Mechanical Systems and Signal Processing最新文献_第7页

Synchronous phase-switching mechanism of a 2-DOF self-excited and forced combined excitation nonlinear mechanical resonator 二自由度自激与强制组合激励非线性机械谐振器的同步换相机理

IF 8.9 1区工程技术 Q1 ENGINEERING, MECHANICAL

Mechanical Systems and Signal Processing

Pub Date : 2026-01-27 DOI: 10.1016/j.ymssp.2026.113927

Lei Li , Peiyuan Tang , Taiyu Li , Chuansen Zhang , Yan Qiao , Wenming Zhang

The synchronous vibration of mechanical resonators is of great significance for improving the frequency stability and measurement resolution of resonant devices. Mechanical resonators can exhibit both in-phase synchronous and out-of-phase synchronous vibrations, and their switching behavior is highly susceptible to various driving parameters and nonlinear factors. In this study, a two-degree-of-freedom (2-DOF) magnetic coupled resonant system driven by both self-excitation and forced excitation is designed, which can realize the synchronous vibration and phase-switching of two resonators by adjusting the feedback phase, self-excitation driving force and forced driving force. Firstly, a 2-DOF self-excited and forced combined excitation (SFCE) dynamic model is proposed, and experimental measurements are carried out. It is found that the synchronous bandwidth of the coupled resonator depends on the feedback phase and driving intensity of the resonant system. Interestingly, the resonant system exhibits two typical switching phenomena. One is the switching phenomenon among dual-frequency (DF), single-frequency (SF) and multi-frequency (MF) responses, and the other is the transition between in-phase and out-of-phase synchronous vibrations. Specifically, when the vibration of the forced-driven resonator disappears, the synchronous phenomenon of the two resonators changes from in-phase vibration to out-of-phase vibration. The phase-switching conditions between in-phase vibration and out-of-phase vibration are theoretically derived and experimentally verified. Furthermore, based on the phase jump caused by the synchronous phase-switching mechanism, the potential application of mechanical resonator in mass and pressure warning is explored. The research results provide a theoretical and experimental basis for the application of synchronous vibration in mechanical resonators.

机械谐振器的同步振动对提高谐振装置的频率稳定性和测量分辨率具有重要意义。机械谐振器既有同相同步振动，也有非同相同步振动，其开关特性极易受到各种驱动参数和非线性因素的影响。本研究设计了一种自激和强制激励两种驱动方式的二自由度磁耦合谐振系统，通过调节反馈相位、自激驱动力和强制驱动力，实现两个谐振器的同步振动和换相。首先，建立了二自由度自激与强制组合激励（SFCE）动力学模型，并进行了实验测量。研究发现，耦合谐振腔的同步带宽取决于谐振系统的反馈相位和驱动强度。有趣的是，谐振系统表现出两种典型的开关现象。一种是双频（DF）、单频（SF）和多频（MF）响应之间的切换现象，另一种是同相和异相同步振动之间的转换。具体来说，当强制驱动谐振腔的振动消失后，两个谐振腔的同步现象由同相振动变为同相振动。从理论上推导了同相振动与非同相振动的切换条件，并进行了实验验证。此外，基于同步换相机制引起的相位跳变，探讨了机械谐振器在质量和压力预警中的潜在应用。研究结果为同步振动在机械谐振器中的应用提供了理论和实验依据。

{"title":"Synchronous phase-switching mechanism of a 2-DOF self-excited and forced combined excitation nonlinear mechanical resonator","authors":"Lei Li , Peiyuan Tang , Taiyu Li , Chuansen Zhang , Yan Qiao , Wenming Zhang","doi":"10.1016/j.ymssp.2026.113927","DOIUrl":"10.1016/j.ymssp.2026.113927","url":null,"abstract":"<div><div>The synchronous vibration of mechanical resonators is of great significance for improving the frequency stability and measurement resolution of resonant devices. Mechanical resonators can exhibit both in-phase synchronous and out-of-phase synchronous vibrations, and their switching behavior is highly susceptible to various driving parameters and nonlinear factors. In this study, a two-degree-of-freedom (2-DOF) magnetic coupled resonant system driven by both self-excitation and forced excitation is designed, which can realize the synchronous vibration and phase-switching of two resonators by adjusting the feedback phase, self-excitation driving force and forced driving force. Firstly, a 2-DOF self-excited and forced combined excitation (SFCE) dynamic model is proposed, and experimental measurements are carried out. It is found that the synchronous bandwidth of the coupled resonator depends on the feedback phase and driving intensity of the resonant system. Interestingly, the resonant system exhibits two typical switching phenomena. One is the switching phenomenon among dual-frequency (DF), single-frequency (SF) and multi-frequency (MF) responses, and the other is the transition between in-phase and out-of-phase synchronous vibrations. Specifically, when the vibration of the forced-driven resonator disappears, the synchronous phenomenon of the two resonators changes from in-phase vibration to out-of-phase vibration. The phase-switching conditions between in-phase vibration and out-of-phase vibration are theoretically derived and experimentally verified. Furthermore, based on the phase jump caused by the synchronous phase-switching mechanism, the potential application of mechanical resonator in mass and pressure warning is explored. The research results provide a theoretical and experimental basis for the application of synchronous vibration in mechanical resonators.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"246 ","pages":"Article 113927"},"PeriodicalIF":8.9,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146047935","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

Baseline-free local stress measurement based on the multi-directional polarization characteristics of zero-group velocity Lamb waves 基于零群速度兰姆波多向极化特征的无基线局部应力测量

IF 8.9 1区工程技术 Q1 ENGINEERING, MECHANICAL

Mechanical Systems and Signal Processing

Pub Date : 2026-01-27 DOI: 10.1016/j.ymssp.2026.113887

Tianchen Sheng , Yichao Tang , Hao Cong , Jiaxin Li , WeiJia Shi , Xinqi Tian , Bo Zhao , Jiubin Tan

Localized stress concentrations in critical structural regions pose serious risks to the integrity and safety of components in aerospace, transportation, and wind power applications. Conventional ultrasonic stress measurement techniques, such as time-of-flight (TOF) methods, provide only averaged stress information along the wave path and suffer from limited spatial resolution. Importantly, these methods often rely on baseline data, which can be difficult to obtain or affected by environmental variations. To address these limitations, this study proposes a baseline-free localized stress monitoring method based on stress-induced multi-directional frequency shifts of zero-group-velocity (ZGV) Lamb wave modes, ensuring greater flexibility and robustness in real-world applications. Based on the anisotropy caused by stress, a piezoelectric (PZT) transducer array is employed to capture ZGV responses in multiple directions, and the frequency differences among these directions are fused to estimate localized stress. Experimental validation demonstrates that the proposed method achieves a maximum absolute error of 3.3 MPa within a stress range of 20–100 MPa, over a distance of 7.5 mm, under the condition of eliminating reference signal dependence. Compared with conventional TOF-based approaches, the method offers significantly enhanced spatial resolution and measurement accuracy, highlighting its potential for high-precision local stress monitoring in critical structural components.

关键结构区域的局部应力集中对航空航天、交通运输和风力发电应用中部件的完整性和安全性构成严重威胁。传统的超声应力测量技术，如飞行时间（TOF）方法，只能提供沿波程的平均应力信息，并且空间分辨率有限。重要的是，这些方法通常依赖于基线数据，而这些数据很难获得或受到环境变化的影响。为了解决这些限制，本研究提出了一种基于应力诱导的零群速度（ZGV）兰姆波模式多向频移的无基线局部应力监测方法，确保在实际应用中具有更大的灵活性和鲁棒性。基于应力引起的各向异性，采用压电换能器阵列捕获多个方向的ZGV响应，并融合各方向之间的频率差来估计局部应力。实验验证表明，在消除参考信号依赖的条件下，在20 ~ 100 MPa的应力范围内，在7.5 mm的距离内，该方法的最大绝对误差为3.3 MPa。与传统的基于tof的方法相比，该方法具有显著提高的空间分辨率和测量精度，突出了其在关键结构部件高精度局部应力监测中的潜力。

{"title":"Baseline-free local stress measurement based on the multi-directional polarization characteristics of zero-group velocity Lamb waves","authors":"Tianchen Sheng , Yichao Tang , Hao Cong , Jiaxin Li , WeiJia Shi , Xinqi Tian , Bo Zhao , Jiubin Tan","doi":"10.1016/j.ymssp.2026.113887","DOIUrl":"10.1016/j.ymssp.2026.113887","url":null,"abstract":"<div><div>Localized stress concentrations in critical structural regions pose serious risks to the integrity and safety of components in aerospace, transportation, and wind power applications. Conventional ultrasonic stress measurement techniques, such as time-of-flight (TOF) methods, provide only averaged stress information along the wave path and suffer from limited spatial resolution. Importantly, these methods often rely on baseline data, which can be difficult to obtain or affected by environmental variations. To address these limitations, this study proposes a baseline-free localized stress monitoring method based on stress-induced multi-directional frequency shifts of zero-group-velocity (ZGV) Lamb wave modes, ensuring greater flexibility and robustness in real-world applications. Based on the anisotropy caused by stress, a piezoelectric (PZT) transducer array is employed to capture ZGV responses in multiple directions, and the frequency differences among these directions are fused to estimate localized stress. Experimental validation demonstrates that the proposed method achieves a maximum absolute error of 3.3 MPa within a stress range of 20–100 MPa, over a distance of 7.5 mm, under the condition of eliminating reference signal dependence. Compared with conventional TOF-based approaches, the method offers significantly enhanced spatial resolution and measurement accuracy, highlighting its potential for high-precision local stress monitoring in critical structural components.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"246 ","pages":"Article 113887"},"PeriodicalIF":8.9,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146047938","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 Multi-scale attention fusion framework based on KAN for rolling mill fault diagnosis 基于KAN的多尺度注意力融合框架轧机故障诊断

IF 8.9 1区工程技术 Q1 ENGINEERING, MECHANICAL

Mechanical Systems and Signal Processing

Pub Date : 2026-01-27 DOI: 10.1016/j.ymssp.2026.113931

Peiming Shi , Junjie He , Xuefang Xu

Three challenges remain in the area of rolling mill fault diagnosis: (1) Existing deep learning methods often struggle to extract key fault information; (2) Redundant fault features can interfere with models; (3) The existing multi-source information fusion models seldom study the complementarity of different modal characteristics. These challenges not only lead to an increase in defective steel products but also may cause production accidents during the operation, ultimately affecting the normal production of steel by enterprises. To solve these challenges, the study proposes a multi-scale attention fusion framework based on KAN (MAFF-KAN) for rolling mill fault diagnosis. Firstly, the dual shock feature approach is proposed to extract shock characteristics from signals. Secondly, KAN is first used in the field of multi-sensor-based rolling mill fault diagnosis to boost the effectiveness of the model. Additionally, KAN is used to develop the feature enhancement module that improves the rolling mill’s fault characteristics. Then, a multi-level parallel attention fusion improvement structure is suggested to achieve incremental fusion and utilization from shallow to deep fusion characteristics. The effectiveness of the suggested approach is validated by two case studies. The findings demonstrate that MAFF-KAN is superior to existing fault diagnosis methods and has better diagnostic capability.

轧机故障诊断存在三大挑战：(1)现有深度学习方法难以提取关键故障信息；(2)冗余故障特征会干扰模型；(3)现有的多源信息融合模型很少研究不同模态特征的互补性。这些挑战不仅导致不良钢材数量增加，而且在生产过程中还可能发生生产事故，最终影响企业钢材的正常生产。针对这些问题，研究提出了一种基于KAN的多尺度注意力融合框架（MAFF-KAN）用于轧机故障诊断。首先，提出了双冲击特征提取方法，从信号中提取冲击特征。其次，首次将KAN应用于基于多传感器的轧机故障诊断领域，提高了模型的有效性。此外，还利用KAN开发了改善轧机故障特征的特征增强模块。在此基础上，提出了多级并行注意力融合改进结构，实现从浅到深的融合特征增量融合和利用。通过两个案例研究验证了所建议方法的有效性。结果表明，MAFF-KAN优于现有的故障诊断方法，具有更好的诊断能力。

{"title":"A Multi-scale attention fusion framework based on KAN for rolling mill fault diagnosis","authors":"Peiming Shi , Junjie He , Xuefang Xu","doi":"10.1016/j.ymssp.2026.113931","DOIUrl":"10.1016/j.ymssp.2026.113931","url":null,"abstract":"<div><div>Three challenges remain in the area of rolling mill fault diagnosis: (1) Existing deep learning methods often struggle to extract key fault information; (2) Redundant fault features can interfere with models; (3) The existing multi-source information fusion models seldom study the complementarity of different modal characteristics. These challenges not only lead to an increase in defective steel products but also may cause production accidents during the operation, ultimately affecting the normal production of steel by enterprises. To solve these challenges, the study proposes a multi-scale attention fusion framework based on KAN (MAFF-KAN) for rolling mill fault diagnosis. Firstly, the dual shock feature approach is proposed to extract shock characteristics from signals. Secondly, KAN is first used in the field of multi-sensor-based rolling mill fault diagnosis to boost the effectiveness of the model. Additionally, KAN is used to develop the feature enhancement module that improves the rolling mill’s fault characteristics. Then, a multi-level parallel attention fusion improvement structure is suggested to achieve incremental fusion and utilization from shallow to deep fusion characteristics. The effectiveness of the suggested approach is validated by two case studies. The findings demonstrate that MAFF-KAN is superior to existing fault diagnosis methods and has better diagnostic capability.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"246 ","pages":"Article 113931"},"PeriodicalIF":8.9,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072619","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

Identification of complete vibration mode shapes of beams from moving load-induced responses measured by spatiotemporally sparse-sampled videos 由时空稀疏采样视频测量的移动荷载引起的响应中梁的完整振型的识别

IF 8.9 1区工程技术 Q1 ENGINEERING, MECHANICAL

Mechanical Systems and Signal Processing

Pub Date : 2026-01-27 DOI: 10.1016/j.ymssp.2026.113920

Tong Wu , Liang Tang , Lin Lang , Jun Xu , Quan Yuan , Fengli Zhou , Zhixiang Zhou

The full-field vibration modes can provide sufficient information for structural health monitoring (SHM). In general, the spatial resolution and orders of identified modes are decided by the number of sensors and sampling rate, respectively, resulting in incomplete mode identification by the conventional methods. This work developed a novel response decomposition method for complete modes identification of beam-like structures from the moving load-induced responses measured by sparse-sampled video both in spatial and temporal dimensions. Specially, the higher-order full-field modes can be obtained by analyzing a small number of points in videos recorded at sampling frequencies lower than the requirements of the Shannon-Nyquist sampling theorem. To establish a theoretical foundation, the structural responses of multi-span beams subjected to a moving concentrated force under various boundary conditions were initially analyzed. Subsequently, a novel response decomposition method was presented to identify complete modes from the computer vision measurements. Specifically, the spatial-sparse modes were identified from the vision-based displacements by the blind source separation (BSS) method. The displacements were subsequently converted into modal coordinates and employed to identify the complete modes after filtering out the dynamic components and normalizing the data. The proposed method is validated by numerical simulations on multi-span beams with different boundary conditions, a dynamic load test on a simply supported beam in the laboratory, and a field test on a two-span continuous beam bridge. The results demonstrate that the identified modes matched well with those obtained by traditional methods, while the spatial resolutions depended on the speed of the moving load and the displacement sampling frequency.

全场振动模态可以为结构健康监测提供充分的信息。通常，识别模式的空间分辨率和阶数分别由传感器数量和采样率决定，导致传统方法识别模式不完全。本文提出了一种新的响应分解方法，用于从空间和时间维度的稀疏采样视频测量的移动荷载引起的响应中识别梁状结构的完整模态。特别是，通过分析在低于Shannon-Nyquist采样定理要求的采样频率下录制的视频中的少量点，可以获得高阶的全场模式。为建立理论基础，对不同边界条件下多跨梁在运动集中力作用下的结构响应进行了初步分析。随后，提出了一种新的响应分解方法，从计算机视觉测量数据中识别完整模态。具体来说，利用盲源分离（BSS）方法从基于视觉的位移中识别出空间稀疏模式。然后将位移转换为模态坐标，过滤出动态分量并对数据进行归一化后用于识别完整模态。通过不同边界条件下多跨梁的数值模拟、简支梁的室内动载试验和两跨连续梁桥的现场试验，验证了所提方法的有效性。结果表明，所识别的模态与传统方法的模态吻合较好，但空间分辨率取决于移动载荷的速度和位移采样频率。

{"title":"Identification of complete vibration mode shapes of beams from moving load-induced responses measured by spatiotemporally sparse-sampled videos","authors":"Tong Wu , Liang Tang , Lin Lang , Jun Xu , Quan Yuan , Fengli Zhou , Zhixiang Zhou","doi":"10.1016/j.ymssp.2026.113920","DOIUrl":"10.1016/j.ymssp.2026.113920","url":null,"abstract":"<div><div>The full-field vibration modes can provide sufficient information for structural health monitoring (SHM). In general, the spatial resolution and orders of identified modes are decided by the number of sensors and sampling rate, respectively, resulting in incomplete mode identification by the conventional methods. This work developed a novel response decomposition method for complete modes identification of beam-like structures from the moving load-induced responses measured by sparse-sampled video both in spatial and temporal dimensions. Specially, the higher-order full-field modes can be obtained by analyzing a small number of points in videos recorded at sampling frequencies lower than the requirements of the Shannon-Nyquist sampling theorem. To establish a theoretical foundation, the structural responses of multi-span beams subjected to a moving concentrated force under various boundary conditions were initially analyzed. Subsequently, a novel response decomposition method was presented to identify complete modes from the computer vision measurements. Specifically, the spatial-sparse modes were identified from the vision-based displacements by the blind source separation (BSS) method. The displacements were subsequently converted into modal coordinates and employed to identify the complete modes after filtering out the dynamic components and normalizing the data. The proposed method is validated by numerical simulations on multi-span beams with different boundary conditions, a dynamic load test on a simply supported beam in the laboratory, and a field test on a two-span continuous beam bridge. The results demonstrate that the identified modes matched well with those obtained by traditional methods, while the spatial resolutions depended on the speed of the moving load and the displacement sampling frequency.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"246 ","pages":"Article 113920"},"PeriodicalIF":8.9,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146072621","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

Adaptive filtering and multi-scale MixStyle network for cross-domain fault diagnosis of marine electric thruster bearings 基于自适应滤波和多尺度MixStyle网络的船舶电动推力器轴承跨域故障诊断

IF 8.9 1区工程技术 Q1 ENGINEERING, MECHANICAL

Mechanical Systems and Signal Processing

Pub Date : 2026-01-27 DOI: 10.1016/j.ymssp.2026.113910

Chenxing Sheng , Meng Zhang , Xiang Rao , Mei Huang , Xueqin Zhang

Variations in operational conditions and noise interference deteriorate the performance of intelligent fault diagnosis models for marine electric thruster bearings. To address this challenge, we propose an adaptive filtering and multi-scale mixStyle network for cross-domain fault diagnosis of marine electric thruster bearings (AFMM). First, the multiscale attention-based MixStyle framework with multi-scale attention is developed based on MixStyle, which explicitly combines multi-scale decomposition and attention mechanisms to capture hierarchical fault patterns. Furthermore, to mitigate noise amplification in random feature mixing, a Hybrid Adaptive Filter is proposed, combining the frequency-domain selectivity of Butterworth filters with the dynamic adaptability of deep neural networks. Additionally, a source domain prototype refinement strategy is developed and combined with JMMD to achieve cross-domain fault diagnosis. Finally, extensive validation experiments are conducted by integrating the developed marine electric thruster simulation platform. Experimental results demonstrate that the proposed method exhibits significant noise robustness and cross-domain fault diagnosis accuracy.

船舶电动推力器轴承运行工况的变化和噪声干扰会降低其智能故障诊断模型的性能。为了解决这一挑战，我们提出了一种自适应滤波和多尺度mixStyle网络用于船用电动推力器轴承（AFMM）的跨域故障诊断。首先，在MixStyle的基础上开发了基于多尺度注意力的MixStyle框架，该框架明确地将多尺度分解和注意力机制结合起来，捕获分层故障模式；在此基础上，结合巴特沃斯滤波器的频域选择性和深度神经网络的动态适应性，提出了一种混合自适应滤波器，以减轻随机特征混合中的噪声放大。在此基础上，提出了一种源域原型细化策略，并结合JMMD实现了跨域故障诊断。最后，结合所开发的船用电动推力器仿真平台进行了大量的验证实验。实验结果表明，该方法具有较好的噪声鲁棒性和跨域故障诊断精度。

{"title":"Adaptive filtering and multi-scale MixStyle network for cross-domain fault diagnosis of marine electric thruster bearings","authors":"Chenxing Sheng , Meng Zhang , Xiang Rao , Mei Huang , Xueqin Zhang","doi":"10.1016/j.ymssp.2026.113910","DOIUrl":"10.1016/j.ymssp.2026.113910","url":null,"abstract":"<div><div>Variations in operational conditions and noise interference deteriorate the performance of intelligent fault diagnosis models for marine electric thruster bearings. To address this challenge, we propose an adaptive filtering and multi-scale mixStyle network for cross-domain fault diagnosis of marine electric thruster bearings (AFMM). First, the multiscale attention-based MixStyle framework with multi-scale attention is developed based on MixStyle, which explicitly combines multi-scale decomposition and attention mechanisms to capture hierarchical fault patterns. Furthermore, to mitigate noise amplification in random feature mixing, a Hybrid Adaptive Filter is proposed, combining the frequency-domain selectivity of Butterworth filters with the dynamic adaptability of deep neural networks. Additionally, a source domain prototype refinement strategy is developed and combined with JMMD to achieve cross-domain fault diagnosis. Finally, extensive validation experiments are conducted by integrating the developed marine electric thruster simulation platform. Experimental results demonstrate that the proposed method exhibits significant noise robustness and cross-domain fault diagnosis accuracy.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"246 ","pages":"Article 113910"},"PeriodicalIF":8.9,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146047936","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 novel online milling chatter detection using natural observation filters and mean filter index 基于自然观测滤波器和平均滤波指数的铣削颤振在线检测方法

IF 8.9 1区工程技术 Q1 ENGINEERING, MECHANICAL

Mechanical Systems and Signal Processing

Pub Date : 2026-01-26 DOI: 10.1016/j.ymssp.2026.113905

Khairul Jauhari, Achmad Zaki Rahman, Fitriana Nur Hasanah Aji Pramesti, Sri Kliwati, Wahyu Widada, Mahfudz Al Huda

Chatter detection plays a critical role in modern milling operations, as regenerative vibrations can severely degrade surface quality, accelerate tool wear, and destabilize the cutting process. Although existing techniques such as Fast Fourier Transform (FFT), Short-Time Fourier Transform (STFT), and wavelet-based analysis have been widely adopted, their performance is often limited by high computational demand and reduced effectiveness when dealing with rapidly changing or non-stationary signals. To overcome these limitations, this study introduces a novel time-domain chatter detection approach based on the Natural Observation Filter (NOF). The method decomposes vibration signals using lightweight recursive filters and employs a Mean Filter Index (MFI) to capture energy shifts associated with transitions from stable cutting to chatter. The proposed framework is validated through both numerical simulations and controlled milling experiments using a 3-axis CNC machine. Results show that the method can accurately distinguish stable, transition, and chatter states even under varying dynamic conditions. With a computational complexity of O(M·N), the NOF algorithm achieves ultra-low processing latency, enabling real-time deployment on low-power embedded platforms such as microcontrollers. These advantages highlight its potential for practical, industry-scale chatter monitoring and integration into intelligent machining systems.

颤振检测在现代铣削作业中起着至关重要的作用，因为再生振动会严重降低表面质量，加速刀具磨损，并使切削过程不稳定。尽管现有的快速傅立叶变换（FFT）、短时傅立叶变换（STFT）和基于小波的分析等技术已被广泛采用，但在处理快速变化或非平稳信号时，它们的性能往往受到高计算需求和效率降低的限制。为了克服这些局限性，本研究提出了一种基于自然观测滤波器（NOF）的时域颤振检测方法。该方法使用轻量级递归滤波器分解振动信号，并采用平均滤波指数（MFI）捕获与从稳定切割到颤振转变相关的能量转移。通过三轴数控机床的数值模拟和控制铣削实验验证了所提出的框架。结果表明，该方法在不同的动态条件下也能准确地区分稳态、过渡态和颤振态。NOF算法的计算复杂度为0 (M·N)，实现了超低的处理延迟，可在微控制器等低功耗嵌入式平台上实时部署。这些优点突出了它在实际工业规模颤振监测和集成到智能加工系统中的潜力。

{"title":"A novel online milling chatter detection using natural observation filters and mean filter index","authors":"Khairul Jauhari, Achmad Zaki Rahman, Fitriana Nur Hasanah Aji Pramesti, Sri Kliwati, Wahyu Widada, Mahfudz Al Huda","doi":"10.1016/j.ymssp.2026.113905","DOIUrl":"10.1016/j.ymssp.2026.113905","url":null,"abstract":"<div><div>Chatter detection plays a critical role in modern milling operations, as regenerative vibrations can severely degrade surface quality, accelerate tool wear, and destabilize the cutting process. Although existing techniques such as Fast Fourier Transform (FFT), Short-Time Fourier Transform (STFT), and wavelet-based analysis have been widely adopted, their performance is often limited by high computational demand and reduced effectiveness when dealing with rapidly changing or non-stationary signals. To overcome these limitations, this study introduces a novel time-domain chatter detection approach based on the Natural Observation Filter (NOF). The method decomposes vibration signals using lightweight recursive filters and employs a Mean Filter Index (MFI) to capture energy shifts associated with transitions from stable cutting to chatter. The proposed framework is validated through both numerical simulations and controlled milling experiments using a 3-axis CNC machine. Results show that the method can accurately distinguish stable, transition, and chatter states even under varying dynamic conditions. With a computational complexity of <em>O</em>(<em>M</em>·<em>N</em>), the NOF algorithm achieves ultra-low processing latency, enabling real-time deployment on low-power embedded platforms such as microcontrollers. These advantages highlight its potential for practical, industry-scale chatter monitoring and integration into intelligent machining systems.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"246 ","pages":"Article 113905"},"PeriodicalIF":8.9,"publicationDate":"2026-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146047937","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

Advances and prospects of phase space reconstruction-based high-dimensional signal processing for fault diagnosis: A systematic review 基于相空间重构的高维信号处理技术在故障诊断中的研究进展与展望

IF 8.9 1区工程技术 Q1 ENGINEERING, MECHANICAL

Mechanical Systems and Signal Processing

Pub Date : 2026-01-24 DOI: 10.1016/j.ymssp.2026.113919

Yong Lv , Hengyu Liu , Rui Yuan , Xun Dong , Yifei Wang , Hao Wu , David Chelidze

High-dimensional signal representations have demonstrated significant advantages in fault diagnosis by effectively capturing the nonlinear, nonstationary, and dynamic behaviors inherent in complex industrial systems. Phase space reconstruction (PSR) has emerged as a powerful framework for constructing high-dimensional trajectories from one-dimensional time series, enabling the recovery of intrinsic system dynamics without full observability. With increasing demand for accurate, interpretable, and robust diagnostic tools, PSR-based techniques have been widely explored across various engineering domains. However, a comprehensive review that synthesizes theoretical advances and practical implementations of PSR in fault diagnosis remains lacking. This review systematically analyzes the development of PSR-based high-dimensional signal processing methods for fault diagnosis. Existing approaches are categorized according to their integration with PSR into seven representative frameworks, i.e., entropy analysis, ergodic theory, tensor decomposition, manifold learning, trajectory morphology, complex networks, and AI-based approaches. Theoretical foundations, engineering applications, and diagnostic performance of these PSR-integrated frameworks are then reviewed and compared. Applications of PSR-based methods are further summarized in key domains such as rotating machinery, electrical systems, and structural components, highlighting their adaptability. Based on this review, several key challenges are identified, including embedding parameter selection, signal distortion during reconstruction, real-time deployment, and interpretability in AI models. To address these issues, future research is expected to focus on adaptive embedding strategies, interpretable hybrid frameworks, and efficient implementations to enable scalable and intelligent industrial diagnostics.

高维信号表示通过有效捕获复杂工业系统中固有的非线性、非平稳和动态行为，在故障诊断中显示出显著的优势。相空间重建（PSR）已经成为从一维时间序列构建高维轨迹的强大框架，可以在没有完全可观测性的情况下恢复系统的内在动力学。随着对准确、可解释和健壮的诊断工具的需求不断增加，基于psr的技术在各个工程领域得到了广泛的探索。然而，综合PSR在故障诊断中的理论进展和实际应用的全面综述仍然缺乏。本文系统地分析了基于psr的高维信号处理方法在故障诊断中的研究进展。现有的方法根据其与PSR的集成分为七个代表性框架，即熵分析、遍历理论、张量分解、流形学习、轨迹形态、复杂网络和基于人工智能的方法。然后对这些psr集成框架的理论基础、工程应用和诊断性能进行了审查和比较。进一步总结了基于psr的方法在旋转机械、电气系统和结构部件等关键领域的应用，强调了它们的适应性。基于这一综述，确定了几个关键挑战，包括嵌入参数选择、重建过程中的信号失真、实时部署和人工智能模型的可解释性。为了解决这些问题，未来的研究预计将集中在自适应嵌入策略、可解释的混合框架和高效实现上，以实现可扩展和智能的工业诊断。

{"title":"Advances and prospects of phase space reconstruction-based high-dimensional signal processing for fault diagnosis: A systematic review","authors":"Yong Lv , Hengyu Liu , Rui Yuan , Xun Dong , Yifei Wang , Hao Wu , David Chelidze","doi":"10.1016/j.ymssp.2026.113919","DOIUrl":"10.1016/j.ymssp.2026.113919","url":null,"abstract":"<div><div>High-dimensional signal representations have demonstrated significant advantages in fault diagnosis by effectively capturing the nonlinear, nonstationary, and dynamic behaviors inherent in complex industrial systems. Phase space reconstruction (PSR) has emerged as a powerful framework for constructing high-dimensional trajectories from one-dimensional time series, enabling the recovery of intrinsic system dynamics without full observability. With increasing demand for accurate, interpretable, and robust diagnostic tools, PSR-based techniques have been widely explored across various engineering domains. However, a comprehensive review that synthesizes theoretical advances and practical implementations of PSR in fault diagnosis remains lacking. This review systematically analyzes the development of PSR-based high-dimensional signal processing methods for fault diagnosis. Existing approaches are categorized according to their integration with PSR into seven representative frameworks, i.e., entropy analysis, ergodic theory, tensor decomposition, manifold learning, trajectory morphology, complex networks, and AI-based approaches. Theoretical foundations, engineering applications, and diagnostic performance of these PSR-integrated frameworks are then reviewed and compared. Applications of PSR-based methods are further summarized in key domains such as rotating machinery, electrical systems, and structural components, highlighting their adaptability. Based on this review, several key challenges are identified, including embedding parameter selection, signal distortion during reconstruction, real-time deployment, and interpretability in AI models. To address these issues, future research is expected to focus on adaptive embedding strategies, interpretable hybrid frameworks, and efficient implementations to enable scalable and intelligent industrial diagnostics.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"246 ","pages":"Article 113919"},"PeriodicalIF":8.9,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039780","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 phenomenological model for planetary gear crack fault by integrating meshing impact and dynamic load sharing response 基于啮合冲击和动载荷共享响应的行星齿轮裂纹故障现象学模型

IF 8.9 1区工程技术 Q1 ENGINEERING, MECHANICAL

Mechanical Systems and Signal Processing

Pub Date : 2026-01-24 DOI: 10.1016/j.ymssp.2026.113908

Hongxiang Jing , Dong Zhen , Guojin Feng , Libin Zang , Zhe Cheng , Niaoqing Hu , Fengshou Gu

Vibration in planetary gear train originates from meshing excitation, with local tooth cracks introducing impact forces that further alter load sharing and complicate the dynamic responses. Existing phenomenological models generally assume uniform load sharing and fail to capture the time-series relationship between fault impacts and the meshing process. This study developed a novel phenomenological model that considers the impacts and dynamic load redistribution caused by cracks in planetary gear teeth. A time-varying mesh stiffness model was employed to analyze stiffness variations caused by cracks of different sizes in planetary gear teeth. Based on meshing phase relationships, the temporal link between fault impact force and the meshing cycle was derived, and a meshing impact function was formulated. A load floating response factor was then introduced to describe the influence of excitation force variation on load sharing. These elements were integrated into the model to investigate vibration responses under sun–planet and ring–planet meshing conditions. Simulation and experimental results show that fault impacts generate transient excitations and enhance higher-order modulation sidebands. While internal and external meshing produce similar spectral pattern, it shows differences in amplitude. The vibration signals under different fault sizes were analyzed using statistical indicators, including the sideband index, zero-order figure of merit, and spectral kurtosis. The results demonstrate the effectiveness of the proposed model in reflecting the fault evolution trend.

行星轮系的振动源于啮合激励，局部齿裂纹引入的冲击力进一步改变了负载分担并使动态响应复杂化。现有的现象学模型一般假设负荷均匀分担，无法捕捉故障冲击与网格划分过程之间的时间序列关系。本研究建立了一种新的现象模型，该模型考虑了行星齿轮齿面裂纹引起的冲击和动载荷重分布。采用时变啮合刚度模型分析了行星齿轮齿面不同尺寸裂纹对啮合刚度的影响。基于啮合相位关系，推导了故障冲击力与啮合周期的时间联系，建立了啮合冲击函数。然后引入负载浮动响应因子来描述激振力变化对负载分担的影响。将这些因素整合到模型中，研究太阳行星和环行星网格条件下的振动响应。仿真和实验结果表明，故障冲击产生瞬态激励，增强了高阶调制边带。内外网格的谱图相似，但幅值不同。采用边带指数、零阶优值和谱峭度等统计指标对不同故障大小下的振动信号进行分析。结果表明，该模型能较好地反映断层演化趋势。

{"title":"A phenomenological model for planetary gear crack fault by integrating meshing impact and dynamic load sharing response","authors":"Hongxiang Jing , Dong Zhen , Guojin Feng , Libin Zang , Zhe Cheng , Niaoqing Hu , Fengshou Gu","doi":"10.1016/j.ymssp.2026.113908","DOIUrl":"10.1016/j.ymssp.2026.113908","url":null,"abstract":"<div><div>Vibration in planetary gear train originates from meshing excitation, with local tooth cracks introducing impact forces that further alter load sharing and complicate the dynamic responses. Existing phenomenological models generally assume uniform load sharing and fail to capture the time-series relationship between fault impacts and the meshing process. This study developed a novel phenomenological model that considers the impacts and dynamic load redistribution caused by cracks in planetary gear teeth. A time-varying mesh stiffness model was employed to analyze stiffness variations caused by cracks of different sizes in planetary gear teeth. Based on meshing phase relationships, the temporal link between fault impact force and the meshing cycle was derived, and a meshing impact function was formulated. A load floating response factor was then introduced to describe the influence of excitation force variation on load sharing. These elements were integrated into the model to investigate vibration responses under sun–planet and ring–planet meshing conditions. Simulation and experimental results show that fault impacts generate transient excitations and enhance higher-order modulation sidebands. While internal and external meshing produce similar spectral pattern, it shows differences in amplitude. The vibration signals under different fault sizes were analyzed using statistical indicators, including the sideband index, zero-order figure of merit, and spectral kurtosis. The results demonstrate the effectiveness of the proposed model in reflecting the fault evolution trend.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"246 ","pages":"Article 113908"},"PeriodicalIF":8.9,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039849","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

An integrated monitoring method for dam displacement considering model deficiencies 考虑模型缺陷的大坝位移综合监测方法

IF 8.9 1区工程技术 Q1 ENGINEERING, MECHANICAL

Mechanical Systems and Signal Processing

Pub Date : 2026-01-24 DOI: 10.1016/j.ymssp.2026.113921

Zhenxiang Jiang, Hui Chen, Luwan Chen

The displacement monitoring method, comprising both a monitoring model and monitoring indicators, is a crucial tool for assessing the health status of dams. High-precision displacement monitoring methods can effectively detect abnormal dam conditions and trigger alarms, thereby alerting management authorities to undertake necessary maintenance actions. The advancement of automated monitoring technologies has provided abundant foundational data for dam surveillance. However, traditional monitoring methods often overlook the inherent limitations of modeling algorithms when processing large volumes of data, which may lead to false alarms in practical engineering applications and consequently reduce the operational management efficiency of dams. To enhance the reliability of monitoring outcomes, this study first discusses the deficiencies of traditional modeling algorithms and, based on this analysis, proposes a multi-model integrated monitoring approach. Specifically, Support Vector Machine (SVM) and Autoregressive Moving Average (ARMA) models are separately utilized to establish monitoring models. On this basis, a distribution analysis of the joint residuals from the two types of models is conducted, revealing that the joint residuals exhibit weak correlation in the non-tail regions and strong correlation in the tail regions. Subsequently, the copula function is employed to fit the joint residuals, and a novel joint monitoring indicator based on the cumulative distribution function is proposed, enabling multi-model integrated monitoring of dam displacement. Case studies demonstrate that the integrated monitoring method significantly reduces the frequency of false alarms and can provide valuable reference for improving dam safety management.

位移监测方法包括监测模型和监测指标，是评价大坝健康状况的重要工具。高精度位移监测方法可以有效地发现大坝异常状况并触发警报，从而提醒管理部门采取必要的维护措施。自动化监测技术的进步，为大坝监测提供了丰富的基础数据。然而，传统的监测方法在处理大量数据时往往忽略了建模算法的固有局限性，在实际工程应用中可能出现虚警，从而降低大坝的运行管理效率。为了提高监测结果的可靠性，本研究首先讨论了传统建模算法的不足，并在此基础上提出了一种多模型集成监测方法。具体而言，分别利用支持向量机（SVM）和自回归移动平均（ARMA）模型建立监测模型。在此基础上，对两类模型的联合残差进行分布分析，发现联合残差在非尾区呈弱相关，在尾区呈强相关。随后，利用copula函数对联合残差进行拟合，提出了一种基于累积分布函数的联合监测指标，实现了大坝位移的多模型综合监测。实例研究表明，该综合监测方法可显著降低误报频率，为提高大坝安全管理水平提供有价值的参考。

{"title":"An integrated monitoring method for dam displacement considering model deficiencies","authors":"Zhenxiang Jiang, Hui Chen, Luwan Chen","doi":"10.1016/j.ymssp.2026.113921","DOIUrl":"10.1016/j.ymssp.2026.113921","url":null,"abstract":"<div><div>The displacement monitoring method, comprising both a monitoring model and monitoring indicators, is a crucial tool for assessing the health status of dams. High-precision displacement monitoring methods can effectively detect abnormal dam conditions and trigger alarms, thereby alerting management authorities to undertake necessary maintenance actions. The advancement of automated monitoring technologies has provided abundant foundational data for dam surveillance. However, traditional monitoring methods often overlook the inherent limitations of modeling algorithms when processing large volumes of data, which may lead to false alarms in practical engineering applications and consequently reduce the operational management efficiency of dams. To enhance the reliability of monitoring outcomes, this study first discusses the deficiencies of traditional modeling algorithms and, based on this analysis, proposes a multi-model integrated monitoring approach. Specifically, Support Vector Machine (SVM) and Autoregressive Moving Average (ARMA) models are separately utilized to establish monitoring models. On this basis, a distribution analysis of the joint residuals from the two types of models is conducted, revealing that the joint residuals exhibit weak correlation in the non-tail regions and strong correlation in the tail regions. Subsequently, the copula function is employed to fit the joint residuals, and a novel joint monitoring indicator based on the cumulative distribution function is proposed, enabling multi-model integrated monitoring of dam displacement. Case studies demonstrate that the integrated monitoring method significantly reduces the frequency of false alarms and can provide valuable reference for improving dam safety management.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"246 ","pages":"Article 113921"},"PeriodicalIF":8.9,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039781","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

Modeling and comprehensive analysis of parameters on a compact pendulum-based electromagnetic energy harvester 小型摆式电磁能量采集器的建模与参数综合分析

IF 8.9 1区工程技术 Q1 ENGINEERING, MECHANICAL

Mechanical Systems and Signal Processing

Pub Date : 2026-01-24 DOI: 10.1016/j.ymssp.2026.113909

Qitao Lu , Mingjing Cai , Xin Li , Junyi Cao , Wei-Hsin Liao

Pendulum-based electromagnetic energy harvesters (PEEHs) convert human motion, particularly walking, into electrical power through intricate electromechanical coupling to power Internet of Things devices. Despite their potential, most studies represent PEEHs using equivalent or simplified models, which often limit dynamic analysis or result in significant computational errors. To address this limitation, we develop an accurate electromechanical coupling model for a compact PEEH incorporating a compound planetary gear train. The magnetic field is evaluated using the magnetic scalar potential method, while system dynamics are formulated via the Euler–Lagrange approach. Magnetic field predictions are validated against finite element simulations in COMSOL. Under geometric constraints, extensive parameter sweeps—including swing frequency, swing amplitude, gear ratio, load resistance, and the pendulum-to-rotor moment of inertia ratio—are performed through numerical simulations. Experimental studies further validate the numerical results. Both simulation and experimental results demonstrate that PEEH performance converges to a local optimum under fixed conditions, and that increasing the moment of inertia ratio enhances performance under parameter optimization. The developed prototype achieves a maximum output power of 18.70 mW and a normalized power density of 29.92 W/(m³·Hz·°), confirming the proposed model’s efficacy to guide the design of high-performance energy harvesters.

基于钟摆的电磁能量采集器（PEEHs）通过复杂的机电耦合将人体运动（特别是行走）转换为电能，为物联网设备供电。尽管它们具有潜力，但大多数研究使用等效或简化的模型来表示PEEHs，这往往限制了动态分析或导致显著的计算误差。为了解决这一限制，我们开发了一个精确的机电耦合模型的紧凑型PEEH结合复合行星齿轮系。磁场计算采用磁标量势法，系统动力学计算采用欧拉-拉格朗日方法。通过COMSOL有限元模拟验证了磁场预测结果。在几何约束下，通过数值模拟进行了广泛的参数扫描，包括摆频、摆幅、传动比、负载阻力和摆转子惯性矩比。实验研究进一步验证了数值结果。仿真和实验结果均表明，在固定条件下，PEEH性能收敛于局部最优；在参数优化条件下，增大惯性矩比可提高PEEH性能。所开发的样机最大输出功率为18.70 mW，归一化功率密度为29.92 W/（m3·Hz·°），证实了所提出模型指导高性能能量采集器设计的有效性。

{"title":"Modeling and comprehensive analysis of parameters on a compact pendulum-based electromagnetic energy harvester","authors":"Qitao Lu , Mingjing Cai , Xin Li , Junyi Cao , Wei-Hsin Liao","doi":"10.1016/j.ymssp.2026.113909","DOIUrl":"10.1016/j.ymssp.2026.113909","url":null,"abstract":"<div><div>Pendulum-based electromagnetic energy harvesters (PEEHs) convert human motion, particularly walking, into electrical power through intricate electromechanical coupling to power Internet of Things devices. Despite their potential, most studies represent PEEHs using equivalent or simplified models, which often limit dynamic analysis or result in significant computational errors. To address this limitation, we develop an accurate electromechanical coupling model for a compact PEEH incorporating a compound planetary gear train. The magnetic field is evaluated using the magnetic scalar potential method, while system dynamics are formulated via the Euler–Lagrange approach. Magnetic field predictions are validated against finite element simulations in COMSOL. Under geometric constraints, extensive parameter sweeps—including swing frequency, swing amplitude, gear ratio, load resistance, and the pendulum-to-rotor moment of inertia ratio—are performed through numerical simulations. Experimental studies further validate the numerical results. Both simulation and experimental results demonstrate that PEEH performance converges to a local optimum under fixed conditions, and that increasing the moment of inertia ratio enhances performance under parameter optimization. The developed prototype achieves a maximum output power of 18.70 mW and a normalized power density of 29.92 W/(m<sup>3</sup>·Hz·°), confirming the proposed model’s efficacy to guide the design of high-performance energy harvesters.</div></div>","PeriodicalId":51124,"journal":{"name":"Mechanical Systems and Signal Processing","volume":"246 ","pages":"Article 113909"},"PeriodicalIF":8.9,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039848","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