Measurement最新文献_第10页

Advanced data reconstruction of vibration signals using a Cauchy–Schwarz variational autoencoder with a spatial-auto-temporal thermal consistency model 基于时空热一致性模型的Cauchy-Schwarz变分自编码器的振动信号高级数据重构

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-29 DOI: 10.1016/j.measurement.2026.120633

Kah Hong Lee , Norhisham Bakhary , Khairul H. Padil , Jun Li , Yon Kong Chen

The effectiveness of vibration-based damage detection depends heavily on the accuracy and completeness of the measured data. However, data loss is inevitable in structural health monitoring, making data reconstruction crucial to ensuring structural safety. However, revealing all complex correlations between the input and the output data in machine learning approaches remains a challenge. Beyond that, even though it is known that spatiotemporal, auto-temporal, and temperature data is influential to the fluctuations of acceleration data, using all three correlations in machine-learning approaches simultaneously remains a bottleneck. To address these challenges, this study presents a novel approach for dynamic response data recovery employing a Cauchy–Schwarz variational autoencoder with a hybrid data arrangement model called the spatial-auto-temporal thermal consistency model. The model uses a probabilistic encoder-decoder structure that leverages the rich expressiveness of a mixed Gaussian as the latent representation to reveal complex relationships between input and output data. The unique architecture of the deep learning model also enables it to be trained using spatiotemporal, auto-temporal, and temperature dependencies simultaneously in the SATTC configuration. The effectiveness of the proposed approach is demonstrated through a case study of field data from the Guangzhou New TV Tower (GNTT). The effects of input channels and measurement noise are also investigated. The quantitative analysis and modal identification results indicate that the proposed approach yields more accurate data reconstruction than VAE-ST, and GAN-ST, and slightly more accurate than CSVAE-ST.

基于振动的损伤检测的有效性在很大程度上取决于测量数据的准确性和完整性。然而，在结构健康监测中，数据丢失是不可避免的，因此数据重建对于保证结构安全至关重要。然而，揭示机器学习方法中输入和输出数据之间的所有复杂相关性仍然是一个挑战。除此之外，尽管已知时空、自动时间和温度数据对加速度数据的波动有影响，但在机器学习方法中同时使用这三种相关性仍然是一个瓶颈。为了应对这些挑战，本研究提出了一种动态响应数据恢复的新方法，采用Cauchy-Schwarz变分自编码器和称为空间-自-时热一致性模型的混合数据排列模型。该模型使用概率编码器-解码器结构，该结构利用混合高斯的丰富表达性作为潜在表示来揭示输入和输出数据之间的复杂关系。深度学习模型的独特架构还使其能够在satc配置中同时使用时空、自动时间和温度依赖关系进行训练。通过对广州新电视塔（GNTT）现场数据的分析，验证了该方法的有效性。研究了输入通道和测量噪声的影响。定量分析和模态识别结果表明，该方法的数据重构精度高于VAE-ST和GAN-ST，略高于CSVAE-ST。

{"title":"Advanced data reconstruction of vibration signals using a Cauchy–Schwarz variational autoencoder with a spatial-auto-temporal thermal consistency model","authors":"Kah Hong Lee , Norhisham Bakhary , Khairul H. Padil , Jun Li , Yon Kong Chen","doi":"10.1016/j.measurement.2026.120633","DOIUrl":"10.1016/j.measurement.2026.120633","url":null,"abstract":"<div><div>The effectiveness of vibration-based damage detection depends heavily on the accuracy and completeness of the measured data. However, data loss is inevitable in structural health monitoring, making data reconstruction crucial to ensuring structural safety. However, revealing all complex correlations between the input and the output data in machine learning approaches remains a challenge. Beyond that, even though it is known that spatiotemporal, auto-temporal, and temperature data is influential to the fluctuations of acceleration data, using all three correlations in machine-learning approaches simultaneously remains a bottleneck. To address these challenges, this study presents a novel approach for dynamic response data recovery employing a Cauchy–Schwarz variational autoencoder with a hybrid data arrangement model called the spatial-auto-temporal thermal consistency model. The model uses a probabilistic encoder-decoder structure that leverages the rich expressiveness of a mixed Gaussian as the latent representation to reveal complex relationships between input and output data. The unique architecture of the deep learning model also enables it to be trained using spatiotemporal, auto-temporal, and temperature dependencies simultaneously in the SATTC configuration. The effectiveness of the proposed approach is demonstrated through a case study of field data from the Guangzhou New TV Tower (GNTT). The effects of input channels and measurement noise are also investigated. The quantitative analysis and modal identification results indicate that the proposed approach yields more accurate data reconstruction than VAE-ST, and GAN-ST, and slightly more accurate than CSVAE-ST.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120633"},"PeriodicalIF":5.6,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146191093","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

Mechanism-consistent probabilistic model for base resistance of long piles in soft soil using optimized sparse bayesian learning and factor correlation analysis 基于优化稀疏贝叶斯学习和因子相关分析的软土地基长桩基础阻力机制一致概率模型

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-29 DOI: 10.1016/j.measurement.2026.120639

Kailiang Weng , Mincai Jia , Gang Zhang , Qingyuan Zeng

Machine learning (ML) has been used for predicting pile base resistance, yet prediction for long piles remains difficult because significant base resistance is mobilized only after a certain degree of settlement. This study proposes a hybrid framework that combines sparse Bayesian learning (SBL) for probabilistic prediction with a Tent Chaotic Gaussian Sparrow Search Algorithm (TCGSSA) for hyperparameter tuning under cross-validation. Model performance and uncertainty quantification are evaluated using point and interval metrics on field data from 37 projects in Ho Chi Minh City, Vietnam. This study applies the maximum information coefficient (MIC) as a data-level diagnostic to quantify input–output associations and to check consistency with established geotechnical understanding. The framework delivers accurate point predictions together with sharp, well-covered intervals and compares favorably with the baselines considered. The diagnostics indicate that the displacement at the point of loading exerts the greatest influence on base resistance among the variables examined. The approach provides a mechanism-consistent, uncertainty-quantified tool for the design and assessment of long piles in soft soils.

机器学习（ML）已被用于预测桩基阻力，但对于长桩的预测仍然很困难，因为只有在一定程度的沉降后才会调动显著的基础阻力。本研究提出了一种混合框架，将用于概率预测的稀疏贝叶斯学习（SBL）与用于交叉验证超参数调优的Tent混沌高斯麻雀搜索算法（TCGSSA）相结合。利用越南胡志明市37个项目现场数据的点和区间指标，对模型性能和不确定性量化进行了评估。本研究将最大信息系数（MIC）作为数据级诊断来量化投入产出关联，并检查与已建立的岩土工程理解的一致性。该框架提供准确的点预测以及清晰、覆盖良好的间隔，并与所考虑的基线进行比较。诊断结果表明，在检查的变量中，加载点的位移对基础阻力的影响最大。该方法为软土中长桩的设计和评价提供了一种机制一致、不确定性量化的工具。

{"title":"Mechanism-consistent probabilistic model for base resistance of long piles in soft soil using optimized sparse bayesian learning and factor correlation analysis","authors":"Kailiang Weng , Mincai Jia , Gang Zhang , Qingyuan Zeng","doi":"10.1016/j.measurement.2026.120639","DOIUrl":"10.1016/j.measurement.2026.120639","url":null,"abstract":"<div><div>Machine learning (ML) has been used for predicting pile base resistance, yet prediction for long piles remains difficult because significant base resistance is mobilized only after a certain degree of settlement. This study proposes a hybrid framework that combines sparse Bayesian learning (SBL) for probabilistic prediction with a Tent Chaotic Gaussian Sparrow Search Algorithm (TCGSSA) for hyperparameter tuning under cross-validation. Model performance and uncertainty quantification are evaluated using point and interval metrics on field data from 37 projects in Ho Chi Minh City, Vietnam. This study applies the maximum information coefficient (MIC) as a data-level diagnostic to quantify input–output associations and to check consistency with established geotechnical understanding. The framework delivers accurate point predictions together with sharp, well-covered intervals and compares favorably with the baselines considered. The diagnostics indicate that the displacement at the point of loading exerts the greatest influence on base resistance among the variables examined. The approach provides a mechanism-consistent, uncertainty-quantified tool for the design and assessment of long piles in soft soils.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120639"},"PeriodicalIF":5.6,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146171991","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

A dual-stage dimensionless method for simultaneous estimation of electrical conductivity, lift-off, and thickness in Eddy Current Testing 在涡流测试中同时估计电导率、升力和厚度的双级无量纲方法

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-29 DOI: 10.1016/j.measurement.2026.120640

Alessandro Sardellitti , Filippo Milano , Vincenzo Mottola , Luigi Ferrigno , Antonello Tamburrino , Marco Laracca

This paper presents a dual-stage method for simultaneously estimating electrical conductivity, lift-off, and thickness in conductive samples using Eddy Current Testing (ECT) and dimensional analysis. The approach takes advantage of Buckingham’s

π

theorem to reformulate the physical model in a reduced set of dimensionless variables, allowing for a computationally efficient and geometrically intuitive inversion based on intersections of level curves.

The proposed method integrates these principles into a two-step estimation process that requires only one high-frequency and one low-frequency measurement. In the first stage, a high-frequency thickness-independent measurement is used for the simultaneous estimation of electrical conductivity and lift-off. In the second stage, a low-frequency measurement is used to recover the remaining thickness through one of the three proposed strategies, depending on which parameters of Stage 1 are retained.

The method was experimentally validated on six conductive samples under five lift-off conditions. The results show high accuracy, with relative errors below 2.5% for both electrical conductivity and lift-off, and 3.2% for thickness, and excellent repeatability with standard deviations generally lower than 1%. The observed repeatability in both stages supports implementation in either a multi-frequency or a single-frequency configuration, enabling simplified hardware, reduced measurement time, and real-time applicability. Overall, the methodology represents a practical and flexible framework suitable for integration into modern industrial ECT systems and Industry 4.0 quality-control environments.

本文提出了一种利用涡流测试（ECT）和量纲分析同时估计导电样品的电导率、升力和厚度的双阶段方法。该方法利用Buckingham的π定理在一组简化的无量纲变量中重新表述物理模型，允许基于水平曲线相交的计算效率和几何直观的反演。所提出的方法将这些原理集成到一个两步估计过程中，只需要一次高频和一次低频测量。在第一阶段，使用与厚度无关的高频测量来同时估计电导率和升力。在第二阶段，根据保留阶段1的哪些参数，使用低频测量通过三种提出的策略之一来恢复剩余的厚度。在6个导电样品上进行了5种发射条件下的实验验证。结果表明，该方法具有较高的准确度，电导率和离程的相对误差均小于2.5%，厚度的相对误差小于3.2%，重复性好，标准偏差一般小于1%。在这两个阶段中观察到的可重复性支持在多频或单频配置中实现，从而简化了硬件，减少了测量时间，并具有实时适用性。总体而言，该方法代表了一个实用而灵活的框架，适合集成到现代工业ECT系统和工业4.0质量控制环境中。

{"title":"A dual-stage dimensionless method for simultaneous estimation of electrical conductivity, lift-off, and thickness in Eddy Current Testing","authors":"Alessandro Sardellitti , Filippo Milano , Vincenzo Mottola , Luigi Ferrigno , Antonello Tamburrino , Marco Laracca","doi":"10.1016/j.measurement.2026.120640","DOIUrl":"10.1016/j.measurement.2026.120640","url":null,"abstract":"<div><div>This paper presents a dual-stage method for simultaneously estimating electrical conductivity, lift-off, and thickness in conductive samples using Eddy Current Testing (ECT) and dimensional analysis. The approach takes advantage of Buckingham’s <span><math><mi>π</mi></math></span> theorem to reformulate the physical model in a reduced set of dimensionless variables, allowing for a computationally efficient and geometrically intuitive inversion based on intersections of level curves.</div><div>The proposed method integrates these principles into a two-step estimation process that requires only one high-frequency and one low-frequency measurement. In the first stage, a high-frequency thickness-independent measurement is used for the simultaneous estimation of electrical conductivity and lift-off. In the second stage, a low-frequency measurement is used to recover the remaining thickness through one of the three proposed strategies, depending on which parameters of Stage 1 are retained.</div><div>The method was experimentally validated on six conductive samples under five lift-off conditions. The results show high accuracy, with relative errors below 2.5% for both electrical conductivity and lift-off, and 3.2% for thickness, and excellent repeatability with standard deviations generally lower than 1%. The observed repeatability in both stages supports implementation in either a multi-frequency or a single-frequency configuration, enabling simplified hardware, reduced measurement time, and real-time applicability. Overall, the methodology represents a practical and flexible framework suitable for integration into modern industrial ECT systems and Industry 4.0 quality-control environments.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120640"},"PeriodicalIF":5.6,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146190445","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

Hierarchical multiscale attention entropy-based fault identification for SDP images and ResNet of water diversion pumping station 基于分层多尺度关注熵的引水泵站SDP图像和ResNet故障识别

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-29 DOI: 10.1016/j.measurement.2026.120638

Jing Feng , Yu Tian , Xiaolian Liu , Zhaolong Xie , Hao Wang

To address fault signal recognition challenges in pumping station units under high noise and small sample conditions, this study proposes a measurement method based on Hierarchical Multiscale Attention Entropy (HMATE), a novel complexity index. HMATE, a quantifiable and interpretable measurand for non-stationary signal complexity, is integrated with attention-weighted entropy in the wavelet multiscale domain, enhancing noise resistance and stability. The HMATE results are visualized through Symmetrized Dot Pattern (SDP) and classified using ResNet50, achieving accurate identification of multiple fault types. Experimental results show over 95% accuracy under four types of strong noise and high reliability with small samples. t-SNE visualization confirms distinct separability of operational states. This study provides an effective fault diagnosis solution and introduces a metrologically grounded method for precise measurement of non-stationary signals.

针对高噪声、小样本条件下泵站机组故障信号识别的挑战，提出了一种基于层次多尺度注意熵（HMATE）的测量方法。HMATE是一种可量化且可解释的非平稳信号复杂度度量，在小波多尺度域中与注意加权熵相结合，增强了信号的抗噪性和稳定性。HMATE结果通过对称点图（SDP）可视化，并使用ResNet50进行分类，实现了多种故障类型的准确识别。实验结果表明，在4种强噪声条件下，该方法准确率超过95%，在小样本条件下具有较高的可靠性。t-SNE可视化证实了操作状态的明显可分性。该研究提供了有效的故障诊断解决方案，并介绍了一种精确测量非平稳信号的计量接地方法。

引用次数: 0

A novel denoising method for bearing vibration signals in rotating machinery based on CEEMDAN-PSO-TV combined AWR and DRC 基于CEEMDAN-PSO-TV结合AWR和DRC的旋转机械轴承振动信号去噪方法

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-29 DOI: 10.1016/j.measurement.2026.120641

Wenkai Yong , Yulong Li , Lijun Yu , Xiaogang Zhang

To address the issues of poor noise suppression in strong noise environments and the difficulty in preserving fault characteristics, this paper proposes an efficient denoising method combining CEEMDAN-PSO-TV (complete ensemble empirical mode decomposition with adaptive noise particle swarm optimization total variation)combined with AWR (adaptive weighting reconstruction) and DRC (dynamic residual compensation). First, the bearing vibration signal is decomposed using CEEMDAN (complete ensemble empirical mode decomposition with adaptive noise) to obtain a series of IMF (intrinsic mode function) components. The variance contribution rate of each IMF is calculated, and IMFs with a rate exceeding 5% are retained, while the remaining IMFs are filtered out. Second, each selected IMF undergoes denoising using the TV (total variation) algorithm. The PSO (particle swarm optimization) algorithm is incorporated during this process to adaptively select the regularization parameter λ of the TV algorithm for each selected IMF. Multiple initialization strategy enhances the PSO algorithm’s ability to find the global optimum, thereby enabling the adaptive selection of the optimal λ parameter for each IMF. Third, the denoised IMFs undergo AWR, DRC is then applied to the reconstructed signal to enhance both the denoising effectiveness and fault feature retention capability of the bearing vibration signal, ultimately yielding the final denoised signal. The proposed method was applied to both simulated and actual bearing vibration signals, and its performance was compared against multiple denoising methods. The results demonstrate that the denoising method based on CEEMDAN-PSO-TV combined AWR and DRC significantly outperforms existing methods in both noise suppression and fault feature retention for bearing vibration signals, thereby providing a robust foundation for accurate fault feature extraction.

针对强噪声环境下噪声抑制差、故障特征难以保持的问题，本文提出了一种将CEEMDAN-PSO-TV（全系综经验模态分解与自适应噪声粒子群优化总变分）与AWR（自适应加权重构）和DRC（动态残差补偿）相结合的高效去噪方法。首先，对轴承振动信号进行CEEMDAN（含自适应噪声的完全集合经验模态分解）分解，得到一系列的内禀模态函数IMF分量；计算每个IMF的方差贡献率，保留贡献率超过5%的IMF，过滤掉剩余的IMF。其次，使用TV（总变分）算法对每个选定的IMF进行去噪。在此过程中引入粒子群优化算法（PSO），对每个选定的IMF自适应选择TV算法的正则化参数λ。多重初始化策略增强了粒子群算法寻找全局最优的能力，从而能够自适应地选择每个IMF的最优λ参数。第三，对去噪后的imf进行AWR处理，然后对重构信号进行DRC处理，增强轴承振动信号的去噪效果和故障特征保留能力，最终得到去噪后的信号。将该方法分别应用于模拟和实际轴承振动信号，并与多种去噪方法进行了性能比较。结果表明，基于CEEMDAN-PSO-TV结合AWR和DRC的轴承振动信号去噪方法在噪声抑制和故障特征保留方面明显优于现有方法，为准确提取故障特征提供了坚实的基础。

{"title":"A novel denoising method for bearing vibration signals in rotating machinery based on CEEMDAN-PSO-TV combined AWR and DRC","authors":"Wenkai Yong , Yulong Li , Lijun Yu , Xiaogang Zhang","doi":"10.1016/j.measurement.2026.120641","DOIUrl":"10.1016/j.measurement.2026.120641","url":null,"abstract":"<div><div>To address the issues of poor noise suppression in strong noise environments and the difficulty in preserving fault characteristics, this paper proposes an efficient denoising method combining CEEMDAN-PSO-TV (complete ensemble empirical mode decomposition with adaptive noise particle swarm optimization total variation)combined with AWR (adaptive weighting reconstruction) and DRC (dynamic residual compensation). First, the bearing vibration signal is decomposed using CEEMDAN (complete ensemble empirical mode decomposition with adaptive noise) to obtain a series of IMF (intrinsic mode function) components. The variance contribution rate of each IMF is calculated, and IMFs with a rate exceeding 5% are retained, while the remaining IMFs are filtered out. Second, each selected IMF undergoes denoising using the TV (total variation) algorithm. The PSO (particle swarm optimization) algorithm is incorporated during this process to adaptively select the regularization parameter <em>λ</em> of the TV algorithm for each selected IMF. Multiple initialization strategy enhances the PSO algorithm’s ability to find the global optimum, thereby enabling the adaptive selection of the optimal <em>λ</em> parameter for each IMF. Third, the denoised IMFs undergo AWR, DRC is then applied to the reconstructed signal to enhance both the denoising effectiveness and fault feature retention capability of the bearing vibration signal, ultimately yielding the final denoised signal. The proposed method was applied to both simulated and actual bearing vibration signals, and its performance was compared against multiple denoising methods. The results demonstrate that the denoising method based on CEEMDAN-PSO-TV combined AWR and DRC significantly outperforms existing methods in both noise suppression and fault feature retention for bearing vibration signals, thereby providing a robust foundation for accurate fault feature extraction.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120641"},"PeriodicalIF":5.6,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146096173","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 on real-time inversion method of rock strength parameters while drilling considering confining pressure effect 考虑围压影响的钻时岩石强度参数实时反演方法研究

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-29 DOI: 10.1016/j.measurement.2026.120611

Zhaoyu Wen , Mingming He , Yangping Yao , Haoteng Wang

The in-situ acquisition of deep rock mechanical properties remains a major challenge in underground engineering. In this study, a prediction model for triaxial compressive strength, cohesion, and internal friction angle was developed by incorporating confining-pressure effects. The model was validated using drilling experiments on four rock types under five confining pressures, demonstrating high accuracy and reliability. The relationships between drilling parameters—such as torque and thrust—and the torque–thrust slope in the frictional stage were examined. The results indicate that these parameters exhibit linear positive correlations with confining pressure. The predicted parameters enabled continuous reconstruction of the spatial distributions of rock strength, cohesion, and internal friction angle, with accuracies generally within 25%. The study further reveals the nonlinear evolution of deep rock strength, characterized by increasing cohesion and decreasing friction angle with depth. Overall, this work provides a new technical pathway for the in-situ mechanical characterization of deep formations and supports intelligent, real-time drilling decision-making.

深部岩石力学特性的原位获取一直是地下工程中的一大难题。本文建立了考虑围压效应的三轴抗压强度、黏聚力和内摩擦角预测模型。通过5种围压下4种岩石类型的钻井实验，验证了该模型的准确性和可靠性。研究了摩擦阶段扭矩和推力等钻井参数与扭矩-推力斜率之间的关系。结果表明，这些参数与围压呈线性正相关。预测参数可以连续重建岩石强度、黏聚力和内摩擦角的空间分布，精度一般在25%以内。研究进一步揭示了深部岩石强度的非线性演化特征，即黏聚力随深度增大，摩擦角随深度减小。总的来说，这项工作为深部地层的原位力学表征提供了新的技术途径，并支持智能、实时的钻井决策。

{"title":"Research on real-time inversion method of rock strength parameters while drilling considering confining pressure effect","authors":"Zhaoyu Wen , Mingming He , Yangping Yao , Haoteng Wang","doi":"10.1016/j.measurement.2026.120611","DOIUrl":"10.1016/j.measurement.2026.120611","url":null,"abstract":"<div><div>The in-situ acquisition of deep rock mechanical properties remains a major challenge in underground engineering. In this study, a prediction model for triaxial compressive strength, cohesion, and internal friction angle was developed by incorporating confining-pressure effects. The model was validated using drilling experiments on four rock types under five confining pressures, demonstrating high accuracy and reliability. The relationships between drilling parameters—such as torque and thrust—and the torque–thrust slope in the frictional stage were examined. The results indicate that these parameters exhibit linear positive correlations with confining pressure. The predicted parameters enabled continuous reconstruction of the spatial distributions of rock strength, cohesion, and internal friction angle, with accuracies generally within 25%. The study further reveals the nonlinear evolution of deep rock strength, characterized by increasing cohesion and decreasing friction angle with depth. Overall, this work provides a new technical pathway for the in-situ mechanical characterization of deep formations and supports intelligent, real-time drilling decision-making.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120611"},"PeriodicalIF":5.6,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172036","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

High precision measurement and monitoring of the magnetic dominance and annular air shield impact on thermal energy and spray flow field for a coaxial liquid burner 同轴液体燃烧器磁优势和环形空气屏蔽对热能和喷雾流场影响的高精度测量与监测

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-29 DOI: 10.1016/j.measurement.2026.120606

Karim Emara , Ahmed Mahfouz M. M. Abd-Elgawad , Mohamed S. Gad , Ahmed Emara

The supremacy of magnetic field impact on spray and flame characteristics of traditional, renewable, and alternative liquid fuels has been rarely investigated particularly with annular air shields. Premeasurements are conducted like Fourier Transform Infrared Spectroscopy, Gas Chromatography-Mass Spectrometry, to characterize the physical and chemical properties for tested fuels. These measurements are carried out to characterize each fuel type and investigating the effect of changing some parameters. The current measurement’s purpose to address this research gap measurements for light diesel oil (LDO), waste cooking oil, waste tire pyrolysis oil, and their blends with the former (5% and 10% on a mass basis). The impact of magnetic field on flow field were conducted using the particle image velocimetry laser system under stagnant atmospheric conditions. Thermal and visual flame measurements are directed to visualize and predict flame temperatures. The findings emphasize that Ionization, de-clustering, and realignment will therefore result in an increase in the kinetic energy of the free electrons in the LDO fuel. Those phenomena, which include ionization and de-clustering, will speed up the vaporization and combining of oxygen atoms with the ionized hydrocarbon chain of LDO fuel to obtain a greater reaction rate during the combustion process, hence raising the combustion enthalpy. The magnetic field speeds up the droplets, facilitating better penetration into the combustion space and enhancing overall combustion efficiency. Moreover, the magnetic field would lower the system’s initial cost and enhance power consumption.

磁场对传统、可再生和可替代液体燃料的喷雾和火焰特性的影响一直很少被研究，特别是在环形空气屏蔽的情况下。预测量如傅里叶变换红外光谱，气相色谱-质谱，以表征物理和化学性质的测试燃料。这些测量是用来表征每种燃料类型，并研究改变某些参数的影响。当前测量的目的是解决轻柴油（LDO）、废食用油、废轮胎裂解油及其与前者的混合物（5%和10%的质量基础）的研究差距测量。在停滞大气条件下，利用粒子成像激光测速系统研究了磁场对流场的影响。热和目视火焰测量的目的是可视化和预测火焰温度。研究结果强调，电离、去聚和重新排列将导致LDO燃料中自由电子动能的增加。这些现象包括电离和脱簇等，会加速LDO燃料中氧原子与电离烃链的汽化结合，从而在燃烧过程中获得更大的反应速率，从而提高燃烧焓。磁场加速液滴，有利于更好地渗透到燃烧空间，提高整体燃烧效率。此外，磁场将降低系统的初始成本并提高功耗。

{"title":"High precision measurement and monitoring of the magnetic dominance and annular air shield impact on thermal energy and spray flow field for a coaxial liquid burner","authors":"Karim Emara , Ahmed Mahfouz M. M. Abd-Elgawad , Mohamed S. Gad , Ahmed Emara","doi":"10.1016/j.measurement.2026.120606","DOIUrl":"10.1016/j.measurement.2026.120606","url":null,"abstract":"<div><div>The supremacy ofmagnetic field impact on spray and flame characteristics of traditional, renewable, and alternative liquid fuels has been rarely investigated particularly with annular air shields. Premeasurements are conducted like Fourier Transform Infrared Spectroscopy, Gas Chromatography-Mass Spectrometry, to characterize the physical and chemical properties for tested fuels. These measurements are carried out to characterize each fuel type and investigating the effect of changing some parameters. The current measurement’s purpose to address this research gap measurements for light diesel oil (LDO), waste cooking oil, waste tire pyrolysis oil, and their blendswith the former (5% and 10% on a mass basis). The impact of magnetic field on flow field were conducted usingthe particle image velocimetry laser systemunder stagnant atmospheric conditions. Thermal and visual flame measurements are directed to visualize and predict flame temperatures. The findings emphasize that Ionization, de-clustering, and realignment will therefore result in an increase in the kinetic energy of the free electrons in the LDO fuel. Those phenomena, which include ionization and de-clustering, will speed up the vaporization and combining of oxygen atoms with the ionized hydrocarbon chain of LDO fuel to obtain a greater reaction rate during the combustion process, hence raising the combustion enthalpy. The magnetic field speeds up the droplets, facilitating better penetration into the combustion space and enhancing overall combustion efficiency. Moreover, the magnetic field would lower the system’s initial cost and enhance power consumption.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120606"},"PeriodicalIF":5.6,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146190444","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

Optimising image acquisition geometry by uncrewed aerial vehicles: A framework for 3D reconstruction of coastal cliffs 优化图像采集几何由无人驾驶飞行器：一个框架的三维重建海岸悬崖

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-29 DOI: 10.1016/j.measurement.2026.120626

Diogo Gonçalves , Gil Gonçalves , Maria da Conceição Cunha , Umberto Andriolo

The complex geometry of coastal cliffs poses challenges for achieving complete and detailed 3D reconstruction by Uncrewed Aerial Vehicle (UAV)-based imagery. This study proposes a novel four-step framework to optimise the UAV image acquisition geometry and flight path. It introduces curvature-based surface simplification and a linear quality metric to guide viewpoint selection, while guaranteeing consistent pixel size across images.

The optimised flight path was verified in the field. Only 2% of the candidate viewpoints comprised the optimised set, with an 84% reduction in flight duration. The acquired images generated a 3D point cloud that captured the entire cliff face, with a mean surface density of 3076 points per m². In addition, acquiring images with a consistent ground sample distance across pixels improved the performance of Structure from Motion and Multi-View Stereo techniques in terms of geometric quality and the surface density of the 3D point cloud.

This study demonstrates that optimising UAV flight paths allows for faster and more effective surveys of complex coastal environments. It promotes the use of remote sensing technologies for cliff inspection and monitoring, thereby contributing for risk assessment and coastal management.

沿海悬崖的复杂几何形状为实现基于无人机（UAV）的图像的完整和详细的3D重建提出了挑战。该研究提出了一种新的四步框架来优化无人机图像采集几何形状和飞行路径。它引入了基于曲率的表面简化和线性质量度量来指导视点选择，同时保证了图像之间像素大小的一致。优化后的飞行路径在现场进行了验证。只有2%的候选视点构成了优化集，飞行时间减少了84%。获取的图像生成了一个3D点云，它捕获了整个悬崖表面，平均表面密度为每平方米3076个点。此外，从几何质量和三维点云的表面密度方面来看，获取具有一致的地面样本距离的图像可以改善Structure from Motion和Multi-View Stereo技术的性能。这项研究表明，优化无人机飞行路径可以更快、更有效地调查复杂的沿海环境。它促进使用遥感技术进行悬崖检查和监测，从而有助于风险评估和沿海管理。

{"title":"Optimising image acquisition geometry by uncrewed aerial vehicles: A framework for 3D reconstruction of coastal cliffs","authors":"Diogo Gonçalves , Gil Gonçalves , Maria da Conceição Cunha , Umberto Andriolo","doi":"10.1016/j.measurement.2026.120626","DOIUrl":"10.1016/j.measurement.2026.120626","url":null,"abstract":"<div><div>The complex geometry of coastal cliffs poses challenges for achieving complete and detailed 3D reconstruction by Uncrewed Aerial Vehicle (UAV)-based imagery. This study proposes a novel four-step framework to optimise the UAV image acquisition geometry and flight path. It introduces curvature-based surface simplification and a linear quality metric to guide viewpoint selection, while guaranteeing consistent pixel size across images.</div><div>The optimised flight path was verified in the field. Only 2% of the candidate viewpoints comprised the optimised set, with an 84% reduction in flight duration. The acquired images generated a 3D point cloud that captured the entire cliff face, with a mean surface density of 3076 points per m<sup>2</sup>. In addition, acquiring images with a consistent ground sample distance across pixels improved the performance of Structure from Motion and Multi-View Stereo techniques in terms of geometric quality and the surface density of the 3D point cloud.</div><div>This study demonstrates that optimising UAV flight paths allows for faster and more effective surveys of complex coastal environments. It promotes the use of remote sensing technologies for cliff inspection and monitoring, thereby contributing for risk assessment and coastal management.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120626"},"PeriodicalIF":5.6,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172039","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

High-precision positioning of LTO magnetic track by four-dimensional topographic measurement 基于四维地形测量的LTO磁迹高精度定位

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-29 DOI: 10.1016/j.measurement.2026.120635

Bo Zhang , Zexiao Li , Weisheng Cheng, Xiaodong Zhang

To address the demand for 6-DOF pose measurement of multi-planar composite material micro-components in high-precision Linear Tape-Open (LTO) storage technology, this paper proposes a pose measurement method based on four-dimensional (4D) data fusion. A system integrating high-precision 3D measurement sensors with precision motion control axes was designed and constructed, enabling precision recognition of component feature regions by fusing 3D point cloud data with optical signals. Furthermore, a standard template matching model based on nonlinear optimization algorithms was developed to effectively solve the challenge of precise alignment of multiple-planar components (MPC) under 6-DOF. Experimental results demonstrate that this method achieves stable and high-precision measurements in complex environments, with a maximum standard deviation of relative displacement error in the most significant direction at 1.5 μm, the minimum standard deviation in other directions at 20 nm, and a maximum angular deviation standard deviation of less than 0.05°. This approach enables the differentiation and measurement of composite materials, meets the stringent requirements of industrial multi-component precision measurement, and provides critical technical support for enhancing the reliability and performance of Linear Tape-Open (LTO) storage systems.

针对高精度线性开带（LTO）存储技术中多平面复合材料微部件6自由度位姿测量的需求，提出了一种基于四维数据融合的位姿测量方法。设计并构建了高精度三维测量传感器与精密运动控制轴相结合的系统，通过将三维点云数据与光信号融合，实现对部件特征区域的精确识别。在此基础上，建立了基于非线性优化算法的标准模板匹配模型，有效地解决了6自由度下多平面零件精确对准的难题。实验结果表明，该方法在复杂环境下实现了稳定的高精度测量，相对位移误差在最显著方向的最大标准偏差为1.5 μm，其他方向的最小标准偏差为20 nm，最大角偏差标准偏差小于0.05°。这种方法能够区分和测量复合材料，满足工业多组分精密测量的严格要求，并为提高线性磁带打开（LTO）存储系统的可靠性和性能提供关键的技术支持。

{"title":"High-precision positioning of LTO magnetic track by four-dimensional topographic measurement","authors":"Bo Zhang , Zexiao Li , Weisheng Cheng, Xiaodong Zhang","doi":"10.1016/j.measurement.2026.120635","DOIUrl":"10.1016/j.measurement.2026.120635","url":null,"abstract":"<div><div>To address the demand for 6-DOF pose measurement of multi-planar composite material micro-components in high-precision Linear Tape-Open (LTO) storage technology, this paper proposes a pose measurement method based on four-dimensional (4D) data fusion. A system integrating high-precision 3D measurement sensors with precision motion control axes was designed and constructed, enabling precision recognition of component feature regions by fusing 3D point cloud data with optical signals. Furthermore, a standard template matching model based on nonlinear optimization algorithms was developed to effectively solve the challenge of precise alignment of multiple-planar components (MPC) under 6-DOF. Experimental results demonstrate that this method achieves stable and high-precision measurements in complex environments, with a maximum standard deviation of relative displacement error in the most significant direction at 1.5 μm, the minimum standard deviation in other directions at 20 nm, and a maximum angular deviation standard deviation of less than 0.05°. This approach enables the differentiation and measurement of composite materials, meets the stringent requirements of industrial multi-component precision measurement, and provides critical technical support for enhancing the reliability and performance of Linear Tape-Open (LTO) storage systems.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120635"},"PeriodicalIF":5.6,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146172087","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 on concentration inversion method for mid-infrared laser remote sensing of H2S and CH4 mixed gases H2S和CH4混合气体中红外激光遥感浓度反演方法研究

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-29 DOI: 10.1016/j.measurement.2026.120600

Le Hao , Xiaowei Zhai , Kai Wang , Jun Li , Qingjie Zeng

Gas leak monitoring represents a critical component in the production, transportation, and processing of high-sulfur natural gas, playing a vital role in ensuring operational safety across all stages and enabling environmental impact assessment following potential leaks. This study addresses spectral interference challenges in mid-infrared laser gas monitoring systems by developing a gas concentration inversion model based on a mixed-Lorentzian approach. Focusing on the two primary constituents of high-sulfur natural gas − methane (CH₄) and hydrogen sulfide (H₂S) − we established an 8.309 μm central spectral line suitable for simultaneous detection of both gases and implemented a remote mid-infrared laser system.To resolve signal interference between CH₄ and H₂S during mixed-gas monitoring, we employed spectral line broadening techniques under simulated high-sulfur gas leakage conditions. This enabled effective deployment of the mixed-Lorentzian model for gas signal separation. The parameters derived from the separated Lorentzian components were subsequently integrated into our concentration inversion model, achieving successful decomposition of mixed infrared laser signals.System stability evaluations demonstrated that our mixed-Lorentzian separation model effectively resolves composite gas signals while preserving absorption feature integrity. The model achieved correlation coefficients of 0.9541 for CH₄ and 0.9591 for H₂S, both exceeding the 0.95 threshold. These results confirm the method’s accuracy in simultaneous monitoring of CH₄ and H₂S concentrations within high-sulfur natural gas environments. This methodology shows significant potential for extension to similar challenges across the energy sector.

气体泄漏监测是高硫天然气生产、运输和加工的关键组成部分，在确保所有阶段的操作安全以及在潜在泄漏后进行环境影响评估方面发挥着至关重要的作用。本研究通过开发一种基于混合洛伦兹方法的气体浓度反演模型，解决了中红外激光气体监测系统中的光谱干扰问题。针对高硫天然气的两种主要成分甲烷（CH4）和硫化氢（H2S），建立了适合于同时检测两种气体的8.309 μm中心谱线，并实现了远程中红外激光系统。为了解决混合气体监测中CH4和H2S之间的信号干扰，我们在模拟高硫气体泄漏条件下采用了谱线展宽技术。这使得混合洛伦兹模型能够有效地用于气体信号分离。从分离的洛伦兹分量中得到的参数随后被整合到我们的浓度反演模型中，实现了混合红外激光信号的成功分解。系统稳定性评估表明，我们的混合洛伦兹分离模型在保持吸收特征完整性的同时有效地分解了复合气体信号。模型对CH4和H2S的相关系数分别为0.9541和0.9591，均超过了0.95的阈值。这些结果证实了该方法在高硫天然气环境中同时监测CH4和H2S浓度的准确性。这种方法显示出将类似挑战推广到整个能源部门的巨大潜力。

{"title":"Research on concentration inversion method for mid-infrared laser remote sensing of H2S and CH4 mixed gases","authors":"Le Hao , Xiaowei Zhai , Kai Wang , Jun Li , Qingjie Zeng","doi":"10.1016/j.measurement.2026.120600","DOIUrl":"10.1016/j.measurement.2026.120600","url":null,"abstract":"<div><div>Gas leak monitoring represents a critical component in the production, transportation, and processing of high-sulfur natural gas, playing a vital role in ensuring operational safety across all stages and enabling environmental impact assessment following potential leaks. This study addresses spectral interference challenges in mid-infrared laser gas monitoring systems by developing a gas concentration inversion model based on a mixed-Lorentzian approach. Focusing on the two primary constituents of high-sulfur natural gas − methane (CH<sub>4</sub>) and hydrogen sulfide (H<sub>2</sub>S) − we established an 8.309 μm central spectral line suitable for simultaneous detection of both gases and implemented a remote mid-infrared laser <span><span>system.To</span><svg><path></path></svg></span> resolve signal interference between CH<sub>4</sub> and H<sub>2</sub>S during mixed-gas monitoring, we employed spectral line broadening techniques under simulated high-sulfur gas leakage conditions. This enabled effective deployment of the mixed-Lorentzian model for gas signal separation. The parameters derived from the separated Lorentzian components were subsequently integrated into our concentration inversion model, achieving successful decomposition of mixed infrared laser signals.System stability evaluations demonstrated that our mixed-Lorentzian separation model effectively resolves composite gas signals while preserving absorption feature integrity. The model achieved correlation coefficients of 0.9541 for CH<sub>4</sub> and 0.9591 for H<sub>2</sub>S, both exceeding the 0.95 threshold. These results confirm the method’s accuracy in simultaneous monitoring of CH<sub>4</sub> and H<sub>2</sub>S concentrations within high-sulfur natural gas environments. This methodology shows significant potential for extension to similar challenges across the energy sector.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120600"},"PeriodicalIF":5.6,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081029","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