Measurement最新文献_第8页

Synergistic axial-radial magnetic structure achieving high uniformity and flux density in seismic monitoring sensor 协同轴向-径向磁结构，在地震监测传感器中实现高均匀性和高磁通密度

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-30 DOI: 10.1016/j.measurement.2026.120666

Yihong Liu , Zhenjing Yao , Mengtao Xing , Xianglong Liu

The magnetoelectric seismometer is crucial for seismic monitoring. Its magnetic field structure, core energy conversion component, not only determines sensitivity but also largely dictates nonlinear errors. Conventional dual magnetic ring designs, based on coaxial nesting with radial magnetization, suffer from incomplete magnetic circuits and edge effects. These cause flux leakage at the air gap edges, reducing flux density and distribution uniformity. To address these issues, this study innovatively proposes an axial-radial hybrid magnetic field structure. By combining an axially magnetized compensation ring with a centrally located, radially magnetized annular permanent magnet, with profiled magnetic boots, the structure guides peripheral flux toward the center of the working air gap, enhances magnetic circuit closure, and enlarges the inner magnetic yoke radius to reduce flux saturation; these design features work to suppress edge flux leakage, thereby improving the uniformity of the air gap magnetic field. Simulation and experimental results show that the new structure achieves a magnetic flux density of 1,031 mT, which is a 16.10% increase over the 888 mT of the conventional dual magnetic ring designs. The uniform field region expands to 8.30 mm, with a 12.16% improvement in uniformity. The peak sensitivity reaches 1702 V/(m·s⁻¹), a 10.95% increase from the original structure (1534 V/(m·s⁻¹)), significantly enhancing the detection capability for weak seismic signals. This study presents an innovative axial-radial hybrid magnetic field structure that resolves the conflict between flux density and field uniformity, simultaneously enhancing sensitivity and suppressing nonlinear errors, thus establishing a new paradigm for high-performance magnetoelectric seismometer design.

磁电地震仪是地震监测的重要仪器。它的磁场结构，核心能量转换组件，不仅决定了灵敏度，而且在很大程度上决定了非线性误差。传统的同轴嵌套径向磁化双磁环设计存在磁路不完整和边缘效应。这会导致气隙边缘漏磁，降低磁通密度和分布均匀性。为了解决这些问题，本研究创新性地提出了轴向-径向混合磁场结构。该结构通过将轴向磁化补偿环与位于中心的径向磁化环形永磁体、带异型磁靴相结合，将外围磁通引导至工作气隙中心，增强磁路闭合性，增大内磁轭半径，降低磁通饱和；这些设计特点可以抑制边缘漏磁，从而提高气隙磁场的均匀性。仿真和实验结果表明，该结构的磁通密度为1031 mT，比传统双磁环设计的888 mT提高了16.10%。均匀场区域扩展到8.30 mm，均匀性提高了12.16%。峰值灵敏度达到1702 V/(m·s−1)，比原结构（1534 V/(m·s−1)）提高了10.95%，显著增强了对微弱地震信号的探测能力。本研究提出了一种新颖的轴-径向混合磁场结构，解决了磁通密度与场均匀性之间的冲突，同时提高了灵敏度，抑制了非线性误差，从而为高性能磁电地震仪的设计建立了新的范式。

{"title":"Synergistic axial-radial magnetic structure achieving high uniformity and flux density in seismic monitoring sensor","authors":"Yihong Liu , Zhenjing Yao , Mengtao Xing , Xianglong Liu","doi":"10.1016/j.measurement.2026.120666","DOIUrl":"10.1016/j.measurement.2026.120666","url":null,"abstract":"<div><div>The magnetoelectric seismometer is crucial for seismic monitoring. Its magnetic field structure, core energy conversion component, not only determines sensitivity but also largely dictates nonlinear errors. Conventional dual magnetic ring designs, based on coaxial nesting with radial magnetization, suffer from incomplete magnetic circuits and edge effects. These cause flux leakage at the air gap edges, reducing flux density and distribution uniformity. To address these issues, this study innovatively proposes an axial-radial hybrid magnetic field structure. By combining an axially magnetized compensation ring with a centrally located, radially magnetized annular permanent magnet, with profiled magnetic boots, the structure guides peripheral flux toward the center of the working air gap, enhances magnetic circuit closure, and enlarges the inner magnetic yoke radius to reduce flux saturation; these design features work to suppress edge flux leakage, thereby improving the uniformity of the air gap magnetic field. Simulation and experimental results show that the new structure achieves a magnetic flux density of 1,031 mT, which is a 16.10% increase over the 888 mT of the conventional dual magnetic ring designs. The uniform field region expands to 8.30 mm, with a 12.16% improvement in uniformity. The peak sensitivity reaches 1702 V/(m·s<sup>−1</sup>), a 10.95% increase from the original structure (1534 V/(m·s<sup>−1</sup>)), significantly enhancing the detection capability for weak seismic signals. This study presents an innovative axial-radial hybrid magnetic field structure that resolves the conflict between flux density and field uniformity, simultaneously enhancing sensitivity and suppressing nonlinear errors, thus establishing a new paradigm for high-performance magnetoelectric seismometer design.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120666"},"PeriodicalIF":5.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146171993","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

MRS-Pyramid-TimeXer: A quality-related distributed monitoring framework for multi-rate industrial processes MRS-Pyramid-TimeXer：多速率工业过程的质量相关分布式监控框架

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-30 DOI: 10.1016/j.measurement.2026.120658

Jie Dong , Minjie Zhang , Qichun Zhang , Kaixiang Peng

Modern industrial processes are increasingly large-scale and multi-unit, which renders quality-related fault detection under multi-rate sampling a challenging task. To address this, MRS-Pyramid-TimeXer is proposed,which is a quality-related distributed fault detection framework that integrates pyramid-based multi-scale representations with an enhanced TimeXer architecture. First, it employs a knowledge collaborative strategy based on mutual information–mechanism to partition multi-rate variables into interpretable subsystems, incorporating process parameters as auxiliary variables to enhance multi-rate coupling characterization. Second, Pyramid-TimeXer models are constructed within local subsystems, utilizing patch-wise self-attention and variate-wise cross-attention to capture multi-scale dynamic correlations between quality and process variables. Finally, Bayesian posterior probability fuses statistics from each subsystem to derive global process metrics. Experiments on a Hot Strip Milling Process (HSMP) show that the proposed framework achieves an average fault detection rate of 98% with an average false alarm rate below 2% across three typical fault scenarios, consistently outperforming state-of-the-art baselines and validating its effectiveness and engineering applicability in large-scale manufacturing.

现代工业过程日益大规模和多单元化，使得多速率采样下的质量故障检测成为一项具有挑战性的任务。为了解决这个问题，提出了MRS-Pyramid-TimeXer，这是一个与质量相关的分布式故障检测框架，它将基于金字塔的多尺度表示与增强的TimeXer体系结构相结合。首先，采用基于互信息机制的知识协同策略，将多速率变量划分为可解释的子系统，并将工艺参数作为辅助变量，增强多速率耦合表征；其次，在局部子系统中构建金字塔-时间模型，利用补丁型自关注和变量型交叉关注捕获质量和过程变量之间的多尺度动态相关性。最后，贝叶斯后验概率融合各子系统的统计信息，得到全局过程度量。在热轧带钢工艺（HSMP）上的实验表明，该框架在三种典型故障场景下的平均故障检测率为98%，平均误报率低于2%，始终优于最先进的基线，并验证了其在大规模制造中的有效性和工程适用性。

{"title":"MRS-Pyramid-TimeXer: A quality-related distributed monitoring framework for multi-rate industrial processes","authors":"Jie Dong , Minjie Zhang , Qichun Zhang , Kaixiang Peng","doi":"10.1016/j.measurement.2026.120658","DOIUrl":"10.1016/j.measurement.2026.120658","url":null,"abstract":"<div><div>Modern industrial processes are increasingly large-scale and multi-unit, which renders quality-related fault detection under multi-rate sampling a challenging task. To address this, MRS-Pyramid-TimeXer is proposed,which is a quality-related distributed fault detection framework that integrates pyramid-based multi-scale representations with an enhanced TimeXer architecture. First, it employs a knowledge collaborative strategy based on mutual information–mechanism to partition multi-rate variables into interpretable subsystems, incorporating process parameters as auxiliary variables to enhance multi-rate coupling characterization. Second, Pyramid-TimeXer models are constructed within local subsystems, utilizing patch-wise self-attention and variate-wise cross-attention to capture multi-scale dynamic correlations between quality and process variables. Finally, Bayesian posterior probability fuses statistics from each subsystem to derive global process metrics. Experiments on a Hot Strip Milling Process (HSMP) show that the proposed framework achieves an average fault detection rate of 98% with an average false alarm rate below 2% across three typical fault scenarios, consistently outperforming state-of-the-art baselines and validating its effectiveness and engineering applicability in large-scale manufacturing.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120658"},"PeriodicalIF":5.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081032","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

Adding active damping to hydraulic systems using observed pressure feedback 利用观察到的压力反馈给液压系统增加主动阻尼

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-30 DOI: 10.1016/j.measurement.2026.120610

Damiano Padovani, Xieli Zhang

The growing focus on energy efficiency directs the hydraulics research toward valveless systems. The resulting elimination of functional energy dissipations leads to inherently low system damping, which requires compensating techniques, such as pressure feedback of the actuator’s pressure differential (

Δ p

). It minimizes harmful payload oscillations, improving the accuracy of position control. Thus, we propose replacing the two physical pressure transducers measuring the

Δ p

with a virtual observer implemented via software. We tested this approach on a full-scale robotic manipulator, using a model-based design for the closed-loop position control and high-pass-filtered pressure feedback. Despite the imperfect

Δ p

estimation, the experimental evidence confirmed that feeding back the observed

Δ p

is comparable to the state-of-the-art approach relying upon the measured

Δ p

. Removing two physical pressure transducers without altering the system response is therefore feasible and advantageous for commissioning and cost-effectiveness; this simplification supports the profitable use of hydraulic systems, including the energy-efficient ones.

对能源效率的日益关注将液压研究导向无阀系统。由此产生的消除功能能量耗散导致固有的低系统阻尼，这需要补偿技术，例如执行器压差的压力反馈（Δp）。它最大限度地减少有害的载荷振荡，提高位置控制的精度。因此，我们建议用通过软件实现的虚拟观测器取代测量Δp的两个物理压力传感器。我们在一个全尺寸的机械臂上测试了这种方法，使用基于模型的闭环位置控制和高通滤波压力反馈设计。尽管Δp估计不完美，但实验证据证实，反馈观察到的Δp与依靠测量到的Δp的最先进方法相当。因此，在不改变系统响应的情况下移除两个物理压力传感器是可行的，有利于调试和成本效益；这种简化支持液压系统的盈利使用，包括节能系统。

{"title":"Adding active damping to hydraulic systems using observed pressure feedback","authors":"Damiano Padovani, Xieli Zhang","doi":"10.1016/j.measurement.2026.120610","DOIUrl":"10.1016/j.measurement.2026.120610","url":null,"abstract":"<div><div>The growing focus on energy efficiency directs the hydraulics research toward valveless systems. The resulting elimination of functional energy dissipations leads to inherently low system damping, which requires compensating techniques, such as pressure feedback of the actuator’s pressure differential (<span><math><mrow><mi>Δ</mi><mi>p</mi></mrow></math></span>). It minimizes harmful payload oscillations, improving the accuracy of position control. Thus, we propose replacing the two physical pressure transducers measuring the <span><math><mrow><mi>Δ</mi><mi>p</mi></mrow></math></span> with a virtual observer implemented via software. We tested this approach on a full-scale robotic manipulator, using a model-based design for the closed-loop position control and high-pass-filtered pressure feedback. Despite the imperfect <span><math><mrow><mi>Δ</mi><mi>p</mi></mrow></math></span> estimation, the experimental evidence confirmed thatfeeding back the observed <span><math><mrow><mi>Δ</mi><mi>p</mi></mrow></math></span> is comparable to the state-of-the-art approach relying upon the measured <span><math><mrow><mi>Δ</mi><mi>p</mi></mrow></math></span>. Removing two physical pressure transducers without altering the system response is therefore feasible and advantageous for commissioning and cost-effectiveness; this simplification supports the profitable use of hydraulic systems, including the energy-efficient ones.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120610"},"PeriodicalIF":5.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146171430","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

Towards diagnostics of damage state in self-healing composites using an AI-driven acousto-ultrasonic approach 利用人工智能驱动的声超声方法诊断自修复复合材料的损伤状态

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-30 DOI: 10.1016/j.measurement.2026.120652

Claudia Barile, Vimalathithan Paramsamy Kannan

Self-healing composites are a novel class of composites which can autonomously heal matrix damages and recover their mechanical properties through external stimulus. For efficient recovery of their mechanical properties, it is essential to establish their damage state non-destructively. In this investigation, an AI-driven Acousto-Ultrasonic approach is designed to analyse the damage and heal states of self-healing composite specimens. Accordingly, artificial stress waves are generated and propagated through the self-healing composites in their different damage states and are evaluated. The stress waves in the time domain are transformed into coefficients using Mel frequency spectral analysis. The resulting Mel frequency cepstral coefficients are used to extract the underlying features in the stress waves originating from the different damage states. The features are used to train a lightweight convolutional neural network (CNN) to automatically classify the damage states. The results show that the CNN classifies the damage states in the test specimens with an exceptional accuracy of 98.66% and an F1 score of 99.18%. Therefore, this AI-driven Acousto-Ultrasonic approach has the potential to be upscaled for large structures and be used as an efficient non-destructive tool to evaluate the damage states of the self-healing composites.

自修复复合材料是一种新型的复合材料，它能在外界刺激下自动修复基体损伤并恢复其力学性能。为了有效地恢复其力学性能，必须以非破坏性的方式建立其损伤状态。在这项研究中，设计了一种人工智能驱动的声超声方法来分析自修复复合材料试件的损伤和愈合状态。因此，人工应力波的产生和传播在不同损伤状态下的自愈复合材料，并评估。利用Mel频谱分析将应力波在时域内转换为系数。得到的Mel频率倒谱系数用于提取来自不同损伤状态的应力波的潜在特征。这些特征被用来训练一个轻量级的卷积神经网络（CNN）来自动分类损伤状态。结果表明，CNN对试件损伤状态的分类准确率达到了98.66%，F1分数达到了99.18%。因此，这种人工智能驱动的声超声方法有可能被扩展到大型结构，并被用作评估自修复复合材料损伤状态的有效非破坏性工具。

{"title":"Towards diagnostics of damage state in self-healing composites using an AI-driven acousto-ultrasonic approach","authors":"Claudia Barile, Vimalathithan Paramsamy Kannan","doi":"10.1016/j.measurement.2026.120652","DOIUrl":"10.1016/j.measurement.2026.120652","url":null,"abstract":"<div><div>Self-healing composites are a novel class of composites which can autonomously heal matrix damages and recover their mechanical properties through external stimulus. For efficient recovery of their mechanical properties, it is essential to establish their damage state non-destructively. In this investigation, an AI-driven Acousto-Ultrasonic approach is designed to analyse the damage and heal states of self-healing composite specimens. Accordingly, artificial stress waves are generated and propagated through the self-healing composites in their different damage states and are evaluated. The stress waves in the time domain are transformed into coefficients using Mel frequency spectral analysis. The resulting Mel frequency cepstral coefficients are used to extract the underlying features in the stress waves originating from the different damage states. The features are used to train a lightweight convolutional neural network (CNN) to automatically classify the damage states. The results show that the CNN classifies the damage states in the test specimens with an exceptional accuracy of 98.66% and an F1 score of 99.18%. Therefore, this AI-driven Acousto-Ultrasonic approach has the potential to be upscaled for large structures and be used as an efficient non-destructive tool to evaluate the damage states of the self-healing composites.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120652"},"PeriodicalIF":5.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172088","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

Improved set-membership-filtering-based state estimation for target localization system 目标定位系统的改进集成员滤波状态估计

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-30 DOI: 10.1016/j.measurement.2026.120653

Xinyong Zhang, Jiongqi Wang, Haiyin Zhou, Bowen Hou

Set membership filtering is an effective approach for target localization in scenarios where the statistical model of the noise remains unknown, while the bounds of the noise have been determined. An accurate and efficient target localization state estimation method is proposed based on the set membership filtering framework. The self-fusion method is designed to enhance the accuracy of nonlinear redundant measurement data through the GDOP of the measurement system. Besides, the boundary of target localization errors from the nonlinear measurement model under unknown statistics but bounded of measurement errors is characterized by zonotopes within update steps of the set membership filtering. Meanwhile, the convex relaxation approach is used to solve the filter parameter to ensure both precision and efficiency. Simulations and field experiments corroborate the effectiveness of our proposed method.

集合隶属度滤波是在噪声的统计模型未知而噪声的边界已经确定的情况下进行目标定位的有效方法。提出了一种基于集隶属度滤波框架的准确高效的目标定位状态估计方法。设计了自融合方法，通过测量系统的GDOP来提高非线性冗余测量数据的精度。此外，在统计量未知但测量误差有界的非线性测量模型中，目标定位误差的边界在集合隶属度滤波的更新步内用分区来表征。同时，采用凸松弛法求解滤波参数，保证了滤波的精度和效率。仿真和现场实验验证了该方法的有效性。

引用次数: 0

Zero-shot learning for wind turbine blade defect detection via symptom description transfer 基于症状描述传递的零学习风电叶片缺陷检测

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-30 DOI: 10.1016/j.measurement.2026.120668

Qiuyu Yang , Wenjun Qiu , Jiangjun Ruan , Xue Xue , Jingyi Xie , Yuyi Lin

Ensuring wind turbine blade structural health is crucial for optimal wind power generation. Achieving this requires reliable blade defect detection under complex operating conditions. However, limited access to defect data under harsh environments poses challenges to conventional detection methods. To overcome this, this paper proposes MGAM (multi-scale gated attention mechanism), a zero-shot learning approach for blade defect detection that leverages generalized symptom descriptions. Compared to existing literature, the innovations of this approach include: (1) The proposed symptom-description paradigm enables cross-defect and cross-turbine identification without requiring direct defect samples; (2) A hierarchical feature learning architecture combining wavelet-based multi-scale refinement with dual-branch feature extraction modules—namely, the multi-scale guard (MSG) and gated linear unit with squeeze-and-excitation (GLUSE)—to enhance discriminative features; (3) A semantic alignment mechanism (loss attention mechanism, LAM) establishes a mapping between symptom descriptions and defect features through three distinct vectors via linear space mapping. Turbine blade test results demonstrate that the symptom-based zero-shot learning approach reduces reliance on historical fault data and shows strong potential for practical deployment in wind turbine condition monitoring.

确保风力发电机叶片结构健康是实现最佳风力发电的关键。实现这一目标需要在复杂的操作条件下可靠的叶片缺陷检测。然而，在恶劣环境下对缺陷数据的有限访问给传统检测方法带来了挑战。为了克服这一点，本文提出了MGAM（多尺度门控注意机制），这是一种利用广义症状描述的叶片缺陷检测的零采样学习方法。与现有文献相比，该方法的创新之处包括：(1)提出的症状-描述范式无需直接缺陷样本即可实现跨缺陷和跨涡轮识别；(2)结合基于小波的多尺度细化和双分支特征提取模块（即多尺度保护（MSG）和带挤压激励（GLUSE）的门控线性单元）的分层特征学习架构，增强了判别特征；(3)语义对齐机制（loss attention mechanism， LAM）通过线性空间映射，通过三个不同的向量建立症状描述与缺陷特征之间的映射关系。风力机叶片试验结果表明，基于症状的零射击学习方法减少了对历史故障数据的依赖，在风力机状态监测中具有很强的实际应用潜力。

{"title":"Zero-shot learning for wind turbine blade defect detection via symptom description transfer","authors":"Qiuyu Yang , Wenjun Qiu , Jiangjun Ruan , Xue Xue , Jingyi Xie , Yuyi Lin","doi":"10.1016/j.measurement.2026.120668","DOIUrl":"10.1016/j.measurement.2026.120668","url":null,"abstract":"<div><div>Ensuring wind turbine blade structural health is crucial for optimal wind power generation. Achieving this requires reliable blade defect detection under complex operating conditions. However, limited access to defect data under harsh environments poses challenges to conventional detection methods. To overcome this, this paper proposes MGAM (multi-scale gated attention mechanism), a zero-shot learning approach for blade defect detection that leverages generalized symptom descriptions. Compared to existing literature, the innovations of this approach include: (1) The proposed symptom-description paradigm enables cross-defect and cross-turbine identification without requiring direct defect samples; (2) A hierarchical feature learning architecture combining wavelet-based multi-scale refinement with dual-branch feature extraction modules—namely, the multi-scale guard (MSG) and gated linear unit with squeeze-and-excitation (GLUSE)—to enhance discriminative features; (3) A semantic alignment mechanism (loss attention mechanism, LAM) establishes a mapping between symptom descriptions and defect features through three distinct vectors via linear space mapping. Turbine blade test results demonstrate that the symptom-based zero-shot learning approach reduces reliance on historical fault data and shows strong potential for practical deployment in wind turbine condition monitoring.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120668"},"PeriodicalIF":5.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172076","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

Enhancing measurement accuracy in robot hand-eye systems through integrated geometric and compliance error modeling 通过集成几何和柔度误差建模，提高机器人手眼系统的测量精度

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-30 DOI: 10.1016/j.measurement.2026.120669

Bao Zhu , Xiaoyu Guo , Meng Chi , Yanding Wei , Mingjun Tang , Qiang Fang

In the execution of assembly feature scanning tasks within robot hand-eye measurement systems, the geometric errors of the robot body, together with joint compliance deformations caused by variations in end payload and posture adjustments during assembly, collectively lead to significant measurement position deviations. Considering the characteristics of industrial robots performing such tasks, including wide end motion ranges, frequent payload changes, and complex posture adjustments, which result in complicated coupling of joint errors, a unified error model was proposed in this study. Based on the Modified Denavit-Hartenberg (MDH) framework, an integrated error propagation model was developed, incorporating both geometric deviations of the robot body and joint compliance deformations. The error transmission characteristics were numerically modeled using a combination of finite difference and Lie algebra mapping methods. Additionally, a multi-scale trust region reflective Levenberg-Marquardt (MS-TRLM) algorithm was introduced, employing a three-stage progressive optimization mechanism to enhance the accuracy of parameter identification. Experimental results indicate that the proposed error compensation method demonstrates strong adaptability and stability under various scanning directions, payload conditions, and end-effector postures. Specifically, under the Y_LIN scanning direction and a 13 kg payload condition, the average and maximum measurement errors were reduced to 0.0811 mm and 0.1318 mm, representing reductions of 77.08% and 72.79% compared to the uncompensated results. This study effectively improves the measurement accuracy and reliability of robot hand-eye systems during the assembly process.

在机器人手眼测量系统中执行装配特征扫描任务时，机器人本体的几何误差以及装配过程中末端载荷变化和姿态调整引起的关节顺应性变形共同导致了显著的测量位置偏差。考虑到工业机器人末端运动范围大、载荷变化频繁、姿态调整复杂等特点，导致关节误差耦合复杂，提出了统一的误差模型。基于改进的Denavit-Hartenberg （MDH）框架，建立了综合考虑机器人本体几何偏差和关节柔度变形的误差传播模型。采用有限差分和李代数映射相结合的方法对误差传输特性进行了数值模拟。此外，引入了一种多尺度信任域反射Levenberg-Marquardt （MS-TRLM）算法，采用三阶段渐进优化机制提高参数辨识的准确性。实验结果表明，所提出的误差补偿方法在不同的扫描方向、载荷条件和末端执行器姿态下都具有较强的适应性和稳定性。其中，在Y_LIN扫描方向和13 kg载荷条件下，平均测量误差和最大测量误差分别减小到0.0811 mm和0.1318 mm，分别比未补偿时减小77.08%和72.79%。该研究有效地提高了机器人手眼系统在装配过程中的测量精度和可靠性。

{"title":"Enhancing measurement accuracy in robot hand-eye systems through integrated geometric and compliance error modeling","authors":"Bao Zhu , Xiaoyu Guo , Meng Chi , Yanding Wei , Mingjun Tang , Qiang Fang","doi":"10.1016/j.measurement.2026.120669","DOIUrl":"10.1016/j.measurement.2026.120669","url":null,"abstract":"<div><div>In the execution of assembly feature scanning tasks within robot hand-eye measurement systems, the geometric errors of the robot body, together with joint compliance deformations caused by variations in end payload and posture adjustments during assembly, collectively lead to significant measurement position deviations. Considering the characteristics of industrial robots performing such tasks, including wide end motion ranges, frequent payload changes, and complex posture adjustments, which result in complicated coupling of joint errors, a unified error model was proposed in this study. Based on the Modified Denavit-Hartenberg (MDH) framework, an integrated error propagation model was developed, incorporating both geometric deviations of the robot body and joint compliance deformations. The error transmission characteristics were numerically modeled using a combination of finite difference and Lie algebra mapping methods. Additionally, a multi-scale trust region reflective Levenberg-Marquardt (MS-TRLM) algorithm was introduced, employing a three-stage progressive optimization mechanism to enhance the accuracy of parameter identification. Experimental results indicate that the proposed error compensation method demonstrates strong adaptability and stability under various scanning directions, payload conditions, and end-effector postures. Specifically, under the Y_LIN scanning direction and a 13 kg payload condition, the average and maximum measurement errors were reduced to 0.0811 mm and 0.1318 mm, representing reductions of 77.08% and 72.79% compared to the uncompensated results. This study effectively improves the measurement accuracy and reliability of robot hand-eye systems during the assembly process.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120669"},"PeriodicalIF":5.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172133","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

Quantitative characterization of fractal features of coal rock pore-fracture based on ResUnet-CBAM modeling 基于reunet - cbam模型的煤岩孔隙破裂分形特征定量表征

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-30 DOI: 10.1016/j.measurement.2026.120609

Shuang Song , Linwei Li , Suinan He , Yilun Xue , Hongyu Pan , Ruoyu Bao , Xinshuang Cao , Guoying Liu

Microscopic fractures in coal rocks exhibit multi-scale development and structurally complex features. Conventional image segmentation methods often suffer from insufficient accuracy and blurred boundaries in fracture identification. This severely compromises the reliability of subsequent structural quantification and physical property analysis. Accordingly, we develop a ResUnet-CBAM framework combining three-dimensional CT reconstruction and deep learning. It enables high-precision segmentation and quantitative characterization of pore-fracture structures. By incorporating residual connections, CBAM attention, and ASPP modules, the model substantially improves feature extraction for micro-fractures and blurred boundaries.The results demonstrate that the ResUnet-CBAM model improves the segmentation accuracy of micro-fractures and reduces misclassification of the coal matrix. The model achieved a loss value of 0.2, a mean intersection over union (MIoU) of 76%, a Mean Pixel Accuracy (MPA) of 82.3%, and a precision of 83.6%. Quantitative analysis based on the high-precision segmentation results shows that the fractal dimension is positively correlated with fracture volume fraction and coordination number, while it is negatively correlated with pore-throat radius.This relationship underscores the link between coal structural complexity and its physical properties. It also supports efficient coalbed methane development and hazard prevention.

煤岩微观裂缝具有多尺度发育和构造复杂的特点。传统的图像分割方法在裂缝识别中存在精度不足、边界模糊等问题。这严重损害了后续结构量化和物理性质分析的可靠性。因此，我们开发了一个结合三维CT重建和深度学习的reunet - cbam框架。它可以实现孔隙破裂结构的高精度分割和定量表征。通过结合残余连接、CBAM关注和ASPP模块，该模型大大提高了微裂缝和模糊边界的特征提取。结果表明，ResUnet-CBAM模型提高了微裂缝的分割精度，减少了煤基质的误分类。该模型的损失值为0.2，平均交联（MIoU）为76%，平均像素精度（MPA）为82.3%，精度为83.6%。基于高精度分割结果的定量分析表明，分形维数与裂缝体积分数、配位数呈正相关，与孔喉半径呈负相关。这种关系强调了煤的结构复杂性与其物理性质之间的联系。它还支持有效的煤层气开发和危害预防。

{"title":"Quantitative characterization of fractal features of coal rock pore-fracture based on ResUnet-CBAM modeling","authors":"Shuang Song , Linwei Li , Suinan He , Yilun Xue , Hongyu Pan , Ruoyu Bao , Xinshuang Cao , Guoying Liu","doi":"10.1016/j.measurement.2026.120609","DOIUrl":"10.1016/j.measurement.2026.120609","url":null,"abstract":"<div><div>Microscopic fractures in coal rocks exhibit multi-scale development and structurally complex features. Conventional image segmentation methods often suffer from insufficient accuracy and blurred boundaries in fracture identification. This severely compromises the reliability of subsequent structural quantification and physical property analysis. Accordingly, we develop a ResUnet-CBAM framework combining three-dimensional CT reconstruction and deep learning. It enables high-precision segmentation and quantitative characterization of pore-fracture structures. By incorporating residual connections, CBAM attention, and ASPP modules, the model substantially improves feature extraction for micro-fractures and blurred boundaries.The results demonstrate that the ResUnet-CBAM model improves the segmentation accuracy of micro-fractures and reduces misclassification of the coal matrix. The model achieved a loss value of 0.2, a mean intersection over union (MIoU) of 76%, a Mean Pixel Accuracy (MPA) of 82.3%, and a precision of 83.6%. Quantitative analysis based on the high-precision segmentation results shows that the fractal dimension is positively correlated with fracture volume fraction and coordination number, while it is negatively correlated with pore-throat radius.This relationship underscores the link between coal structural complexity and its physical properties. It also supports efficient coalbed methane development and hazard prevention.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120609"},"PeriodicalIF":5.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146190446","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

Implementation and performance analysis of an integrated control strategy for high-precision positioning of a parallel 3-DOF piezoelectric stage 并联三自由度压电平台高精度定位集成控制策略的实现与性能分析

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-30 DOI: 10.1016/j.measurement.2026.120661

Guilian Wang, Shuai Hao, Jiangyang Chen, Hang Zhang

Parallel three-degree-of-freedom (3-DOF) piezoelectric stages are essential for applications demanding high-precision positioning, yet their performance is inherently limited by piezoelectric hysteresis, nonlinearities, and mechanical cross-coupling. This paper presents an integrated feedforward-feedback control strategy specifically designed to overcome these limitations. The kinematic and dynamic characteristics of the stage are established through analytical modeling, including forward and inverse kinematics and stiffness analysis, alongside finite element simulation. The proposed control architecture synergistically combines two components: a feedforward compensator that integrates the Prandtl-Ishlinskii (P-I) hysteresis model with a BP neural network for proactive hysteresis and nonlinearity mitigation, and a PID feedback controller for real-time error correction. Experimental validation demonstrates the effectiveness of this integrated approach, achieving high positioning accuracies of 20 nm in the x and y axes, and 50 nm in the z axis. The system exhibits robust performance with excellent three-dimensional tracking capabilities for signals above 2 μm amplitude within the 0–40 Hz frequency range. This research contributes a reliable and high-performance control framework for multi-axis piezoelectric nanopositioning systems, paving the way for enhanced applications in demanding fields such as AFM, MEMS, and biomedicine, while suggesting avenues for future work in adaptive control and dynamic performance optimization.

并联三自由度（3-DOF）压电级对于要求高精度定位的应用至关重要，但其性能受到压电滞后、非线性和机械交叉耦合的固有限制。本文提出了一种集成的前馈-反馈控制策略，专门用于克服这些限制。通过解析建模，包括正、逆运动学和刚度分析，以及有限元仿真，建立了平台的运动学和动力学特性。所提出的控制体系结构协同结合了两个组件：一个前馈补偿器，集成了Prandtl-Ishlinskii （P-I）滞后模型和BP神经网络，用于主动滞后和非线性缓解，以及一个PID反馈控制器，用于实时纠错。实验验证了该方法的有效性，在x轴和y轴上实现了20 nm的高精度定位，在z轴上实现了50 nm的高精度定位。该系统在0-40 Hz频率范围内，对幅度大于2 μm的信号具有良好的三维跟踪能力。该研究为多轴压电纳米定位系统提供了可靠、高性能的控制框架，为其在AFM、MEMS和生物医学等苛刻领域的应用铺平了道路，同时为自适应控制和动态性能优化的未来工作提供了途径。

{"title":"Implementation and performance analysis of an integrated control strategy for high-precision positioning of a parallel 3-DOF piezoelectric stage","authors":"Guilian Wang, Shuai Hao, Jiangyang Chen, Hang Zhang","doi":"10.1016/j.measurement.2026.120661","DOIUrl":"10.1016/j.measurement.2026.120661","url":null,"abstract":"<div><div>Parallel three-degree-of-freedom (3-DOF) piezoelectric stages are essential for applications demanding high-precision positioning, yet their performance is inherently limited by piezoelectric hysteresis, nonlinearities, and mechanical cross-coupling. This paper presents an integrated feedforward-feedback control strategy specifically designed to overcome these limitations. The kinematic and dynamic characteristics of the stage are established through analytical modeling, including forward and inverse kinematics and stiffness analysis, alongside finite element simulation. The proposed control architecture synergistically combines two components: a feedforward compensator that integrates the Prandtl-Ishlinskii (P-I) hysteresis model with a BP neural network for proactive hysteresis and nonlinearity mitigation, and a PID feedback controller for real-time error correction. Experimental validation demonstrates the effectiveness of this integrated approach, achieving high positioning accuracies of 20 nm in the <em>x</em> and <em>y</em> axes, and 50 nm in the <em>z</em> axis. The system exhibits robust performance with excellent three-dimensional tracking capabilities for signals above 2 μm amplitude within the 0–40 Hz frequency range. This research contributes a reliable and high-performance control framework for multi-axis piezoelectric nanopositioning systems, paving the way for enhanced applications in demanding fields such as AFM, MEMS, and biomedicine, while suggesting avenues for future work in adaptive control and dynamic performance optimization.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120661"},"PeriodicalIF":5.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172131","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

Development and static performance analysis of a six-axis force sensor with rigid–flexible collaborative decoupling 刚柔协同解耦六轴力传感器的研制与静态性能分析

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-30 DOI: 10.1016/j.measurement.2026.120608

Yongli Wang , Ke Jin , Huimei Pan , Zikang Xu , Hongxu Liu , Lin Chang

To achieve high-precision measurement, robust anti-interference capability, and good interchangeability, a six-axis force sensor based on a six-bar closed-loop parallel mechanism is proposed. The sensor employs a modular split-body design and incorporates a rigid–flexible coordinated decoupling unit within the measuring branch. By introducing a steel ball and a spring-guided mechanism, the load transmission path is optimized, and the coupling among six-dimensional channels is effectively suppressed, thereby realizing mechanical decoupling. A static model is developed based on the structural equivalence principle, and finite element simulation and static calibration are conducted to verify the effectiveness of the structural design and modeling method, as well as their engineering feasibility. The static calibration results indicate that the linearity, repeatability, hysteresis, and coupling error of the sensor are within 1.38 % FS, 0.43 % FS, 1.13 % FS, and 1.73 % FS, respectively, demonstrating high measurement accuracy and promising potential for engineering applications.

为了实现测量精度高、抗干扰能力强、互换性好，提出了一种基于六杆闭环并联机构的六轴力传感器。该传感器采用模块化分体设计，并在测量分支内集成了刚柔协调解耦单元。通过引入钢球和弹簧导向机构，优化了载荷传递路径，有效抑制了六维通道间的耦合，实现了机械解耦。基于结构等效原理建立了静力模型，并进行了有限元仿真和静力标定，验证了结构设计和建模方法的有效性及工程可行性。静态标定结果表明，传感器的线性度、重复性、滞后度和耦合误差分别在1.38% FS、0.43% FS、1.13% FS和1.73% FS以内，具有较高的测量精度和工程应用潜力。

{"title":"Development and static performance analysis of a six-axis force sensor with rigid–flexible collaborative decoupling","authors":"Yongli Wang , Ke Jin , Huimei Pan , Zikang Xu , Hongxu Liu , Lin Chang","doi":"10.1016/j.measurement.2026.120608","DOIUrl":"10.1016/j.measurement.2026.120608","url":null,"abstract":"<div><div>To achieve high-precision measurement, robust anti-interference capability, and good interchangeability, a six-axis force sensor based on a six-bar closed-loop parallel mechanism is proposed. The sensor employs a modular split-body design and incorporates a rigid–flexible coordinated decoupling unit within the measuring branch. By introducing a steel ball and a spring-guided mechanism, the load transmission path is optimized, and the coupling among six-dimensional channels is effectively suppressed, thereby realizing mechanical decoupling. A static model is developed based on the structural equivalence principle, and finite element simulation and static calibration are conducted to verify the effectiveness of the structural design and modeling method, as well as their engineering feasibility. The static calibration results indicate that the linearity, repeatability, hysteresis, and coupling error of the sensor are within 1.38 % FS, 0.43 % FS, 1.13 % FS, and 1.73 % FS, respectively, demonstrating high measurement accuracy and promising potential for engineering applications.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120608"},"PeriodicalIF":5.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172132","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