Pub Date : 2026-05-05Epub Date: 2026-03-03DOI: 10.1016/j.measurement.2026.121037
Mingyu Gao , Fei Wang , Chuang Wei , Yulong Zhou , Zhipeng Liang , Guohui Yang , Honghao Yue , Junyan Liu
The non-destructive testing (NDT) of internal debonding defects in Polymethacrylimide (PMI) foam is challenged by the inherent low density and closed-cell structure of the material. This study proposes and validates a microwave near-field reflection inspection method using an open-ended rectangular waveguide (OERW) to address this challenge. A multi-layer dielectric electromagnetic model was first established to elucidate the detection mechanism. Key parameters of stand-off distance (SOD), operating frequency, and scanning step were then systematically optimized through numerical simulations and experimental studies to achieve high-sensitivity detection. An anisotropic diffusion filtering strategy was applied to suppress inherent Rayleigh-distributed speckle noise while preserving defect edge details. The developed automated inspection system under optimal parameters successfully imaged a specimen with 16 artificial defects of various sizes (Φ4−12 mm) and depths (h = 0.1–0.4 mm). Quantitative signal-to-noise ratio (SNR) analysis confirms the proposed filtering algorithm significantly outperforms conventional mean and Gaussian filters by substantially enhancing image quality and defect recognizability. This research provides a complete solution integrating theory, system development, and algorithms for the reliable inspection of internal defects in PMI thermal protection structures.
{"title":"Near-field microwave imaging and speckle noise suppression for internal debonding defects in thermal protection composite materials","authors":"Mingyu Gao , Fei Wang , Chuang Wei , Yulong Zhou , Zhipeng Liang , Guohui Yang , Honghao Yue , Junyan Liu","doi":"10.1016/j.measurement.2026.121037","DOIUrl":"10.1016/j.measurement.2026.121037","url":null,"abstract":"<div><div>The non-destructive testing (NDT) of internal debonding defects in Polymethacrylimide (PMI) foam is challenged by the inherent low density and closed-cell structure of the material. This study proposes and validates a microwave near-field reflection inspection method using an open-ended rectangular waveguide (OERW) to address this challenge. A multi-layer dielectric electromagnetic model was first established to elucidate the detection mechanism. Key parameters of stand-off distance (SOD), operating frequency, and scanning step were then systematically optimized through numerical simulations and experimental studies to achieve high-sensitivity detection. An anisotropic diffusion filtering strategy was applied to suppress inherent Rayleigh-distributed speckle noise while preserving defect edge details. The developed automated inspection system under optimal parameters successfully imaged a specimen with 16 artificial defects of various sizes (Φ4−12 mm) and depths (<em>h</em> = 0.1–0.4 mm). Quantitative signal-to-noise ratio (SNR) analysis confirms the proposed filtering algorithm significantly outperforms conventional mean and Gaussian filters by substantially enhancing image quality and defect recognizability. This research provides a complete solution integrating theory, system development, and algorithms for the reliable inspection of internal defects in PMI thermal protection structures.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121037"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388110","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}
Pub Date : 2026-05-05Epub Date: 2026-02-28DOI: 10.1016/j.measurement.2026.121001
Zhen-Ying Xu, Chang Wang, Ying-Jun Lei, Yue Yang, Le Yin, Yu Guo, Li-Ling Han, Yun Wang
Current defect detection systems struggle to meet the requirements for high-precision and high-speed identification of multi-scale weld defects in large-sized thin-walled metal tubes. This paper proposes an intelligent weld defect detection system for thin-walled metal tubes, centered on a novel defect detection algorithm (A Dual-Stage SPD-Enhanced Network with TripletAttention mechanism). The proposed algorithm incorporates SPDConv modules to enhance its capability in detecting small targets. The TripletAttention mechanism optimizes feature fusion and lowers computational complexity and the WIoU loss function promotes generalization ability. The proposed system achieves [email protected] of 90.2% on the thin-walled metal tube weld defects dataset according to the experimental results. Notably, it attains 73.1% accuracy rate for small weld defects which is 6.3% better than the baseline models, and a frame rate of 118 frames per second (FPS), which is a significant improvement over current methods in terms of accuracy and efficiency. This detection system can be effectively deployed for non-destructive testing of various welded structures made of thin-walled metal tube, particularly demonstrating utility in weld quality inspection and defect identification for lightweight mobility device frames, such as bicycles, motorcycles, and wheelchairs.
{"title":"DualSPD-YOLO: A Dual-Stage SPD-Enhanced network with TripletAttention for High-Speed weld defect detection in Thin-Walled metal tubes","authors":"Zhen-Ying Xu, Chang Wang, Ying-Jun Lei, Yue Yang, Le Yin, Yu Guo, Li-Ling Han, Yun Wang","doi":"10.1016/j.measurement.2026.121001","DOIUrl":"10.1016/j.measurement.2026.121001","url":null,"abstract":"<div><div>Current defect detection systems struggle to meet the requirements for high-precision and high-speed identification of multi-scale weld defects in large-sized thin-walled metal tubes. This paper proposes an intelligent weld defect detection system for thin-walled metal tubes, centered on a novel defect detection algorithm (A Dual-Stage SPD-Enhanced Network with TripletAttention mechanism). The proposed algorithm incorporates SPDConv modules to enhance its capability in detecting small targets. The TripletAttention mechanism optimizes feature fusion and lowers computational complexity and the WIoU loss function promotes generalization ability. The proposed system achieves [email protected] of 90.2% on the thin-walled metal tube weld defects dataset according to the experimental results. Notably, it attains 73.1% accuracy rate for small weld defects which is 6.3% better than the baseline models, and a frame rate of 118 frames per second (FPS), which is a significant improvement over current methods in terms of accuracy and efficiency. This detection system can be effectively deployed for non-destructive testing of various welded structures made of thin-walled metal tube, particularly demonstrating utility in weld quality inspection and defect identification for lightweight mobility device frames, such as bicycles, motorcycles, and wheelchairs.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121001"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388280","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}
Extreme weather and earthquakes are boosting demand for slope protection on steep terrain. Retaining grid frame structures—lattice-type concrete grids—must be surveyed precisely, yet rope-work-based manual measurements are labour-intensive and hazardous. This paper presents an as-built management method that (1) separates grid frame and slope surfaces in point cloud data, (2) creates cross-sections, and (3) outputs width, height and frame center distance for every frame. Tests on a 60 m curved site in Fukushima showed mean dimensional error of 0.02 m, satisfying the Japanese guideline tolerances (±0.10 m for frame center distance and ± 0.03 m for both width and height), and an F1-score of 0.92 for slope surface detection with reliable rope removal. The workflow lowers field labour, enhances safety.
{"title":"Automatic as-built management of retaining grid frame structures using point cloud data","authors":"Yoshimasa Umehara , Yoshinori Tsukada , Kenji Nakamura , Satoshi Fujita , Tarou Yamanashi , Yasuhito Niina , Yutaka Matsubayashi , Ryuichi Imai","doi":"10.1016/j.measurement.2026.121018","DOIUrl":"10.1016/j.measurement.2026.121018","url":null,"abstract":"<div><div>Extreme weather and earthquakes are boosting demand for slope protection on steep terrain. Retaining grid frame structures—lattice-type concrete grids—must be surveyed precisely, yet rope-work-based manual measurements are labour-intensive and hazardous. This paper presents an as-built management method that (1) separates grid frame and slope surfaces in point cloud data, (2) creates cross-sections, and (3) outputs width, height and frame center distance for every frame. Tests on a 60 m curved site in Fukushima showed mean dimensional error of 0.02 m, satisfying the Japanese guideline tolerances (±0.10 m for frame center distance and ± 0.03 m for both width and height), and an F1-score of 0.92 for slope surface detection with reliable rope removal. The workflow lowers field labour, enhances safety.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121018"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388351","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}
Pub Date : 2026-05-05Epub Date: 2026-02-28DOI: 10.1016/j.measurement.2026.121011
Md. Zikrul Bari Chowdhury , Mohammad Tariqul Islam , Mohamad A. Alawad , Phumin Kirawanich , Badariah Bais , Mohamed Ouda , Yazeed Alkhrijah , Abdulmajeed M. Alenezi
Powdered food products such as milk powder, horlicks, lactogen, and coffee powder exhibit varying dielectric properties that can be leveraged for material characterization and quality monitoring. Conventional methods for analyzing such materials often involve complex, time-consuming procedures. This paper presents a novel dual-band octagonal symmetric metamaterial absorber designed to detect dielectric variations in powdered foods through high-sensitivity electromagnetic sensing. The absorber operates at resonant frequencies of 9.86 GHz and 12.50 GHz with a unit cell dimension of approximately 0.591 × 0.591 at the lower frequency, corresponding to an effective medium ratio of 1.69. The X-band is dedicated to powdered food sensing, while the Ku-band supports general microwave absorption applications. The structure achieves an absorption rate of up to 99.99% at both bands and maintains stable performance under oblique incidence angles up to 60°, demonstrating strong angular resilience. Numerical simulations validate the absorber’s electromagnetic response and confirm close alignment with theoretical predictions. The proposed design offers a compact, sensitive, and angularly stable solution involving dielectric property detection and electromagnetic wave absorption, making it suitable for wireless and sensing technologies.
{"title":"A novel oblique-incident stable dual-band octagonal symmetric metamaterial absorber for sensing applications","authors":"Md. Zikrul Bari Chowdhury , Mohammad Tariqul Islam , Mohamad A. Alawad , Phumin Kirawanich , Badariah Bais , Mohamed Ouda , Yazeed Alkhrijah , Abdulmajeed M. Alenezi","doi":"10.1016/j.measurement.2026.121011","DOIUrl":"10.1016/j.measurement.2026.121011","url":null,"abstract":"<div><div>Powdered food products such as milk powder, horlicks, lactogen, and coffee powder exhibit varying dielectric properties that can be leveraged for material characterization and quality monitoring. Conventional methods for analyzing such materials often involve complex, time-consuming procedures. This paper presents a novel dual-band octagonal symmetric metamaterial absorber designed to detect dielectric variations in powdered foods through high-sensitivity electromagnetic sensing. The absorber operates at resonant frequencies of 9.86 GHz and 12.50 GHz with a unit cell dimension of approximately 0.591 <span><math><msub><mi>λ</mi><mn>0</mn></msub></math></span> × 0.591 <span><math><msub><mi>λ</mi><mn>0</mn></msub></math></span> at the lower frequency, corresponding to an effective medium ratio of 1.69. The X-band is dedicated to powdered food sensing, while the Ku-band supports general microwave absorption applications. The structure achieves an absorption rate of up to 99.99% at both bands and maintains stable performance under oblique incidence angles up to 60°, demonstrating strong angular resilience. Numerical simulations validate the absorber’s electromagnetic response and confirm close alignment with theoretical predictions. The proposed design offers a compact, sensitive, and angularly stable solution involving dielectric property detection and electromagnetic wave absorption, making it suitable for wireless and sensing technologies.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121011"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388353","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}
Pub Date : 2026-05-05Epub Date: 2026-03-09DOI: 10.1016/j.measurement.2026.121083
Xin Tian , Zheng Yin , Dalong Tan , Yixin He , Changzhe Li , Tian Chen , Yijie Peng , Min Yang
In battery X-ray computed tomography (CT) imaging, artifacts emanating from internal metallic implants markedly distort electrode morphology, thereby impeding precise interpretation and quantification of internal structural features. Although sinogram‑domain correction techniques can mitigate these artifacts to some extent, they frequently introduce secondary artifacts due to localized correction errors and entail substantial computational overhead. Meanwhile, most deep learning metal artifact reduction (MAR) methods treat the task as a generic image‑restoration problem, overlooking the inherent physics of CT imaging and relying solely on off‑the‑shelf network components, which limits their interpretability. To overcome these challenges, we propose FID‑Net, an interpretable Fourier and image domain convolutional dictionary network for industrial CT metal artifact reduction. FID‑Net constructs spatial and spectral dictionaries in parallel and uses the fast Fourier transform as an efficient bridge to jointly capture global context and local details, iteratively refining dictionary coefficients to accurately model and remove metal artifacts while fully preserving electrode morphology. To validate the effectiveness of the method, we constructed a real battery CT dataset and conducted comparative experiments on extensive synthetic and real data, and the results show that the proposed FID‑Net demonstrates superior performance in MAR effectiveness and structural fidelity.
{"title":"FID-Net: an interpretable fourier and image domain convolutional dictionary network for industrial CT metal artifact reduction","authors":"Xin Tian , Zheng Yin , Dalong Tan , Yixin He , Changzhe Li , Tian Chen , Yijie Peng , Min Yang","doi":"10.1016/j.measurement.2026.121083","DOIUrl":"10.1016/j.measurement.2026.121083","url":null,"abstract":"<div><div>In battery X-ray computed tomography (CT) imaging, artifacts emanating from internal metallic implants markedly distort electrode morphology, thereby impeding precise interpretation and quantification of internal structural features. Although sinogram‑domain correction techniques can mitigate these artifacts to some extent, they frequently introduce secondary artifacts due to localized correction errors and entail substantial computational overhead. Meanwhile, most deep learning metal artifact reduction (MAR) methods treat the task as a generic image‑restoration problem, overlooking the inherent physics of CT imaging and relying solely on off‑the‑shelf network components, which limits their interpretability. To overcome these challenges, we propose FID‑Net, an interpretable Fourier and image domain convolutional dictionary network for industrial CT metal artifact reduction. FID‑Net constructs spatial and spectral dictionaries in parallel and uses the fast Fourier transform as an efficient bridge to jointly capture global context and local details, iteratively refining dictionary coefficients to accurately model and remove metal artifacts while fully preserving electrode morphology. To validate the effectiveness of the method, we constructed a real battery CT dataset and conducted comparative experiments on extensive synthetic and real data, and the results show that the proposed FID‑Net demonstrates superior performance in MAR effectiveness and structural fidelity.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121083"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388160","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}
Pub Date : 2026-05-05Epub Date: 2026-03-01DOI: 10.1016/j.measurement.2026.121014
Xiaoyu Ma , Long Ma , Guifu Huang , Igor Konyakhin , Xinhai Zou , Renpu Li , Xiao Feng , Junsen Yuan
Investigating the mathematical relationship between the structure of a nonstandard corner cube prism and its reflection-angle sensitivity can guide the engineering design of large-range autocollimators. However, four measurement challenges are introduced: (1) degradation of measurement resolution; (2) redundant and confusing intersection points among multiple cross images; (3) reduced stability of pixel assignment within overlapping regions; and (4) ambiguity in fitting the imaging profile within pixel aggregation regions. This study overcomes these challenges by optimizing the utilization of the multidimensional reflective surfaces of a nonstandard corner cube prism to modulate the reflection-angle sensitivity of each cross-shaped imaging beam, in addition to designing a quadruple algorithm fusion recognition mechanism. Experimental results indicate that, whereas the measurement range of the autocollimator increases by a factor of 10, the mean measurement accuracy of the pitch and yaw angles reaches 0.41″ and 0.44″, respectively, within a 350″ range. Meanwhile, the mean measurement accuracy of the angles are less than 4.34″ and 4.62″, respectively, within an extended range of 10°.
{"title":"Large-range high-accuracy autocollimator using non-standard corner cube prism and multi-cross imaging","authors":"Xiaoyu Ma , Long Ma , Guifu Huang , Igor Konyakhin , Xinhai Zou , Renpu Li , Xiao Feng , Junsen Yuan","doi":"10.1016/j.measurement.2026.121014","DOIUrl":"10.1016/j.measurement.2026.121014","url":null,"abstract":"<div><div>Investigating the mathematical relationship between the structure of a nonstandard corner cube prism and its reflection-angle sensitivity can guide the engineering design of large-range autocollimators. However, four measurement challenges are introduced: (1) degradation of measurement resolution; (2) redundant and confusing intersection points among multiple cross images; (3) reduced stability of pixel assignment within overlapping regions; and (4) ambiguity in fitting the imaging profile within pixel aggregation regions. This study overcomes these challenges by optimizing the utilization of the multidimensional reflective surfaces of a nonstandard corner cube prism to modulate the reflection-angle sensitivity of each cross-shaped imaging beam, in addition to designing a quadruple algorithm fusion recognition mechanism. Experimental results indicate that, whereas the measurement range of the autocollimator increases by a factor of 10, the mean measurement accuracy of the pitch and yaw angles reaches 0.41″ and 0.44″, respectively, within a 350″ range. Meanwhile, the mean measurement accuracy of the angles are less than 4.34″ and 4.62″, respectively, within an extended range of 10°.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121014"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388278","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}
Pub Date : 2026-05-05Epub Date: 2026-02-27DOI: 10.1016/j.measurement.2026.120998
Tingting Zhang, Yefeng Gu, Qinduan Zhang, Yubin Wei, Li Wang, Chunsheng Li, Wei Wang
Methane (CH4) has extensive applications in energy and chemical industries. However, its flammability and greenhouse effects necessitate timely detection of gas leakage for safety and environmental protection. This study presents a novel second harmonic signal denoising method for quartz-enhanced photoacoustic spectroscopy (QEPAS) CH4 gas sensors. The proposed method integrates successive variational mode decomposition (SVMD), permutation entropy (PE), and Savitzky-Golay (SG) filtering. The methodology involves initial signal decomposition via SVMD, followed by PE threshold implementation to preserve effective signals during reconstruction, and subsequent SG filtering for enhanced denoising. In addition, we develop a parameter optimization approach to determine an appropriate balance parameter for SVMD-based second harmonic signal decomposition. Both simulation and experimental results validate the superior performance of the SVMD-PE-SG method. The approach significantly improved the signal-to-noise ratio (SNR) from 54.58 to 298.49 and the correlation coefficient (R2) from 0.99682 to 0.99922, achieving a minimum detection limit (MDL) of 2.01 ppm for CH4 gas. Through Allan variance analysis, the MDL reaches 67 ppb when the integration time is 121 s. Implementation of the SVMD-PE-SG algorithm substantially enhanced the detection accuracy of the CH4-QEPAS gas sensor.
甲烷（CH4）在能源和化学工业中有着广泛的应用。然而，其可燃性和温室效应需要及时检测气体泄漏，以确保安全和环保。提出了一种用于石英增强光声光谱（QEPAS） CH4气体传感器的二次谐波信号去噪方法。该方法集成了连续变分模态分解（SVMD）、置换熵（PE）和Savitzky-Golay （SG）滤波。该方法包括通过SVMD进行初始信号分解，然后通过PE阈值实现以在重建过程中保留有效信号，然后进行SG滤波以增强去噪。此外，我们开发了一种参数优化方法来确定合适的平衡参数α，用于基于支持向量机的二次谐波信号分解。仿真和实验结果都验证了svm - pe - sg方法的优越性能。该方法将CH4气体的信噪比（SNR）从54.58提高到298.49，相关系数（R2）从0.99682提高到0.99922，最低检出限（MDL）为2.01 ppm。通过Allan方差分析，当积分时间为121 s时，MDL达到67 ppb。SVMD-PE-SG算法的实现大大提高了CH4-QEPAS气体传感器的检测精度。
{"title":"SVMD-PE-SG for improved QEPAS-based methane concentration detection","authors":"Tingting Zhang, Yefeng Gu, Qinduan Zhang, Yubin Wei, Li Wang, Chunsheng Li, Wei Wang","doi":"10.1016/j.measurement.2026.120998","DOIUrl":"10.1016/j.measurement.2026.120998","url":null,"abstract":"<div><div>Methane (CH<sub>4</sub>) has extensive applications in energy and chemical industries. However, its flammability and greenhouse effects necessitate timely detection of gas leakage for safety and environmental protection. This study presents a novel second harmonic signal denoising method for quartz-enhanced photoacoustic spectroscopy (QEPAS) CH<sub>4</sub> gas sensors. The proposed method integrates successive variational mode decomposition (SVMD), permutation entropy (PE), and Savitzky-Golay (SG) filtering. The methodology involves initial signal decomposition via SVMD, followed by PE threshold implementation to preserve effective signals during reconstruction, and subsequent SG filtering for enhanced denoising. In addition, we develop a parameter optimization approach to determine an appropriate balance parameter <span><math><mrow><mi>α</mi></mrow></math></span> for SVMD-based second harmonic signal decomposition. Both simulation and experimental results validate the superior performance of the SVMD-PE-SG method. The approach significantly improved the signal-to-noise ratio (SNR) from 54.58 to 298.49 and the correlation coefficient (R<sup>2</sup>) from 0.99682 to 0.99922, achieving a minimum detection limit (MDL) of 2.01 ppm for CH<sub>4</sub> gas. Through Allan variance analysis, the MDL reaches 67 ppb when the integration time is 121 s. Implementation of the SVMD-PE-SG algorithm substantially enhanced the detection accuracy of the CH<sub>4</sub>-QEPAS gas sensor.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 120998"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388281","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}
The occurrence of multiple faults simultaneously in industrial machines is very complex. If maintenance operation is not carried out promptly, it leads to financial losses. Moreover, the unavailability of the multi-fault dataset imposes a problem for automated diagnosis model development. The model developed with a single sensor could not work with multiple sensor data and needs a separate fusion process. To tackle these problems, a robust methodology involving an Adaptive Information Fusion (AIF) process along with a Multi-Stream Convolutional Neural Network (MSCNN) is developed in this work. Initially, a large acoustics-based dataset is made by experimentally simulating 225 multi-fault conditions of motor, mechanical unbalance, system bearing and shaft misalignment at varying speeds. The acoustic signatures show distinct features with different fault conditions, passing through the AIF process for generating a time–frequency spectrum dataset using Wavelet Synchrosqueezed Transform (WSST). The spectrums are eventually used for training the optimised MSCNN model. A maximum overall classification accuracy of 97.19% is achieved. The performance of the methodology is rigorously tested under various background noise, data imbalance, speed conditions and with multi-sensor data fusion. The proposed method is useful for single as well as multiple-sensor data and can be helpful for incipient fault diagnosis in industries.
{"title":"Adaptive information fusion for multi-fault diagnosis using multi-stream convolutional neural network","authors":"R.K. Mishra , Anurag Choudhary , S. Fatima , A.R. Mohanty , B.K. Panigrahi","doi":"10.1016/j.measurement.2026.121043","DOIUrl":"10.1016/j.measurement.2026.121043","url":null,"abstract":"<div><div>The occurrence of multiple faults simultaneously in industrial machines is very complex. If maintenance operation is not carried out promptly, it leads to financial losses. Moreover, the unavailability of the multi-fault dataset imposes a problem for automated diagnosis model development. The model developed with a single sensor could not work with multiple sensor data and needs a separate fusion process. To tackle these problems, a robust methodology involving an Adaptive Information Fusion (AIF) process along with a Multi-Stream Convolutional Neural Network (MSCNN) is developed in this work. Initially, a large acoustics-based dataset is made by experimentally simulating 225 multi-fault conditions of motor, mechanical unbalance, system bearing and shaft misalignment at varying speeds. The acoustic signatures show distinct features with different fault conditions, passing through the AIF process for generating a time–frequency spectrum dataset using Wavelet Synchrosqueezed Transform (WSST). The spectrums are eventually used for training the optimised MSCNN model. A maximum overall classification accuracy of 97.19% is achieved. The performance of the methodology is rigorously tested under various background noise, data imbalance, speed conditions and with multi-sensor data fusion. The proposed method is useful for single as well as multiple-sensor data and can be helpful for incipient fault diagnosis in industries.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121043"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388334","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}
Pub Date : 2026-05-05Epub Date: 2026-03-06DOI: 10.1016/j.measurement.2026.121050
Han Wang , Xuejian Zhang , Zheyuan Zhang , Xiaobing Hu , Zhipeng Du
To address the trade-off between measurement accuracy, field of view, and efficiency in large-scale dimensional metrology, this paper proposes a high-precision three-dimensional measurement method integrating a laser tracker and a binocular vision system. A Multi-Target Spatial Pose Determination Device (MSPD) is designed to establish a unified global reference frame and a stable pose linkage among heterogeneous measurement data, thereby reducing error accumulation and operational overhead in multi-station measurements. A multi-physics, multi-objective optimization model is formulated for the spatial layout of MSPD targets by incorporating geometric visibility, pose solvability, and structural constraints. The target configuration is optimized using a non-dominated sorting genetic algorithm (NSGA-II) to improve omnidirectional visibility and pose determination robustness. Experimental results show that the optimized MSPD achieves approximately 90% visibility coverage and maintains reliable target identification under varying observation poses. Calibration validation experiments yield a mean reprojection error of 0.1331 mm, confirming the accuracy and repeatability of the coordinate transformation and pose prediction models. Comparative multi-station experiments further demonstrate that, with only one initial calibration, the proposed MSPD-based tracking integration strategy achieves an average point position error of 0.3336 mm and a closed-loop consistency error of 0.2363 mm, while limiting additional operation time per station to 1–2 min. In contrast, station-by-station repeated calibration requires 15–20 min per station, and feature-based alignment exhibits significant error accumulation. These results indicate that the proposed fusion measurement framework provides an effective balance between accuracy, robustness, and operational efficiency, making it suitable for engineering-scale measurement of large and complex components and high-precision robotic calibration.
{"title":"Design of a high-precision 3D measurement system via multi-source fusion and structural optimization of a Multi-Target Spatial Pose Determination Device (MSPD)","authors":"Han Wang , Xuejian Zhang , Zheyuan Zhang , Xiaobing Hu , Zhipeng Du","doi":"10.1016/j.measurement.2026.121050","DOIUrl":"10.1016/j.measurement.2026.121050","url":null,"abstract":"<div><div>To address the trade-off between measurement accuracy, field of view, and efficiency in large-scale dimensional metrology, this paper proposes a high-precision three-dimensional measurement method integrating a laser tracker and a binocular vision system. A Multi-Target Spatial Pose Determination Device (MSPD) is designed to establish a unified global reference frame and a stable pose linkage among heterogeneous measurement data, thereby reducing error accumulation and operational overhead in multi-station measurements. A multi-physics, multi-objective optimization model is formulated for the spatial layout of MSPD targets by incorporating geometric visibility, pose solvability, and structural constraints. The target configuration is optimized using a non-dominated sorting genetic algorithm (NSGA-II) to improve omnidirectional visibility and pose determination robustness. Experimental results show that the optimized MSPD achieves approximately 90% visibility coverage and maintains reliable target identification under varying observation poses. Calibration validation experiments yield a mean reprojection error of 0.1331 mm, confirming the accuracy and repeatability of the coordinate transformation and pose prediction models. Comparative multi-station experiments further demonstrate that, with only one initial calibration, the proposed MSPD-based tracking integration strategy achieves an average point position error of 0.3336 mm and a closed-loop consistency error of 0.2363 mm, while limiting additional operation time per station to 1–2 min. In contrast, station-by-station repeated calibration requires 15–20 min per station, and feature-based alignment exhibits significant error accumulation. These results indicate that the proposed fusion measurement framework provides an effective balance between accuracy, robustness, and operational efficiency, making it suitable for engineering-scale measurement of large and complex components and high-precision robotic calibration.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 121050"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147388335","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}
Pub Date : 2026-05-05Epub Date: 2026-03-04DOI: 10.1016/j.measurement.2026.120977
Tengsheng Zhang , Zixuan Liu , Mengmeng Chen , Namita Sahoo , Kaiming Zhou , Bing Sun , Zuxing Zhang , Xiankui Zeng , Lin Zhang
An ultrasensitive optical fiber sensor is developed for real-time and selective detection of lead (Pb2+) ions in aqueous environments. The device integrates a tapered two-mode fiber (T2MF) coated with a polydopamine–maleic acid (PDA–MA) film, which exhibits strong Pb2+ affinity through multidentate carboxyl coordination. The optimized taper geometry enables efficient excitation of the and modes, yeilding high refractive index sensitivity while retaining mechanical robustness. Upon Pb2+ binding, the local refractive index increases, including a redshift in the interference spectrum. Across concentrations from 2.07 to 207.2 ppb, the sensor demonstrates a linear spectral response (R20.99) with a peak sensitivity of 0.219 nm/ppb and a detection limit of 2.4 ppb—well below the WHO drinking water standard. Incorporating a cascaded fiber Bragg grating (FBG) enables dual-parameter demodulation for real-time temperature compensation, ensuring high measurement fidelity. This compact and low-cost platform demonstrates strong selectivity, excellent reproducibility, and scalable potential for in situ heavy metal monitoring in environmental and public health applications.
{"title":"Ultrasensitive lead ion detection using polydopamine–maleic acid coated two-mode fiber with cascaded fiber Bragg grating compensation","authors":"Tengsheng Zhang , Zixuan Liu , Mengmeng Chen , Namita Sahoo , Kaiming Zhou , Bing Sun , Zuxing Zhang , Xiankui Zeng , Lin Zhang","doi":"10.1016/j.measurement.2026.120977","DOIUrl":"10.1016/j.measurement.2026.120977","url":null,"abstract":"<div><div>An ultrasensitive optical fiber sensor is developed for real-time and selective detection of lead (Pb<sup>2+</sup>) ions in aqueous environments. The device integrates a tapered two-mode fiber (<em>T</em><sup>2</sup>MF) coated with a polydopamine–maleic acid (PDA–MA) film, which exhibits strong Pb<sup>2+</sup> affinity through multidentate carboxyl coordination. The optimized taper geometry enables efficient excitation of the <span><math><mrow><mi>H</mi><msub><mrow><mi>E</mi></mrow><mrow><mtext>11</mtext></mrow></msub></mrow></math></span> and <span><math><mrow><mi>H</mi><msub><mrow><mi>E</mi></mrow><mrow><mtext>12</mtext></mrow></msub></mrow></math></span> modes, yeilding high refractive index sensitivity while retaining mechanical robustness. Upon Pb<sup>2+</sup> binding, the local refractive index increases, including a redshift in the interference spectrum. Across concentrations from 2.07 to 207.2 ppb, the sensor demonstrates a linear spectral response (R<sup>2</sup><span><math><mo>></mo></math></span>0.99) with a peak sensitivity of 0.219 nm/ppb and a detection limit of 2.4 ppb—well below the WHO drinking water standard. Incorporating a cascaded fiber Bragg grating (FBG) enables dual-parameter demodulation for real-time temperature compensation, ensuring high measurement fidelity. This compact and low-cost platform demonstrates strong selectivity, excellent reproducibility, and scalable potential for in situ heavy metal monitoring in environmental and public health applications.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"272 ","pages":"Article 120977"},"PeriodicalIF":5.6,"publicationDate":"2026-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147387728","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号