Mechanical Systems and Signal Processing最新文献

{"title":"Research on phenomenological vibration signal model for reflecting the fault feature sidebands of crack on planetary carrier plate","authors":"Xiaosong Lin ,&nbsp;Niaoqing Hu ,&nbsp;Yi Yang ,&nbsp;Zhengyang Yin ,&nbsp;Zuanbo Zhou ,&nbsp;Yuehao Li ,&nbsp;Zihao Deng","doi":"10.1016/j.ymssp.2025.113044","DOIUrl":"10.1016/j.ymssp.2025.113044","url":null,"abstract":"<div><div>The detection of early fatigue cracks in the high-stress regions of planetary gear carrier represents a significant challenge in the field of helicopter fault diagnosis. As an essential tool for fault diagnosis, the frequency spectrum of vibration signals contains abundant fault feature information. However, owing to insufficient fault mechanism analysis, some key spectral features remain unexplored and underutilized. Considering the influence of carrier crack-induced stiffness variation and carrier eccentricity on the vibration characteristics of the planetary gear system, a novel phenomenological vibration signal model is established by comprehensively incorporating the carrier eccentricity modulation function, transfer path function, excitation force function, and projection direction function. By this model, a new sideband modulation phenomenon of planetary gear distributed fault frequency induced by planetary carrier crack is revealed. The newly discovered phenomenon shows that planetary gear distributed fault characteristic frequency appears in the vibration signal spectrum of planetary carrier crack as both carrier frequency and modulation frequency. Finally, the experimental results prove the effectiveness of the proposed model, which offers a novel perspective for analyzing the fault mechanism of crack on carrier plate.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113044"},"PeriodicalIF":7.9,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517309","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":"Analytical and experimental study on the bi-misaligned lubrication characteristics of a novel grooved bearing","authors":"Zhongliang Xie ,&nbsp;Ming Yang ,&nbsp;Wenjun Gao ,&nbsp;Bin Zhao ,&nbsp;Peng Du","doi":"10.1016/j.ymssp.2025.113050","DOIUrl":"10.1016/j.ymssp.2025.113050","url":null,"abstract":"<div><div>Lubrication states of water-lubricated grooved bearing are unclear, revised models considering bi-misaligned status and turbulent effect are established. Effects of bi-misaligned status on the internal fluid dynamics and how they further act on lubrication characteristics are thoroughly analyzed. Influences of velocity, external load, clearance ratio and groove number on key parameters under different misaligned conditions are probed. Bi-misalignment leads to uneven increase of film thickness and pressure. It affects the interface lubrication effect and operation stability. When the number of grooves is 5, the multiple dynamic parameters reach extreme values. This study provides valuable references for the understanding of bi-misaligned status on the lubrication characteristics for such grooved bearings.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113050"},"PeriodicalIF":7.9,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517832","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":"Research on feedrate intelligent control method for machining efficiency","authors":"Yeming Jiang ,&nbsp;Kuo Liu ,&nbsp;Jiadong Huang ,&nbsp;Di Zhao ,&nbsp;Pengxiang Gao ,&nbsp;Mingyu Li ,&nbsp;Haibo Liu ,&nbsp;Yongqing Wang","doi":"10.1016/j.ymssp.2025.113037","DOIUrl":"10.1016/j.ymssp.2025.113037","url":null,"abstract":"<div><div>Traditional constant feed machining methods have several limitations, including the lack of online adjustment capabilities for machining parameters and an over-reliance on operator experience. Additionally, for workpieces with uneven allowance and high removal rates, the efficiency of traditional constant feed machining is notably low. To address these issues, this paper investigates a feedrate adaptive control method aimed at improving machining efficiency. Firstly, we analyze the signal data collected by the system. Using spindle system power as the signal for machining state monitoring, we study the composition of the collected spindle system power signals based on the energy transfer relationship of the spindle system. On this basis, we design a preprocessing algorithm for the power data, incorporating the concept of the command domain to mark the power data collected when the spindle speed is not constant. Secondly, a feedrate adaptive control method based on fuzzy theory is proposed to achieve adaptive control under constant power constraints. The proposed algorithm is validated through simulations, demonstrating good convergence. Furthermore, we design a control parameter self-adjustment algorithm. By constructing a reference model of the cutting process, this algorithm enables real-time adjustment of the fuzzy control parameters, ensuring the controller optimally performs at each stage of the machining process. Finally, experiments were conducted on an aerospace workpiece under both traditional constant feed machining mode and the proposed feedrate adaptive control machining mode. The results show that the proposed method improves machining efficiency by 23.1% and maintains spindle system power near the target power. This study is significant for enhancing CNC machining efficiency and promoting the intelligent transformation of the equipment manufacturing industry.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113037"},"PeriodicalIF":7.9,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144517214","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":"Neural networks with dual-view fusion feature learning mechanism: A method to improve micro-milling tool wear monitoring performance by enhancing feature generalization capabilities","authors":"Hongfei Wang ,&nbsp;Qingshun Bai ,&nbsp;Shandeng Chen ,&nbsp;Tingting Wang ,&nbsp;Wanmin Guo ,&nbsp;Yuhao Dou","doi":"10.1016/j.ymssp.2025.113038","DOIUrl":"10.1016/j.ymssp.2025.113038","url":null,"abstract":"<div><div>Tool wear monitoring methods can effectively monitor cutting tool condition, workpiece machining accuracy, and machined surface quality. Current tool wear monitoring models typically focus on repeatable sensitive features in the machining signal that are consistent with the wear trend, and can only monitor from a single view. However, in actual machining processes, especially in micro-milling where signal strength is weak and signals are not sensitive to wear, such a single-view monitoring approach struggles to handle the complex nonlinear variations in signals during actual machining. Therefore, this study innovatively proposes a tool wear monitoring model based on a dual-view fusion feature learning mechanism. The feature extraction capability of the model is enhanced by introducing a Siamese neural network. First, the Kolmogorov-Arnold neural network (KAN) is constructed to extract repeatable features that align with the tool wear trend from all signals, building the original feature vector. Secondly, the Siamese neural network is embedded to capture the feature differences between the current signal and the initial machining signal, and the feature difference vector is constructed. The feature difference vector effectively avoids interference of signals caused by sudden variation of cutting state and environment during the intermediate machining process, which enhances the differentiation of the machining signals under different wear states and provides a new training perspective for tool wear monitoring. Then, the original feature vectors are fused with the feature difference vectors to form a dual-view fusion feature with excellent generalization ability. The model not only captures global features in all signals, but also adaptively learns correlations and differences between signals, providing richer feature information for tool wear monitoring. To verify the superiority of the model, ablation experiments and micro-milling experiments are performed. The results show that the Siamese network effectively enhanced the feature extraction and generalization capabilities of models. The performance of models with dual-view features for tool wear monitoring is improved compared to standalone models. Among all models, the dual-view model based on the Siamese and the KAN network has the highest accuracy, with 97.95 % prediction accuracy. In addition, it is still highly accurate under new micro-milling conditions.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113038"},"PeriodicalIF":7.9,"publicationDate":"2025-06-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144510756","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":"Fast and high-resolution ultrasonic sparse TFM imaging based on CycleSR","authors":"Li Lin ,&nbsp;Jiachen Xiao ,&nbsp;Donghui Zhang ,&nbsp;Jingyu Liao ,&nbsp;Zhiyuan Ma","doi":"10.1016/j.ymssp.2025.113030","DOIUrl":"10.1016/j.ymssp.2025.113030","url":null,"abstract":"<div><div>The low imaging efficiency of the Total Focusing Method (TFM) limits its application in high real-time performance scenarios. This paper proposes a fast imaging framework for TFM that combines sparse array techniques with deep learning. First, a binary dung beetle optimization (BDBO) algorithm is employed to optimize the effective array number and layout, achieving a 75 % sparsity rate to reduce the amount of data acquisition. Then, an unpaired image super-resolution method, unsupervised image super-resolution with an indirect supervised path (CycleSR), utilizes simulation data for training. The CycleSR leverages sparse TFM-generated low-resolution (LR) images and synthetic LR images to train the unsupervised translation module, while using high-resolution images (HR) and synthetic LR images to train the super-resolution module, within a collaborative training framework to achieve sparse TFM image reconstruction. Experiments were conducted on carbon steel blocks with eight side drilled holes (SDHs) ranging in diameter from 0.8 mm to 3.0 mm, and the initial sparse TFM-generated LR images were reconstructed through the CycleSR. The experimental results show that the proposed DAS(Sparse+LR)+CycleSR method reduces imaging time from 21.06 s to 0.13 s, the maximum positioning error of the SDH is less than 0.3 mm, and the API error is less than 0.25. Compared to the representative unpaired super-resolution method, cycle-in-cycle generative adversarial network (CinCGAN), the proposed method achieves a peak signal-to-noise ratio (PSNR) of 39.32 dB and a learned perceptual image patch similarity (LPIPS) of 0.002, which are significantly superior to the 36.42 dB and 0.007 of the CinCGAN method, enabling fast and high-resolution imaging inspection.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113030"},"PeriodicalIF":7.9,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144502275","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":"Vibration energy analysis of rub-impact rotor system: Nonlinear fault-induced energy response and transfer state","authors":"Xiantao Zhang ,&nbsp;Yongfeng Yang ,&nbsp;Wangqun Deng ,&nbsp;Shibo Zhao ,&nbsp;Qingyu Zhu ,&nbsp;Hui Ma ,&nbsp;Qinkai Han ,&nbsp;Zhaoye Qin ,&nbsp;Wenkui Liu","doi":"10.1016/j.ymssp.2025.113035","DOIUrl":"10.1016/j.ymssp.2025.113035","url":null,"abstract":"<div><div>The nonlinear dynamic behavior induced by rub-impact faults constitutes a critical factor that constrains the safe and stable operation of rotor systems. Existing research primarily focuses on analyzing the characteristics of nonlinear responses, but rarely addresses the specific influence of such nonlinear behavior on the system performance. In this paper, the vibration energy method was introduced to investigate the nonlinear response caused by rub-impact of rotor system. The energy characteristics corresponding to different fault states were achieved, and the energy variation law induced by nonlinear bifurcation was elucidated. In addition, the energy transfer ratio was defined to quantify the system’s utilization ability of input energy, and the influence of key rub-impact parameters on vibration energy was analyzed. The results demonstrated that the rub-impact fault results in the dissipation of input energy while simultaneously enhancing the energy utilization efficiency of the system. Particularly, when fault induce instability in periodic motion, the energy utilization ability increases sharply, potentially threatening structural safety. Finally, the validity of the proposed method and theoretical analysis was experimentally confirmed. The results provide a theoretical foundation for understanding the mechanisms by which nonlinear responses affect rotor systems and for optimizing structural design.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113035"},"PeriodicalIF":7.9,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144502272","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 characterization modeling of space truss structures utilizing transfer matrix method","authors":"Liang Wang ,&nbsp;Rui Qi ,&nbsp;Yuning Ge ,&nbsp;Ziyu Shen ,&nbsp;Jiamei Jin ,&nbsp;Lusheng Yuan","doi":"10.1016/j.ymssp.2025.113048","DOIUrl":"10.1016/j.ymssp.2025.113048","url":null,"abstract":"<div><div>Space truss structures are integral components of spacecraft, and the accurate modeling of their dynamic response is crucial for meeting the rigorous requirements of space missions. Due to the large scale and inherent complexity of space truss structures, existing modeling methods often suffer from slow computational speed and limited accuracy. A novel dynamic response characterization modeling method for truss structures is proposed, integrating the finite element method and the transfer matrix method. First, the basic assumptions and boundary conditions for modeling are determined using the typical truss structure as the research object. Secondly, the transfer matrix is obtained through discretization and derivation, and the truss elements modeling is derived through the connection conditions. In addition, the modeling of the space truss structure system for dynamic response characterization is developed, incorporating connection attachments. Finally, the prototype was fabricated and the experimental system was constructed. The correctness of the proposed modeling method is verified by a comprehensive comparison of FEM simulations, proposed modeling computations and experimental measurements. The comprehensive comparison includes the frequency response, data analysis, correlation comparison and relative error comparison. The results demonstrate that the proposed method offers significant advantages in terms of both computational accuracy and efficiency, and it holds great potential for the dynamic characterization of complex space structures.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113048"},"PeriodicalIF":7.9,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144502276","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":"Trajectory Pre-filtering-Based cooperative tool servo diamond turning of Complex-Shaped optics","authors":"Hao Wu ,&nbsp;YiXuan Meng ,&nbsp;YuHan Niu ,&nbsp;XiangYuan Wang ,&nbsp;ZhiWei Zhu ,&nbsp;MingJun Ren ,&nbsp;XinQuan Zhang ,&nbsp;LiMin Zhu","doi":"10.1016/j.ymssp.2025.113056","DOIUrl":"10.1016/j.ymssp.2025.113056","url":null,"abstract":"<div><div>Traditional single slow slide servo (SSS) and fast tool servo (FTS) diamond turning processes face limitations due to the trade-off between stroke and bandwidth, which hinders high-precision and high-speed machining of complex-shaped optics with large depths. To overcome these limitations, this paper proposes a novel trajectory pre-filtering-based cooperative tool servo (TP-CTS) diamond turning process. The TP-CTS process utilizes a dual-stage feed drive system coordinated through a master–slave strategy, wherein the machine tool axis serves as the master servo and the fast tool axis as the slave servo. Tool trajectories are pre-filtered and assigned to the master servo, while the slave servo compensates in real time for deviations from the reference trajectory. The pre-filtering approach effectively bypasses the kinematic and dynamic constraints of the master servo, enabling full utilization of the dynamic capabilities of both servos. To further enhance system performance, an online data-driven control strategy is developed for the slave servo, integrating a Koopman operator-based feedforward controller with a high-bandwidth feedback controller to achieve near-zero phase-lag tracking. Experimental validation on a commercial lathe with a self-developed fast tool axis demonstrates that the TP-CTS process significantly improves machining efficiency while maintaining accuracy compared to the existing CTS method. The results highlight the substantial improvements offered by the proposed process, emphasizing its potential for high-precision, high-speed manufacturing of complex-shaped optics in industrial applications.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113056"},"PeriodicalIF":7.9,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144502274","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":"Investigation of vector force decoupling utilizing F&D collaborative data reconstruction strategy","authors":"Xinyang Li ,&nbsp;Zongjin Ren ,&nbsp;Wei Liu ,&nbsp;Mingyu Yuan ,&nbsp;Shilu Mai ,&nbsp;Shan Lin ,&nbsp;Jun Zhang ,&nbsp;Wei Zhang ,&nbsp;Wenju Sun","doi":"10.1016/j.ymssp.2025.113033","DOIUrl":"10.1016/j.ymssp.2025.113033","url":null,"abstract":"<div><div>During the measurement of vector engine thrust with a distributed test system, the deformation of the force-measuring units (FMUs) varies, resulting in significant differences in the generated coupling patterns under different vector forces. Consequently, the decoupling accuracy of the traditional orthogonal matrix linear decoupling method undergoes a substantial decrease. To address this issue, this paper introduces a force–displacement (F&amp;D) collaborative decoupling measurement method specifically designed for measuring all-angle vector forces using distributed measurement techniques. Firstly, simulation calculations are used to analyze the error source of vector force coupling when vector forces are applied at different angles. By analyzing the cause of coupling, a multi-layer coordinate system and its corresponding transformation relationship are derived through the placement of multiple measuring points within the test device. Then, the output of the force sensors is decomposed and reconstructed layer by layer according to the multilevel coordinate system transformation, and combined with the orthogonal linear matrix, the coupling error is eliminated under the multi-source interference. Finally, the experiment of unidirectional loading and vector loading is carried out. The results show that the amplitude error of vector force after decoupling is less than 0.95 %, and the angle error of vector force is less than 3.76 %. The error is reduced by over 50 % when compared to the linear decoupling method. The comparison with different decoupling methods shows that this method has the highest decoupling accuracy. This method greatly improves the coupling accuracy and provides a new solution for the accurate measurement of vector engine thrust.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 113033"},"PeriodicalIF":7.9,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144502385","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":"Progressive hybrid hypergraph attention network with channel information fusion for remaining useful life prediction of rolling bearings","authors":"Yuru Zhang ,&nbsp;Chun Su ,&nbsp;Xiaoliang He ,&nbsp;Jiuqiang Tang ,&nbsp;Mingjiang Xie ,&nbsp;Hao Liu","doi":"10.1016/j.ymssp.2025.112987","DOIUrl":"10.1016/j.ymssp.2025.112987","url":null,"abstract":"<div><div>Nowadays, remaining useful life (RUL) prediction of rolling bearings based on graph neural network has attracted extensive attention. However, most existing methods can only model pairwise correlations, while they have not yet concerned the complex higher-order relationships and lack the ability to learn global-local information. This paper proposes a progressive hybrid hypergraph attention network for RUL prediction of rolling bearings with multi-channel signals, where the pair-wise graphs and hypergraphs are incorporated to simultaneously capture high-order and low-order dependencies. Initially, the network constructs a dynamic graph structure to extract global to local temporal information with the progressive elimination of neighbor nodes. Meanwhile, the channel information is fused through the connectivity of hybrid hypergraph construction. Afterwards, node-level and hyperedge-level attention are emphasized to enhance the contribution of significant nodes. Eventually, the high-order and low-order features are blended via an adaptive fusion module. The experimental study on two benchmark datasets and a time-varying condition dataset of rolling bearings indicates the effectiveness, generalization, and comparable efficiency of the proposed approach. Besides, the impact of channel information variability on prediction results is explored with the considerable effectiveness of the fusion strategy.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"236 ","pages":"Article 112987"},"PeriodicalIF":7.9,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144502273","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}