and system identification","authors":"Dukang Huang , Ke Huang , Lei Xiao , Yafei Ma , Ka-Veng Yuen , Lei Wang","doi":"10.1016/j.ymssp.2025.113051","DOIUrl":"10.1016/j.ymssp.2025.113051","url":null,"abstract":"<div><div>This study presents a novel approach for real-time anomaly detection and system identification. The approach eliminates the need for fixed threshold settings in anomaly detection and provides an efficient solution for simultaneous recognition of multiple anomaly types and identification of time-varying systems. Statistical models for random and gross errors are introduced to represent typical measurement anomalies, and Bernoulli random vectors are used for anomaly detection. Once potential anomalies are recognized, they are either excluded or compensated in further real-time system identification through detect-to-reject and detect-to-fix procedures. An adaptive Bayesian scheme updates both the Bernoulli and model parameters, allowing for real-time simultaneous anomaly detection and system identification. The approach is verified through numerical simulation and laboratory experiment. Moreover, it is implemented in a full-scale monitoring system. The proposed method effectively detects multiple anomaly types and achieves reliable identification results for time-varying systems.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113051"},"PeriodicalIF":7.9,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517896","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Wave-structure conversion and rainbow trapping effect in complex cross-sectional structures 复杂截面结构中的波结构转换与彩虹俘获效应

IF 7.9 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2025-07-01 DOI: 10.1016/j.ymssp.2025.113034

Xicheng Feng , Zihan Zhou , Jingmang Xu , Kang Li , Ping Wang , Jun Lai

{"title":"Wave-structure conversion and rainbow trapping effect in complex cross-sectional structures","authors":"Xicheng Feng , Zihan Zhou , Jingmang Xu , Kang Li , Ping Wang , Jun Lai","doi":"10.1016/j.ymssp.2025.113034","DOIUrl":"10.1016/j.ymssp.2025.113034","url":null,"abstract":"<div><div>The straight switch rail of the turnout is an engineering structure with variable cross-sections and special-shaped cross-sections. The propagation characteristics of guided waves in it are very complicated. Even for the guided waves of a single mode, the distribution of wave amplitudes varies greatly at different cross-sections of the straight rail. Therefore, this paper innovatively proposes a wave-structure conversion algorithm for variable cross-section structures. Employing this algorithm enables us to discern the pattern of wave amplitude distribution and the evolution of dispersion characteristics as a single mode traverses a straight switch rail. This paper also established a transient finite element model of the straight switch rail. Through the comb-like excitation method based on the principle of group velocity, the propagation of a single guided wave mode in the straight switch rail was simulated, thereby verifying the correctness of the wave-structure conversion algorithm for variable cross-section structures. Meanwhile, the dispersion information of modes in different key cross-sections is obtained based on the algorithm in this paper, and the rainbow trapping effect in straight switch rails is found for the first time. Eventually, the theoretical analysis results and the rainbow trapping phenomenon in the straight switch rail were verified through experiments. The main conclusions obtained are as follows: Under the excitation of specific center frequencies, some modes will exhibit the rainbow trapping effect, which will cease to propagate forward in the healthy straight switch rail and generate a completely reflected echo.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113034"},"PeriodicalIF":7.9,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144524287","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Unsteady aerodynamics identification of transonic buffet by incorporating shock position 基于冲击位置的跨声速颤振非定常气动辨识

IF 7.9 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2025-07-01 DOI: 10.1016/j.ymssp.2025.112995

Qiyue Ma , Chuanqiang Gao , Dangguo Yang , Weiwei Zhang

{"title":"Unsteady aerodynamics identification of transonic buffet by incorporating shock position","authors":"Qiyue Ma , Chuanqiang Gao , Dangguo Yang , Weiwei Zhang","doi":"10.1016/j.ymssp.2025.112995","DOIUrl":"10.1016/j.ymssp.2025.112995","url":null,"abstract":"<div><div>Transonic shock buffet is characterized by self-excited and large-amplitude oscillations of shock wave and the aerodynamics, resulting in reducing the handling quality and even causing flight accidents. Efficient and accurate prediction of the unsteady aerodynamic loads caused by the shock buffet, thereby, is an urgent and challenging work in the aeronautical engineering. With the sparse identification technique, an unsteady aerodynamic modeling framework is proposed to predict the shock buffeting loads over an airfoil. First of all, dynamic analysis reveals that the oscillating lift coefficient is strongly dominated by shock wave motion. With the sparse regression framework, an algebraic equation is, then, derived from the time series samples to parametrically describe the dynamic behavior of the shock buffet by incorporating the dynamical features of shock wave. By specific analysis of coherent structures with different frequencies, modeling can be achieved in both light and deep buffeting phenomena. The temporal response of the lift coefficient can be effectively predicted at varying Mach numbers and angles of attack with a relative error of less than 4%. This approach establishes a physics-informed function mapping between measurable shock wave dynamics and challenging-to-quantify lift forces, offering a potential solution for predicting aerodynamic force.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 112995"},"PeriodicalIF":7.9,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-objective optimization for vibration suppression and weight reduction in composite sandwich shallow-spherical shells with functionally graded coating 功能梯度涂层复合材料夹层浅球壳减振减重多目标优化

IF 7.9 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2025-07-01 DOI: 10.1016/j.ymssp.2025.113049

Jichuan Cao , Hui Li , Zhaoye Qin , Yang Liu , Haiyang Zhang , Xiangping Wang , Qingkai Han

{"title":"Multi-objective optimization for vibration suppression and weight reduction in composite sandwich shallow-spherical shells with functionally graded coating","authors":"Jichuan Cao , Hui Li , Zhaoye Qin , Yang Liu , Haiyang Zhang , Xiangping Wang , Qingkai Han","doi":"10.1016/j.ymssp.2025.113049","DOIUrl":"10.1016/j.ymssp.2025.113049","url":null,"abstract":"<div><div>This study is the first to integrate the composite sandwich shallow spherical shell (CSSS) with functionally graded coating (FGC) and an improved meta-heuristic algorithm, and optimize the vibration suppression and lightweight performance. Initially, using the mixing principle, high-order displacement field theory, complex modulus method, Lagrange equation and Newmark-<em>β</em> method, a theoretical model of the FGC-CSSS subjected to base excitation is established to analyze the free and forced vibrations of the structure. Then, the vibration suppression, stiffness and lightweight performance of the FGC-CSSS are optimized by a multi-objective approach utilizing the improved walrus optimization algorithm (IWOA). The minimum of the first resonant response amplitude, the minimum of the equivalent overall mass of the FGC-CSSS and the minimum of the reciprocal of the sum of natural frequencies are taken as objective functions, respectively. Also, the core thickness ratio, the coating thickness ratio, and the honeycomb cell thickness are selected as design variables. The effectiveness of the IWOA is validated through an example of engineering, and the data obtained from the theoretical model are compared with the literature and experimental results to ensure robust validation. Finally, an optimization method of the FGC-CSSS is conducted, and the landmark optimization results and corresponding variations on pareto front are displayed. These results demonstrate that the IWOA has better multi-objective optimization capability than NSGA-Ⅱ, and design variables C<sub>1</sub> and C<sub>2</sub> can improve vibration suppression, lightness, and structural stiffness performance by 85.5%, 43.5%, and 8.7%, respectively.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113049"},"PeriodicalIF":7.9,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144524286","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Remaining useful life prediction by degradation distribution transport health indicator and consolidated memory stabilized LSTM 基于退化分布、传输健康指标和整合内存稳定LSTM的剩余使用寿命预测

IF 7.9 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2025-07-01 DOI: 10.1016/j.ymssp.2025.113039

Ting Zhu , Zhen Chen , Di Zhou , Zhaoxiang Chen , Ershun Pan

{"title":"Remaining useful life prediction by degradation distribution transport health indicator and consolidated memory stabilized LSTM","authors":"Ting Zhu , Zhen Chen , Di Zhou , Zhaoxiang Chen , Ershun Pan","doi":"10.1016/j.ymssp.2025.113039","DOIUrl":"10.1016/j.ymssp.2025.113039","url":null,"abstract":"<div><div>Accurate prediction of the remaining useful life (RUL) is of paramount importance for preventing unexpected failures in industrial machinery. This primarily involves the construction of health indicator (HI) to capture degradation information and the establishment of relationships between HI and RUL. However, most previous methods which require complex model structures or rich domain knowledge are not suitable for real industrial conditions with variable operating conditions. To address this challenge, a novel RUL prediction framework is proposed based on a degradation distribution transport health indicator (DDTHI) and a consolidated memory stabilized LSTM (CMsLSTM). First, a new degradation data distribution transport matrix is proposed, requiring no prior domain knowledge, to characterize the transformation process between degradation data distributions. Then, the HI at the current degradation time is constructed by minimizing the distribution transport cost. To streamline the prognostic architecture while preserving degradation information retention capability, a consolidated memory stabilized LSTM is designed via optimizing the internal structure of neurons and relearning the relationship between different time points. It can further store historical knowledge across time scales and provide more information for the RUL prediction. Finally, the effectiveness of the proposed method is verified by a real motor bearing dataset of a combustion fan in a hot strip mill and a public bearing dataset.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113039"},"PeriodicalIF":7.9,"publicationDate":"2025-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517833","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Comprehensive tribodynamic analysis of lubricated gear contacts for direct identification of gear whine noise in electric drive units 为直接识别电传动装置齿轮啸叫噪声，对润滑齿轮接触进行了全面的摩擦动力学分析

IF 7.9 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2025-06-30 DOI: 10.1016/j.ymssp.2025.113054

Hamza Mughal , Nader Dolatabadi , Ramin Rahmani , Paul King , Max Gnanakumarr , Olivier Varnier

{"title":"Comprehensive tribodynamic analysis of lubricated gear contacts for direct identification of gear whine noise in electric drive units","authors":"Hamza Mughal , Nader Dolatabadi , Ramin Rahmani , Paul King , Max Gnanakumarr , Olivier Varnier","doi":"10.1016/j.ymssp.2025.113054","DOIUrl":"10.1016/j.ymssp.2025.113054","url":null,"abstract":"<div><div>This study focuses on gear whine noise of electric drive units (EDUs) with helical gear transmission. A combined analytical and experimental approach is adopted to better understand dynamic transmission error (DTE) as the root cause of emitted gear whine noise. A fully analytical gear model based on potential energy method is developed to calculate time-varying meshing stiffness (TVMS). A lubricated contact model based on identification of regimes of lubrication is applied to precisely account for contact stiffness and damping variations. It is observed that local variations in lubricated contact generate undulations in DTE and dynamic load signatures. The frequency of the undulations, as observed from the comparison between the measured acceleration and sound pressure signals and the predicted DTEs, is determined by the resonance between the meshing frequency (including its multiples) and the harmonic frequencies of the gear pair. Greater amplitudes of undulations in lubricated contact stiffness are observed, followed by excessive vibrations in DTE, when multiples of meshing frequency coincide with gear pair harmonics and super-harmonics simultaneously. The superimposition effect exacerbates the vibrations indicating possibility of higher gear whine noise. Lower gearbox and meshing orders propagate as airborne noise, whilst higher frequency multiples are more noticeable as structural vibrations. Such comprehensive analysis linking the emitted gear whine noise and vibrations to the changes in the regime of lubrication in gear contacts is novel contribution of this research and has not hitherto been reported for helical gear whine noise and EDU NVH analyses.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113054"},"PeriodicalIF":7.9,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Physics informed Long Short-Term Memory neural network for dual state-parameter estimation of linear dynamical systems robust to input forces 基于物理信息的长短期记忆神经网络用于对输入力具有鲁棒性的线性动力系统的双状态参数估计

IF 7.9 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2025-06-30 DOI: 10.1016/j.ymssp.2025.113012

Nikhil Mahar , Subhamoy Sen , Laurent Mevel

{"title":"Physics informed Long Short-Term Memory neural network for dual state-parameter estimation of linear dynamical systems robust to input forces","authors":"Nikhil Mahar , Subhamoy Sen , Laurent Mevel","doi":"10.1016/j.ymssp.2025.113012","DOIUrl":"10.1016/j.ymssp.2025.113012","url":null,"abstract":"<div><div>Structural Health Monitoring (SHM) of structures is significantly challenged by the presence of unknown input forces, e.g, external unmeasured and non-stationary excitations applied to the system such as wind or loads, which hinder accurate damage assessment and system identification. Traditional model-based approaches face inherent limitations related to model inversion, observability, and identifiability, all of which are impaired by the need for input force knowledge. Data-driven methods offer advantages in terms of scalability and automation but still depend on known inputs and often lack physical interpretability. To address these issues, an input-robust Physics-Informed Long Short-Term Memory (rPi-LSTM) framework is introduced, integrating input-robust physical modeling of system dynamics with the temporal learning capabilities of LSTM networks. The framework employs an output injection strategy to reject unknown input forces, enabling estimation of system states and spatial health parameters without requiring input force information. By preserving temporal dependencies with the LSTM network, an aspect that is often neglected in conventional physics-informed networks, the method ensures stable and accurate system estimation while complying with the system physics. Validation through numerical simulations and real laboratory-scale experiments confirms its robustness to input forces, noise, data sparsity, and varying damage scenarios, demonstrating strong potential for real-world SHM applications under uncertain conditions.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113012"},"PeriodicalIF":7.9,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517213","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A digital twin-enabled domain adaptation network for cross-space fault diagnosis of roller bearings 滚子轴承跨空间故障诊断的数字双域自适应网络

IF 7.9 1区工程技术

Mechanical Systems and Signal Processing Pub Date : 2025-06-30 DOI: 10.1016/j.ymssp.2025.113053

Congcong Fang , Qi Chang , Xiuyuan Hu , Wei Zhou , Xianghui Meng

{"title":"A digital twin-enabled domain adaptation network for cross-space fault diagnosis of roller bearings","authors":"Congcong Fang , Qi Chang , Xiuyuan Hu , Wei Zhou , Xianghui Meng","doi":"10.1016/j.ymssp.2025.113053","DOIUrl":"10.1016/j.ymssp.2025.113053","url":null,"abstract":"<div><div>The accuracy of roller bearing diagnosis is crucial for ensuring the reliability and safety of mechanical systems. Data-driven methods, such as deep transfer learning (DTL), have been widely applied in fault diagnosis. However, the performance of these models is still limited in industrial scenarios due to the scarcity of labeled data for roller bearings and the complex operational conditions. This study introduces a digital twin-enabled domain adaptation network (DTDA) for the fault diagnosis of roller bearings across digital and physical space. In the digital space, a numerical model of a cylindrical roller bearing with its support housing is created based on the Augmented Lagrange multibody dynamics methodology. Typical fault modes, including raceway defects and cage pillar fracture, are accurately described in this model. A large amount of labeled pedestal vibration signals for corresponding defective roller bearings is then generated. For the physical space, defective roller bearings are machined, and a series of bench tests are carried out to obtain the vibration data of the bearing pedestal. A closed-set domain adaptation network based on DTL is developed to minimize discrepancies in data distribution from these two spaces. A bearing cross-space diagnosis task is constructed using labeled simulation data and unlabeled measurement data. The proposed digital twin-enabled fault diagnosis framework is experimentally validated, and the results demonstrate its superiority over the latest published methods.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113053"},"PeriodicalIF":7.9,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0