Measurement最新文献_第5页

Deep learning-based phase retrieval with SE-SwinUNet architecture in fringe projection 3D measurement 基于SE-SwinUNet结构的深度学习相位检索在条纹投影三维测量中的应用

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-05-05 Epub Date: 2026-03-06 DOI: 10.1016/j.measurement.2026.121044

Qinkui Ma, Jianhua Wang, Huixin Sun, Tong Shao, Mengmeng Wang

Fringe projection three-dimensional (3D) measurement is an important non-contact measurement technique, and the precision of 3D reconstruction largely depends on the accuracy of phase retrieval. With the development of deep learning (DL), phase retrieval based on deep neural network (DNN) has been widely studied. The performance of neural networks plays a decisive role in determining the accuracy of phase demodulation in fringe projection profilometry (FPP). Currently, most deep learning-based wrapped phase extraction methods are built upon the U-Net architecture. Nevertheless, the hierarchical skip-connection mechanism of U-Net presents inherent limitations in global information transmission and feature fusion, which in turn restricts further improvements in phase retrieval accuracy. For this purpose, we propose a method for phase demodulation grounded in a novel multi-scale feature fusion network, referred to as SE-SwinUNet. The network combines the advantages of the Swin Transformer and residual connections, incorporating an asymmetric design in both the encoder and decoder. Through enhanced global information modeling and local detail refinement, it markedly improves the efficiency of feature propagation and utilization. Furthermore, by incorporating a channel attention mechanism (Squeeze-and-Excitation layer, SE), the network is capable of adaptively allocating appropriate weights to multi-scale features, thereby effectively reinforcing its focus on the most salient features. Experimental results demonstrate that SE-SwinUNet achieves higher accuracy in phase demodulation tasks compared to the conventional U-Net, exhibiting particularly pronounced advantages in complex scenarios.

条纹投影三维测量是一种重要的非接触式测量技术，三维重建的精度很大程度上取决于相位恢复的精度。随着深度学习技术的发展，基于深度神经网络（DNN）的相位检索技术得到了广泛的研究。在条纹投影轮廓术（FPP）中，神经网络的性能对相位解调精度起着决定性的作用。目前，大多数基于深度学习的包装相位提取方法都是建立在U-Net架构之上的。然而，U-Net的分层跳过连接机制在全局信息传输和特征融合方面存在固有的局限性，从而制约了相位检索精度的进一步提高。为此，我们提出了一种基于新型多尺度特征融合网络（SE-SwinUNet）的相位解调方法。该网络结合了Swin变压器和剩余连接的优点，在编码器和解码器中都采用了非对称设计。通过增强的全局信息建模和局部细节细化，显著提高了特征的传播和利用效率。此外，通过引入通道关注机制（Squeeze-and-Excitation layer， SE），该网络能够自适应地为多尺度特征分配适当的权重，从而有效地加强对最显著特征的关注。实验结果表明，与传统的U-Net相比，SE-SwinUNet在相位解调任务中具有更高的精度，在复杂场景中表现出特别明显的优势。

{"title":"Deep learning-based phase retrieval with SE-SwinUNet architecture in fringe projection 3D measurement","authors":"Qinkui Ma, Jianhua Wang, Huixin Sun, Tong Shao, Mengmeng Wang","doi":"10.1016/j.measurement.2026.121044","DOIUrl":"10.1016/j.measurement.2026.121044","url":null,"abstract":"<div><div>Fringe projection three-dimensional (3D) measurement is an important non-contact measurement technique, and the precision of 3D reconstruction largely depends on the accuracy of phase retrieval. With the development of deep learning (DL), phase retrieval based on deep neural network (DNN) has been widely studied. The performance of neural networks plays a decisive role in determining the accuracy of phase demodulation in fringe projection profilometry (FPP). Currently, most deep learning-based wrapped phase extraction methods are built upon the U-Net architecture. Nevertheless, the hierarchical skip-connection mechanism of U-Net presents inherent limitations in global information transmission and feature fusion, which in turn restricts further improvements in phase retrieval accuracy. For this purpose, we propose a method for phase demodulation grounded in a novel multi-scale feature fusion network, referred to as SE-SwinUNet. The network combines the advantages of the Swin Transformer and residual connections, incorporating an asymmetric design in both the encoder and decoder. Through enhanced global information modeling and local detail refinement, it markedly improves the efficiency of feature propagation and utilization. Furthermore, by incorporating a channel attention mechanism (Squeeze-and-Excitation layer, SE), the network is capable of adaptively allocating appropriate weights to multi-scale features, thereby effectively reinforcing its focus on the most salient features. Experimental results demonstrate that SE-SwinUNet achieves higher accuracy in phase demodulation tasks compared to the conventional U-Net, exhibiting particularly pronounced advantages in complex scenarios.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121044"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Torque stability of direct torque control in induction motors based on fuzzy PI adaptive control 基于模糊PI自适应控制的异步电动机直接转矩控制的转矩稳定性

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-05-05 Epub Date: 2026-03-04 DOI: 10.1016/j.measurement.2026.121024

Shuanglong Wu , Shubin Chen , Xiaoxing Ye , Jiajun Rao , Yijie He , Yushen Huang , Caixia Lin , Long Qi

To address the issues of large torque fluctuations, poor low-speed performance, and low control accuracy in traditional DTC for induction motors (IMs), this paper proposes a torque control strategy for IMs based on fuzzy logic control (FLC) combined with space vector pulse width modulation (SVM). First, SVM technology is introduced to optimize the stator flux vector control method, an SVM-DTC control system is designed to mitigate the significant torque fluctuations caused by hysteresis control. Subsequently, based on fuzzy control theory, a hybrid membership function fuzzy PI adaptive torque controller was designed. After parameter fitting, it is integrated into the SVM-DTC system to establish the FL-SVM-DTC control system. This approach mitigates inconsistent torque fluctuation suppression under varying load conditions and reduces excessive torque overshoot during load transients. Root Mean Squared Error (RMSE) is adopted as the torque evaluation metric. Finally, variable-load torque tests were conducted on the IM using an experimental motor test platform. Test results demonstrate that, under the torque controller designed in this paper, the FL-SVM-DTC achieves an average reduction of 70.46% in torque fluctuations compared to the traditional DTC, and an average reduction of 32.48% compared to the SVM-DTC. This validates the reliability and effectiveness of the proposed controller.

针对传统异步电机直接转矩控制存在转矩波动大、低速性能差、控制精度低等问题，提出了一种基于模糊逻辑控制（FLC）与空间矢量脉宽调制（SVM）相结合的异步电机转矩控制策略。首先，引入支持向量机技术对定子磁链矢量控制方法进行优化，设计了一种支持向量机-直接转矩控制系统，以缓解磁滞控制引起的显著转矩波动。随后，基于模糊控制理论，设计了一种混合隶属函数模糊PI自适应转矩控制器。经过参数拟合后，将其集成到SVM-DTC系统中，建立FL-SVM-DTC控制系统。这种方法减轻了在不同负载条件下不一致的转矩波动抑制，并减少了负载瞬态时过大的转矩超调。采用均方根误差（RMSE）作为扭矩评价指标。最后，利用实验电机测试平台对IM进行了变负载转矩测试。测试结果表明，在本文设计的转矩控制器下，FL-SVM-DTC的转矩波动比传统DTC平均降低70.46%，比SVM-DTC平均降低32.48%。这验证了所提出控制器的可靠性和有效性。

{"title":"Torque stability of direct torque control in induction motors based on fuzzy PI adaptive control","authors":"Shuanglong Wu , Shubin Chen , Xiaoxing Ye , Jiajun Rao , Yijie He , Yushen Huang , Caixia Lin , Long Qi","doi":"10.1016/j.measurement.2026.121024","DOIUrl":"10.1016/j.measurement.2026.121024","url":null,"abstract":"<div><div>To address the issues of large torque fluctuations, poor low-speed performance, and low control accuracy in traditional DTC for induction motors (IMs), this paper proposes a torque control strategy for IMs based on fuzzy logic control (FLC) combined with space vector pulse width modulation (SVM). First, SVM technology is introduced to optimize the stator flux vector control method, an SVM-DTC control system is designed to mitigate the significant torque fluctuations caused by hysteresis control. Subsequently, based on fuzzy control theory, a hybrid membership function fuzzy PI adaptive torque controller was designed. After parameter fitting, it is integrated into the SVM-DTC system to establish the FL-SVM-DTC control system. This approach mitigates inconsistent torque fluctuation suppression under varying load conditions and reduces excessive torque overshoot during load transients. Root Mean Squared Error (RMSE) is adopted as the torque evaluation metric. Finally, variable-load torque tests were conducted on the IM using an experimental motor test platform. Test results demonstrate that, under the torque controller designed in this paper, the FL-SVM-DTC achieves an average reduction of 70.46% in torque fluctuations compared to the traditional DTC, and an average reduction of 32.48% compared to the SVM-DTC. This validates the reliability and effectiveness of the proposed controller.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121024"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388354","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Robust five-degree-of-freedom error measurement in complex environments using composite sensing and SLM-based beam modulation 基于复合传感和slm波束调制的复杂环境下鲁棒五自由度误差测量

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-05-05 Epub Date: 2026-03-02 DOI: 10.1016/j.measurement.2026.121031

Chuanzhi Tang , Wenzheng Liu , Fajie Duan , Jinfeng Lan , Jinlai Zhang , Chenhui Deng , Jiawei Chu

The continuous advancement of high-end equipment manufacturing toward higher precision and enhanced robustness has imposed significant challenges on geometric error measurement under complex environments and long-range conditions. To address the significant impact of beam quality degradation on laser collimation measurement performance, a laser-based five-degree-of-freedom (5-DOF) error measurement method incorporating composite sensing and spatial light modulator (SLM)-based beam modulation is proposed in this paper for operation under complex conditions. A charge-coupled device (CCD) is employed as an auxiliary observation channel to enable real-time monitoring of critical beam characteristics, such as spot centroid drift, shape distortion, and energy non-uniformity. Combined with high-accuracy centroid data acquired from a quadrant photodiode (QPD) and a position-sensitive detector (PSD), a multi-sensor fusion-based composite perception model is established. On this basis, a beam deviation fusion estimation algorithm with dynamic weighting is developed to improve the accuracy and robustness of multi-source data integration. Furthermore, a dual-loop feedback control framework is designed to coordinate beam quality optimization and drift compensation, driving the SLM for composite phase modulation and achieving high-precision laser 5-DOF error measurement over extended distances. Experimental validation demonstrates that within a 3 m measurement range, the proposed method reduces translational drift from 37.31 μm to 3.38 μm and angular drift from 24.35 μm to 1.56 μm, enhances beam rotational symmetry to 0.949, and maintains energy concentration above 0.56. The proposed method has been validated to achieve high measurement precision, system stability, and environmental adaptability, providing a robust technical foundation for high-accuracy optical metrology under complex operational conditions.

高端装备制造不断向高精度和鲁棒性方向发展，对复杂环境和远程条件下的几何误差测量提出了重大挑战。针对光束质量退化对激光准直测量性能的显著影响，提出了一种结合复合传感和空间光调制器（SLM）光束调制的激光五自由度误差测量方法，用于复杂条件下的测量。采用电荷耦合器件（CCD）作为辅助观测通道，实时监测光斑质心漂移、形状畸变、能量不均匀性等光束关键特性。结合象限光电二极管（QPD）和位置敏感探测器（PSD）采集的高精度质心数据，建立了基于多传感器融合的复合感知模型。在此基础上，提出了一种动态加权波束偏差融合估计算法，提高了多源数据融合的精度和鲁棒性。此外，设计了双环反馈控制框架，协调光束质量优化和漂移补偿，驱动SLM进行复合相位调制，实现长距离高精度激光五自由度误差测量。实验验证表明，在3 m测量范围内，该方法将光束的平移漂移从37.31 μm减小到3.38 μm，角漂移从24.35 μm减小到1.56 μm，光束的旋转对称性提高到0.949，能量浓度保持在0.56以上。该方法具有较高的测量精度、系统稳定性和环境适应性，为复杂操作条件下的高精度光学计量提供了坚实的技术基础。

{"title":"Robust five-degree-of-freedom error measurement in complex environments using composite sensing and SLM-based beam modulation","authors":"Chuanzhi Tang , Wenzheng Liu , Fajie Duan , Jinfeng Lan , Jinlai Zhang , Chenhui Deng , Jiawei Chu","doi":"10.1016/j.measurement.2026.121031","DOIUrl":"10.1016/j.measurement.2026.121031","url":null,"abstract":"<div><div>The continuous advancement of high-end equipment manufacturing toward higher precision and enhanced robustness has imposed significant challenges on geometric error measurement under complex environments and long-range conditions. To address the significant impact of beam quality degradation on laser collimation measurement performance, a laser-based five-degree-of-freedom (5-DOF) error measurement method incorporating composite sensing and spatial light modulator (SLM)-based beam modulation is proposed in this paper for operation under complex conditions. A charge-coupled device (CCD) is employed as an auxiliary observation channel to enable real-time monitoring of critical beam characteristics, such as spot centroid drift, shape distortion, and energy non-uniformity. Combined with high-accuracy centroid data acquired from a quadrant photodiode (QPD) and a position-sensitive detector (PSD), a multi-sensor fusion-based composite perception model is established. On this basis, a beam deviation fusion estimation algorithm with dynamic weighting is developed to improve the accuracy and robustness of multi-source data integration. Furthermore, a dual-loop feedback control framework is designed to coordinate beam quality optimization and drift compensation, driving the SLM for composite phase modulation and achieving high-precision laser 5-DOF error measurement over extended distances. Experimental validation demonstrates that within a 3 m measurement range, the proposed method reduces translational drift from 37.31 μm to 3.38 μm and angular drift from 24.35 μm to 1.56 μm, enhances beam rotational symmetry to 0.949, and maintains energy concentration above 0.56. The proposed method has been validated to achieve high measurement precision, system stability, and environmental adaptability, providing a robust technical foundation for high-accuracy optical metrology under complex operational conditions.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121031"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387986","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Research of interactive sheep body size measurement based on 3D reconstruction 基于三维重建的交互式羊体尺寸测量方法研究

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-05-05 Epub Date: 2026-03-09 DOI: 10.1016/j.measurement.2026.121080

Lin Zhu , Lina Zhang , Fan Yang , Yuxing Wei , Hua Meng , Lu Yang , Kefan Shang , Jue Zhang , Xinhua Jiang

Body sizes of sheep serve as a comprehensive indicator of their morphological profile, growth characteristics, genetic traits, and so on. Vision-based technologies offer a non-invasive, efficient, and precise approach to livestock body size measurement. Although substantial progress has been achieved in this field, several critical challenges remain. Automated body measurement still faces challenges arising from livestock posture variability, as well as morphological differences across breeds and age groups. Moreover, the required measurement parameters may differ among breeds. To address these limitations, this study proposed an interactive body size measurement approach. A multi-view depth vision imaging system was employed to capture body images, followed by a denoising process to reduce noise arising from fleece heterogeneity and environmental interference. A three-dimensional spatial model of each sheep was then reconstructed, and pose normalization was performed to standardize the spatial orientation for consistent measurements. To improve accessibility and operational efficiency, an interactive body measurement system was developed on the PyQt5 platform. This system supports multi-dimensional morphometric measurement, including body height, width, length, chest girth, surface area, and volume. Farm experiments on sheep of different breeds and ages demonstrated that the proposed method provides high measurement accuracy, with an average relative error below 5%. The method is simple and user-friendly, exhibiting strong adaptability to various imaging conditions and sheep postures, thereby enabling diversified phenotypic characterization of sheep. It therefore serves as a valuable complement to existing approaches for sheep body measurement and demonstrates potential for extension to other livestock species, such as pigs and cattle.

羊体大小是羊的形态特征、生长特征、遗传性状等的综合指标。基于视觉的技术提供了一种非侵入性、高效和精确的牲畜体型测量方法。虽然在这一领域取得了重大进展，但仍然存在若干重大挑战。由于牲畜姿势的变化，以及不同品种和年龄组的形态差异，自动身体测量仍然面临挑战。此外，所需的测量参数可能因品种而异。为了解决这些局限性，本研究提出了一种交互式身体尺寸测量方法。采用多视角深度视觉成像系统采集人体图像，然后进行去噪处理，以降低羊毛异质性和环境干扰引起的噪声。然后重建每只羊的三维空间模型，并进行位姿归一化以标准化空间方向，以保证测量结果的一致性。为了提高可访问性和操作效率，在PyQt5平台上开发了一个交互式身体测量系统。该系统支持多维形态测量，包括身体高度、宽度、长度、胸围、表面积和体积。对不同品种和年龄的绵羊进行的农场实验表明，该方法具有较高的测量精度，平均相对误差在5%以下。该方法简单易用，对各种成像条件和羊的姿态具有较强的适应性，从而实现了羊的多样化表型表征。因此，它是对现有羊体测量方法的宝贵补充，并显示了推广到其他牲畜物种（如猪和牛）的潜力。

{"title":"Research of interactive sheep body size measurement based on 3D reconstruction","authors":"Lin Zhu , Lina Zhang , Fan Yang , Yuxing Wei , Hua Meng , Lu Yang , Kefan Shang , Jue Zhang , Xinhua Jiang","doi":"10.1016/j.measurement.2026.121080","DOIUrl":"10.1016/j.measurement.2026.121080","url":null,"abstract":"<div><div>Body sizes of sheep serve as a comprehensive indicator of their morphological profile, growth characteristics, genetic traits, and so on. Vision-based technologies offer a non-invasive, efficient, and precise approach to livestock body size measurement. Although substantial progress has been achieved in this field, several critical challenges remain. Automated body measurement still faces challenges arising from livestock posture variability, as well as morphological differences across breeds and age groups. Moreover, the required measurement parameters may differ among breeds. To address these limitations, this study proposed an interactive body size measurement approach. A multi-view depth vision imaging system was employed to capture body images, followed by a denoising process to reduce noise arising from fleece heterogeneity and environmental interference. A three-dimensional spatial model of each sheep was then reconstructed, and pose normalization was performed to standardize the spatial orientation for consistent measurements. To improve accessibility and operational efficiency, an interactive body measurement system was developed on the PyQt5 platform. This system supports multi-dimensional morphometric measurement, including body height, width, length, chest girth, surface area, and volume. Farm experiments on sheep of different breeds and ages demonstrated that the proposed method provides high measurement accuracy, with an average relative error below 5%. The method is simple and user-friendly, exhibiting strong adaptability to various imaging conditions and sheep postures, thereby enabling diversified phenotypic characterization of sheep. It therefore serves as a valuable complement to existing approaches for sheep body measurement and demonstrates potential for extension to other livestock species, such as pigs and cattle.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121080"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387726","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Dynamic modelling and vibration characteristic analysis of gear transmission system under the effect of time-varying misalignment from clearance of bearings 轴承间隙时变不对准影响下齿轮传动系统动力学建模及振动特性分析

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-05-05 Epub Date: 2026-02-25 DOI: 10.1016/j.measurement.2026.120961

Minmin Xu , Mingchun Wang , Yixuan Zeng , Yaoyao Han , Junming Huang , Wennian Yu , Fengshou Gu

Gear misalignment error is one of the most common errors in manufacturing or assembling process, which leads to uneven distribution of contact stress and increased vibration and noise due to the biased load. It also influences the diagnosis of gear local defects due to the coupling effect of them. To address the gear misalignment errors caused by discrepancies in bearing clearance at both ends of shaft, a coupled gear-shaft-bearing nonlinear dynamic model with 24 DOF is developed. Effect of time-varying misalignment errors on the characteristics of contact, stiffness and vibration are analyzed. Simulations are performed to evaluate the vibration response of the gear transmission system under the excitation of time-varying misalignment from bearing clearances and gear spall. Results indicate that discrepancies of bearing clearances reduce the time-varying meshing stiffness of the gear pair, which generates additional vibration and affects the vibration responses of the gear transmission system. Time-varying misalignment from bearing clearance also contributes the rising of shaft rotating frequency, which results in difficulty to detect early spalling faults on gear surface. Verification experiments are conducted to assess the precision of the enhanced gear-shaft-bearing model. New indicators are proposed to defect the change of gear misalignment error due to bearing clearances. The proposed coupling model contributes to a deeper insight on condition monitoring strategies and early fault diagnosis methodologies for gear transmission systems.

齿轮对中误差是制造或装配过程中最常见的误差之一，它会导致接触应力分布不均匀，并且由于偏载而增加振动和噪声。由于齿轮局部缺陷与齿轮局部缺陷的耦合作用，影响了齿轮局部缺陷的诊断。为了解决由于轴两端轴承间隙差异引起的齿轮对中误差，建立了24自由度齿轮-轴-轴承耦合非线性动力学模型。分析了时变误差对接触特性、刚度特性和振动特性的影响。对齿轮传动系统在轴承间隙和齿轮脱落等时变不对准激励下的振动响应进行了仿真研究。结果表明，轴承间隙的差异降低了齿轮副的时变啮合刚度，产生附加振动，影响齿轮传动系统的振动响应。由轴承间隙引起的时变对中偏差也会导致轴的旋转频率升高，从而导致齿轮表面剥落故障的早期检测困难。通过实验验证了改进后的齿轮轴-轴承模型的精度。提出了新的指标来缺陷由于轴承间隙引起的齿轮不对准误差的变化。提出的耦合模型有助于深入了解齿轮传动系统的状态监测策略和早期故障诊断方法。

{"title":"Dynamic modelling and vibration characteristic analysis of gear transmission system under the effect of time-varying misalignment from clearance of bearings","authors":"Minmin Xu , Mingchun Wang , Yixuan Zeng , Yaoyao Han , Junming Huang , Wennian Yu , Fengshou Gu","doi":"10.1016/j.measurement.2026.120961","DOIUrl":"10.1016/j.measurement.2026.120961","url":null,"abstract":"<div><div>Gear misalignment error is one of the most common errors in manufacturing or assembling process, which leads to uneven distribution of contact stress and increased vibration and noise due to the biased load. It also influences the diagnosis of gear local defects due to the coupling effect of them. To address the gear misalignment errors caused by discrepancies in bearing clearance at both ends of shaft, a coupled gear-shaft-bearing nonlinear dynamic model with 24 DOF is developed. Effect of time-varying misalignment errors on the characteristics of contact, stiffness and vibration are analyzed. Simulations are performed to evaluate the vibration response of the gear transmission system under the excitation of time-varying misalignment from bearing clearances and gear spall. Results indicate that discrepancies of bearing clearances reduce the time-varying meshing stiffness of the gear pair, which generates additional vibration and affects the vibration responses of the gear transmission system. Time-varying misalignment from bearing clearance also contributes the rising of shaft rotating frequency, which results in difficulty to detect early spalling faults on gear surface. Verification experiments are conducted to assess the precision of the enhanced gear-shaft-bearing model. New indicators are proposed to defect the change of gear misalignment error due to bearing clearances. The proposed coupling model contributes to a deeper insight on condition monitoring strategies and early fault diagnosis methodologies for gear transmission systems.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 120961"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388104","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Waveform-based clustering of fatigue cracking acoustic signatures 基于波形的疲劳裂纹声特征聚类

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-05-05 Epub Date: 2026-03-05 DOI: 10.1016/j.measurement.2026.120965

Théotime de la Selle , Stéphanie Deschanel , Jérome Weiss

Identifying signatures of fatigue crack growth and detecting early warnings in service in Non-Destructive Testing applications remain major challenges for research and industry alike. To address these issues, we propose a waveform-based signal processing technique applied to Acoustic Emission (AE) data. This study explores how a fully unsupervised, data-driven approach can cluster acoustic signals based on their physical emission mechanisms associated with fatigue cracking. Using AE datasets recorded during fatigue crack growth tests on compact tension specimens made of aluminum or steel alloys, we demonstrate that hierarchical clustering of acoustic multiplets – signatures of a unique source – is consistent with the phenomenological classification of multiplets previously established distinguishing multiplets generated by local contacts between crack surfaces from those associated with fatigue crack growth at crack tip. Here we show that we can assign physical emission mechanisms to each multiplet solely based on acoustic information through waveform-based dissimilarity measures — unlike standard AE signal processing techniques. Furthermore, by taking advantage of the multiple levels of analysis offered by hierarchical clustering, we significantly improve multiplets detection by reducing dependency on user-parameters. This improvement opens the possibility of developing a more automated Non-Destructive Testing architecture for fatigue cracking signatures detection, thus reducing computational complexity.

在无损检测应用中，识别疲劳裂纹扩展的特征和检测早期预警仍然是研究和工业领域面临的主要挑战。为了解决这些问题，我们提出了一种应用于声发射数据的基于波形的信号处理技术。本研究探索了一种完全无监督、数据驱动的方法如何基于与疲劳开裂相关的物理发射机制对声信号进行聚类。利用在铝或钢合金致密拉伸试样疲劳裂纹扩展试验中记录的声发射数据集，我们证明了多重声的分层聚类——一个独特来源的特征——与先前建立的多重声的现象学分类是一致的，该分类将裂纹表面之间的局部接触产生的多重声与裂纹尖端疲劳裂纹扩展产生的多重声区分开来。在这里，我们表明，我们可以通过基于波形的不相似性测量，仅根据声学信息为每个多路分配物理发射机制，这与标准的声发射信号处理技术不同。此外，通过利用分层聚类提供的多层次分析，我们通过减少对用户参数的依赖显著提高了多胞胎检测。这一改进开启了开发一种更自动化的无损检测体系结构用于疲劳裂纹特征检测的可能性，从而降低了计算复杂性。

{"title":"Waveform-based clustering of fatigue cracking acoustic signatures","authors":"Théotime de la Selle , Stéphanie Deschanel , Jérome Weiss","doi":"10.1016/j.measurement.2026.120965","DOIUrl":"10.1016/j.measurement.2026.120965","url":null,"abstract":"<div><div>Identifying signatures of fatigue crack growth and detecting early warnings in service in Non-Destructive Testing applications remain major challenges for research and industry alike. To address these issues, we propose a waveform-based signal processing technique applied to Acoustic Emission (AE) data. This study explores how a fully unsupervised, data-driven approach can cluster acoustic signals based on their physical emission mechanisms associated with fatigue cracking. Using AE datasets recorded during fatigue crack growth tests on compact tension specimens made of aluminum or steel alloys, we demonstrate that hierarchical clustering of acoustic multiplets – signatures of a unique source – is consistent with the phenomenological classification of multiplets previously established distinguishing multiplets generated by local contacts between crack surfaces from those associated with fatigue crack growth at crack tip. Here we show that we can assign physical emission mechanisms to each multiplet solely based on acoustic information through waveform-based dissimilarity measures — unlike standard AE signal processing techniques. Furthermore, by taking advantage of the multiple levels of analysis offered by hierarchical clustering, we significantly improve multiplets detection by reducing dependency on user-parameters. This improvement opens the possibility of developing a more automated Non-Destructive Testing architecture for fatigue cracking signatures detection, thus reducing computational complexity.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 120965"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Non-invasive rotor fault measurement in pumped storage systems via enhanced energy entropy analysis and hybrid deep learning with metrological validation 基于增强能量熵分析和混合深度学习的抽水蓄能系统转子无创故障测量

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-05-05 Epub Date: 2026-03-07 DOI: 10.1016/j.measurement.2026.121069

Qingqi Lan , Hui Sun , Kang Hu , Qiaorui Si , Yu Wu , Shouqi Yuan

To advance measurement methodologies for rotor health monitoring in pumped storage systems, this study proposes a novel non-invasive measurement framework integrating Motor Current Signature Analysis (MCSA) with hybrid deep learning. This method quantifies energy entropy shifts by employing Cyclic Autocorrelation Function (CAF) and Empirical Mode Decomposition (EMD), thereby addressing key metrological challenges including signal noise suppression and uncertainty propagation. Principal Component Analysis (PCA) is rigorously applied to reduce feature dimensionality while preserving measurement integrity. A novel SCSO-CNN-BiLSTM-Attention model is proposed, combining CNN’s spatial feature extraction, BiLSTM’s temporal dependency modeling, and attention-based critical weighting, optimized via Sand Cat Swarm Optimization (SCSO) for hyperparameter tuning with traceable uncertainty budgets. Validated on seven fault conditions, the model achieves 95.24% testing accuracy, outperforming traditional methods in both metrological robustness and computational efficiency. Key innovations include IMF1’s high-frequency sensitivity and IMF4’s low-frequency discriminability. T-SNE visualization confirms enhanced clustering through integrated RMS and IMF entropy features. Despite residual errors in cavitation-coupled faults, the framework demonstrates robust generalizability and real-time potential. This work provides a physics-informed AI solution for centrifugal pump health monitoring, balancing diagnostic precision with computational efficiency for industrial deployment.

为了推进抽水蓄能系统转子健康监测的测量方法，本研究提出了一种将电机电流特征分析（MCSA）与混合深度学习相结合的新型无创测量框架。该方法通过使用循环自相关函数（CAF）和经验模态分解（EMD）来量化能量熵移，从而解决了包括信号噪声抑制和不确定性传播在内的关键计量挑战。主成分分析（PCA）严格应用于降低特征维数的同时保持测量的完整性。提出了一种新的SCSO-CNN-BiLSTM- attention模型，该模型结合了CNN的空间特征提取、BiLSTM的时间依赖建模和基于注意力的临界加权，并通过沙猫群优化（SCSO）对具有可追溯不确定性预算的超参数调优进行了优化。在7种故障条件下验证，该模型的测试精度达到95.24%，在计量鲁棒性和计算效率方面均优于传统方法。关键的创新包括IMF1的高频灵敏度和IMF4的低频可分辨性。T-SNE可视化通过集成RMS和IMF熵特征证实了增强的聚类。尽管在空化耦合故障中存在残余误差，但该框架具有鲁棒的通用性和实时性。这项工作为离心泵健康监测提供了一种基于物理的人工智能解决方案，平衡了诊断精度和工业部署的计算效率。

{"title":"Non-invasive rotor fault measurement in pumped storage systems via enhanced energy entropy analysis and hybrid deep learning with metrological validation","authors":"Qingqi Lan , Hui Sun , Kang Hu , Qiaorui Si , Yu Wu , Shouqi Yuan","doi":"10.1016/j.measurement.2026.121069","DOIUrl":"10.1016/j.measurement.2026.121069","url":null,"abstract":"<div><div>To advance measurement methodologies for rotor health monitoring in pumped storage systems, this study proposes a novel non-invasive measurement framework integrating Motor Current Signature Analysis (MCSA) with hybrid deep learning. This method quantifies energy entropy shifts by employing Cyclic Autocorrelation Function (CAF) and Empirical Mode Decomposition (EMD), thereby addressing key metrological challenges including signal noise suppression and uncertainty propagation. Principal Component Analysis (PCA) is rigorously applied to reduce feature dimensionality while preserving measurement integrity. A novel SCSO-CNN-BiLSTM-Attention model is proposed, combining CNN’s spatial feature extraction, BiLSTM’s temporal dependency modeling, and attention-based critical weighting, optimized via Sand Cat Swarm Optimization (SCSO) for hyperparameter tuning with traceable uncertainty budgets. Validated on seven fault conditions, the model achieves 95.24% testing accuracy, outperforming traditional methods in both metrological robustness and computational efficiency. Key innovations include IMF1’s high-frequency sensitivity and IMF4’s low-frequency discriminability. T-SNE visualization confirms enhanced clustering through integrated RMS and IMF entropy features. Despite residual errors in cavitation-coupled faults, the framework demonstrates robust generalizability and real-time potential. This work provides a physics-informed AI solution for centrifugal pump health monitoring, balancing diagnostic precision with computational efficiency for industrial deployment.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121069"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388165","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Application of ensemble-based machine learning models for investigating the importance of asphalt pavement design factors on IRI 应用基于集成的机器学习模型研究沥青路面设计因素对IRI的重要性

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-05-05 Epub Date: 2026-03-09 DOI: 10.1016/j.measurement.2026.121087

Abdulraaof H. Al-Qaili, Abdullah I. Al-Mansour, Hamad Al-Solieman

Asphalt pavement design factors play a critical role in determining the quality and longevity of pavements. Understanding how these factors affect the International Roughness Index (IRI) is essential for optimizing pavement performance and improving road quality. In this study, the importance of asphalt pavement design factors on IRI investigate using ensemble-based machine learning models, providing valuable insights for pavement design and maintenance strategies.

Ensemble models, specifically Random Forest (RF) and Gradient Boosting (GB), were used to analyze the importance of asphalt pavement design factors on the IRI. These algorithms combine the predictive power of multiple decision trees to increase accuracy and identify critical variables that affect IRI.

The results of the study demonstrated the influence of hyperparameters on model performance and highlighted the importance of parameter tuning for optimal results. Additionally, the study emphasizes the critical role of factors such as temperature, traffic volume, and pavement age in predicting IRI across different models. SHAP analysis revealed the primary factors influencing IRI, conforming the importance of environmental conditions and material properties.

This study provide valuable insights into the relationships between asphalt pavement design factors and IRI, offering a roadmap for evidence-based decision making in pavement design and management. The findings can guide stakeholders in prioritizing design factors to optimize pavement quality and overall road performance.

沥青路面设计因素对沥青路面的质量和寿命起着至关重要的作用。了解这些因素如何影响国际粗糙度指数（IRI）对于优化路面性能和改善道路质量至关重要。在本研究中，使用基于集成的机器学习模型研究沥青路面设计因素对IRI的重要性，为路面设计和维护策略提供有价值的见解。采用随机森林（Random Forest， RF）和梯度增强（Gradient Boosting， GB）集成模型分析沥青路面设计因素对IRI的影响。这些算法结合了多个决策树的预测能力，以提高准确性并识别影响IRI的关键变量。研究结果证明了超参数对模型性能的影响，并强调了参数调整对优化结果的重要性。此外，该研究还强调了温度、交通量和路面年龄等因素在不同模型中预测IRI的关键作用。SHAP分析揭示了影响IRI的主要因素，符合环境条件和材料性能的重要性。本研究为沥青路面设计因素与IRI之间的关系提供了有价值的见解，为路面设计和管理的循证决策提供了路线图。研究结果可以指导利益相关者优先考虑设计因素，以优化路面质量和整体道路性能。

{"title":"Application of ensemble-based machine learning models for investigating the importance of asphalt pavement design factors on IRI","authors":"Abdulraaof H. Al-Qaili, Abdullah I. Al-Mansour, Hamad Al-Solieman","doi":"10.1016/j.measurement.2026.121087","DOIUrl":"10.1016/j.measurement.2026.121087","url":null,"abstract":"<div><div>Asphalt pavement design factors play a critical role in determining the quality and longevity of pavements. Understanding how these factors affect the International Roughness Index (IRI) is essential for optimizing pavement performance and improving road quality. In this study, the importance of asphalt pavement design factors on IRI investigate using ensemble-based machine learning models, providing valuable insights for pavement design and maintenance strategies.</div><div>Ensemble models, specifically Random Forest (RF) and Gradient Boosting (GB), were used to analyze the importance of asphalt pavement design factors on the IRI. These algorithms combine the predictive power of multiple decision trees to increase accuracy and identify critical variables that affect IRI.</div><div>The results of the study demonstrated the influence of hyperparameters on model performance and highlighted the importance of parameter tuning for optimal results. Additionally, the study emphasizes the critical role of factors such as temperature, traffic volume, and pavement age in predicting IRI across different models. SHAP analysis revealed the primary factors influencing IRI, conforming the importance of environmental conditions and material properties.</div><div>This study provide valuable insights into the relationships between asphalt pavement design factors and IRI, offering a roadmap for evidence-based decision making in pavement design and management. The findings can guide stakeholders in prioritizing design factors to optimize pavement quality and overall road performance.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121087"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388161","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Ordered WO3-decorated ZrO2 nanotubes with efficient three-phase boundary towards highly sensitive detection of trace nitrogen dioxide 具有高效三相边界的有序wo3修饰ZrO2纳米管用于痕量二氧化氮的高灵敏度检测

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-05-05 Epub Date: 2026-03-03 DOI: 10.1016/j.measurement.2026.121022

Liang-Bo Bo , Xin-Feng Qiao , Xiao-Hong Zheng , Su-Yue Ren , Zhi-Lei Li , Jing-Wen Pan , Cheng-Yong Chen

Long-term exposure to high concentrations of nitrogen dioxide (NO₂) poses severe risks to both the environment and human health, thereby creating an urgent need for gas sensors that enable sensitive and reliable detection at relatively low operating temperatures. However, most reported NO₂ sensors still suffer from either high operating temperatures or insufficient detection sensitivity, which severely limits their practical application in atmospheric monitoring. In this study, a potential NO₂ sensor based on a ordered WO₃-decorated ZrO₂ nanotube film was fabricated via anodic oxidation and in-situ growth. The sensing properties, including a low optimum test temperature, low detection limit, high sensitivity, repeatability, and selectivity, were determined. The ordered WO₃-decorated ZrO₂ nanotube film exhibited a high NO₂ sensitivity (134.8 mV/decade) toward 0.06–10 ppm NO₂ at a medium–low working temperature of 350°C. The high performance of the sensor is primarily attributed to Knudsen diffusion in the ordered nanotube structure and the perfect three-phase boundary (TPB) of the nanocomposite, as the results indicate. This study presents a novel strategy for designing ordered nanostructures and introduces an effective approach for detecting atmospheric NO₂ concentrations.

长期暴露于高浓度二氧化氮（NO2）对环境和人类健康构成严重风险，因此迫切需要能够在相对较低的工作温度下进行敏感和可靠检测的气体传感器。然而，大多数已报道的NO2传感器仍然存在工作温度高或检测灵敏度不足的问题，这严重限制了其在大气监测中的实际应用。在本研究中，采用阳极氧化和原位生长的方法制备了基于有序wo3修饰的ZrO2纳米管薄膜的潜在NO2传感器。测定了该方法的传感性能，包括低最佳测试温度、低检出限、高灵敏度、重复性和选择性。在350℃的中低温下，有序wo3修饰的ZrO2纳米管薄膜对0.06-10 ppm NO2具有较高的NO2灵敏度（134.8 mV/decade）。结果表明，传感器的高性能主要归功于有序纳米管结构中的Knudsen扩散和纳米复合材料的完美三相边界（TPB）。本研究提出了一种设计有序纳米结构的新策略，并介绍了一种检测大气NO2浓度的有效方法。

{"title":"Ordered WO3-decorated ZrO2 nanotubes with efficient three-phase boundary towards highly sensitive detection of trace nitrogen dioxide","authors":"Liang-Bo Bo , Xin-Feng Qiao , Xiao-Hong Zheng , Su-Yue Ren , Zhi-Lei Li , Jing-Wen Pan , Cheng-Yong Chen","doi":"10.1016/j.measurement.2026.121022","DOIUrl":"10.1016/j.measurement.2026.121022","url":null,"abstract":"<div><div>Long-term exposure to high concentrations of nitrogen dioxide (NO<sub>2</sub>) poses severe risks to both the environment and human health, thereby creating an urgent need for gas sensors that enable sensitive and reliable detection at relatively low operating temperatures. However, most reported NO<sub>2</sub> sensors still suffer from either high operating temperatures or insufficient detection sensitivity, which severely limits their practical application in atmospheric monitoring. In this study, a potential NO<sub>2</sub> sensor based on a ordered WO<sub>3</sub>-decorated ZrO<sub>2</sub> nanotube film was fabricated via anodic oxidation and in-situ growth. The sensing properties, including a low optimum test temperature, low detection limit, high sensitivity, repeatability, and selectivity, were determined. The ordered WO<sub>3</sub>-decorated ZrO<sub>2</sub> nanotube film exhibited a high NO<sub>2</sub> sensitivity (134.8 mV/decade) toward 0.06–10 ppm NO<sub>2</sub> at a medium–low working temperature of 350°C. The high performance of the sensor is primarily attributed to Knudsen diffusion in the ordered nanotube structure and the perfect three-phase boundary (TPB) of the nanocomposite, as the results indicate. This study presents a novel strategy for designing ordered nanostructures and introduces an effective approach for detecting atmospheric NO<sub>2</sub> concentrations.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121022"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388296","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Characterisation and decoupling of electromagnetic properties in liquid–solid system using high-frequency electromagnetic tomography 利用高频电磁层析成像技术表征液固系统的电磁特性并进行解耦

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-05-05 Epub Date: 2026-03-05 DOI: 10.1016/j.measurement.2026.120996

Xun Zou, Kuohai Yu, Yu Li, Saibo She, Xinnan Zheng, Shu Lin, Wuliang Yin

In the liquid–solid fluidised system with conductive fluid and magnetic catalysts, the monitoring of electrical conductivity and magnetic permeability is essential to maintain the desired operating conditions. Conventional electromagnetic tomography (EMT) techniques applied to fluidised beds are limited to magnetic phase detection, while conductivity measurements still rely on contact methods or multimodal integration. Therefore, this study presents a non-contact approach for the characterisation and decoupling of electromagnetic properties in solid–liquid systems with a high-frequency EMT sensor. A pre-calibrated vector decomposition method is developed for the simultaneous estimation of conductivity and permeability. The effectiveness of the method is verified through experiments with magnetite saline dispersions in various concentrations, which give average estimation errors of 3.44% and 0.16% for the conductivity and permeability estimation, respectively. In the imaging experiments, a modified Landweber algorithm integrating Nesterov momentum and weight assignment is developed and applied to the reconstruction, contributing to fast convergence and enhanced image quality. Furthermore, distributions of conductive phase and magnetic phase are decoupled through the reconstruction of distinct voltage response components.

在具有导电性流体和磁性催化剂的液固流化系统中，电导率和磁导率的监测对于维持所需的操作条件至关重要。应用于流化床的传统电磁层析成像（EMT）技术仅限于磁相位检测，而电导率测量仍然依赖于接触方法或多模态积分。因此，本研究提出了一种非接触的方法来表征和解耦高频EMT传感器在固液系统中的电磁特性。提出了一种预校正矢量分解方法，用于同时估计电导率和渗透率。通过不同浓度的磁铁矿盐分散体实验验证了该方法的有效性，其电导率和渗透率的平均估计误差分别为3.44%和0.16%。在成像实验中，提出了一种结合Nesterov动量和权值分配的Landweber算法，并将其应用于图像重建，提高了图像的收敛速度和图像质量。此外，通过重构不同的电压响应分量来解耦导电相位和磁相位的分布。

{"title":"Characterisation and decoupling of electromagnetic properties in liquid–solid system using high-frequency electromagnetic tomography","authors":"Xun Zou, Kuohai Yu, Yu Li, Saibo She, Xinnan Zheng, Shu Lin, Wuliang Yin","doi":"10.1016/j.measurement.2026.120996","DOIUrl":"10.1016/j.measurement.2026.120996","url":null,"abstract":"<div><div>In the liquid–solid fluidised system with conductive fluid and magnetic catalysts, the monitoring of electrical conductivity and magnetic permeability is essential to maintain the desired operating conditions. Conventional electromagnetic tomography (EMT) techniques applied to fluidised beds are limited to magnetic phase detection, while conductivity measurements still rely on contact methods or multimodal integration. Therefore, this study presents a non-contact approach for the characterisation and decoupling of electromagnetic properties in solid–liquid systems with a high-frequency EMT sensor. A pre-calibrated vector decomposition method is developed for the simultaneous estimation of conductivity and permeability. The effectiveness of the method is verified through experiments with magnetite saline dispersions in various concentrations, which give average estimation errors of 3.44% and 0.16% for the conductivity and permeability estimation, respectively. In the imaging experiments, a modified Landweber algorithm integrating Nesterov momentum and weight assignment is developed and applied to the reconstruction, contributing to fast convergence and enhanced image quality. Furthermore, distributions of conductive phase and magnetic phase are decoupled through the reconstruction of distinct voltage response components.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 120996"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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