Measurement最新文献_第7页

Giant Goos-Hänchen shifts via quasibound states in the continuum in metastructure-photonic crystals for ultrasensitive detection 超灵敏探测的超结构光子晶体中通过准束缚态的巨大Goos-Hänchen位移

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-26 DOI: 10.1016/j.measurement.2026.120545

Yu-Hang Sun , Jun-Jie Luo , Hai-Feng Zhang

The reflection-type bound states in the continuum (BICs) within the metastructure-photonic crystals (MPCs) architecture are realized by precisely engineering the excitation conditions of guided modes. Leveraging ultrahigh-quality factor (Q factor) quasi-BICs, giant bidirectional Goos-Hänchen (GH) shifts are achieved. While a magnification of four orders of magnitude relative to the operational wavelength (λ₀) is attainable within the standard detection range, our approach allows for precise technical tuning to achieve even greater enhancement up to five orders of magnitude as needed, far surpassing the performance of existing enhancement mechanisms. Critically, these quasi-BIC-mediated GH shifts are occurred at reflection peaks with near-unity amplitudes, ensuring experimental detectability without requiring sophisticated signal recovery techniques. Beyond this fundamental advancement, the rapid advancement of the photovoltaic sector has intensified demand for high-performance solar cells, where organic semiconductor polymers serve as critical enablers due to their tunable optoelectronic properties and processability. By leveraging the refractive index sensitivity of GH shifts, an organic semiconductor polymer sensor was engineered, demonstrating ultrahigh sensitivity (−9.57 × 10⁴ λ₀/RIU), quantifying the differential GH shifts variation rate per refractive index unit (RIU) ultra-wide and continuous detection range (n = 1.7 ∼ 2.3), and high resolution (Δn_min = 0.00001). This system enables precise identification of polymer compositions by correlating GH shifts magnitudes with dielectric permittivity gradients, thereby providing a powerful tool for material characterization in next-generation optoelectronic devices.

在超结构光子晶体（MPCs）结构中，连续介质中的反射型束缚态是通过精确设计引导模的激发条件来实现的。利用超高质量因子（Q因子）准bic，实现了巨大的双向Goos-Hänchen （GH）位移。虽然在标准检测范围内，相对于工作波长（λ0）可以实现4个数量级的放大，但我们的方法允许精确的技术调整，以实现更大的增强，根据需要可达到5个数量级，远远超过现有增强机制的性能。关键的是，这些准bic介导的GH移位发生在反射峰上，其振幅接近一致，确保了实验可探测性，而不需要复杂的信号恢复技术。除了这一基本进步之外，光伏行业的快速发展也增加了对高性能太阳能电池的需求，其中有机半导体聚合物由于其可调谐的光电特性和可加工性而成为关键的推动因素。利用GH位移的折射率灵敏度，设计了一种有机半导体聚合物传感器，具有超高灵敏度（−9.57 × 104 λ0/RIU），量化了每折射率单位（RIU）的差分GH位移变化率，超宽连续检测范围（n = 1.7 ~ 2.3）和高分辨率（Δnmin = 0.00001）。该系统通过将GH位移幅度与介电常数梯度相关联来精确识别聚合物成分，从而为下一代光电器件的材料表征提供了强大的工具。

{"title":"Giant Goos-Hänchen shifts via quasibound states in the continuum in metastructure-photonic crystals for ultrasensitive detection","authors":"Yu-Hang Sun , Jun-Jie Luo , Hai-Feng Zhang","doi":"10.1016/j.measurement.2026.120545","DOIUrl":"10.1016/j.measurement.2026.120545","url":null,"abstract":"<div><div>The reflection-type bound states in the continuum (BICs) within the metastructure-photonic crystals (MPCs) architecture are realized by precisely engineering the excitation conditions of guided modes. Leveraging ultrahigh-quality factor (<em>Q</em> factor) quasi-BICs, giant bidirectional Goos-Hänchen (GH) shifts are achieved. While a magnification of four orders of magnitude relative to the operational wavelength (<em>λ<sub>0</sub></em>) is attainable within the standard detection range, our approach allows for precise technical tuning to achieve even greater enhancement up to five orders of magnitude as needed, far surpassing the performance of existing enhancement mechanisms. Critically, these quasi-BIC-mediated GH shifts are occurred at reflection peaks with near-unity amplitudes, ensuring experimental detectability without requiring sophisticated signal recovery techniques. Beyond this fundamental advancement, the rapid advancement of the photovoltaic sector has intensified demand for high-performance solar cells, where organic semiconductor polymers serve as critical enablers due to their tunable optoelectronic properties and processability. By leveraging the refractive index sensitivity of GH shifts, an organic semiconductor polymer sensor was engineered, demonstrating ultrahigh sensitivity (−9.57 × 10<sup>4</sup> <em>λ</em><sub>0</sub>/RIU), quantifying the differential GH shifts variation rate per refractive index unit (RIU) ultra-wide and continuous detection range (<em>n</em> = 1.7 ∼ 2.3), and high resolution (Δ<em>n</em><sub>min</sub> = 0.00001). This system enables precise identification of polymer compositions by correlating GH shifts magnitudes with dielectric permittivity gradients, thereby providing a powerful tool for material characterization in next-generation optoelectronic devices.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120545"},"PeriodicalIF":5.6,"publicationDate":"2026-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081402","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

Soft microfluidic optical waveguide transducer based on pressure balanced stable two-phase flow for displacement and pressure measurement 基于压力平衡稳定两相流的软微流体光波导传感器用于位移和压力测量

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-26 DOI: 10.1016/j.measurement.2026.120569

Hao Tian, Hong Xu

Microfluidic sensors are known for accuracy in fluid control and versatility in chemical component sensing, making them potential candidates for integration into soft robotic sensing tasks. However, traditional microfluidic sensors often need injection process to load the sample, and requires external signal excitation and processing hardware, such as an electrochemical station, to achieve full functionality. These prerequisites make the sensor integration with a robotic system difficult. Therefore, a standalone soft sensor, with signal excitation and parameter transduction functions fully integrated onboard, can be of great value. This paper seeks a new way to convert the displacement and pressure signals into electrical signals via a novel soft microfluidic optical waveguide transducer. The proposed device is fully self-contained, manufactured using laser lithography and UV-assisted bounding, creating a stable two-phase fluid within a flow channel, balanced by two opposing chambers. By creating a pressure differential across the two chambers, the interface of the two-phase fluid travels, achieving variable displacement optical waveguide function. Finally, using a photovoltaic cell, the optical signal can be converted into voltage readings, achieving the sensing goals. Prototype sensors were fabricated in lab, and the key transduction principles were validated against experiment. Initial results have shown that the proposed sensor can achieve a displacement sensitivity of 0.6263 V/mm and a pressure sensitivity of 0.5809

{\times 10}^{- 5}

V/Pa in standalone tests. The maximum measurement uncertainty was found to be

\pm

5.74%.

微流体传感器以流体控制的准确性和化学成分传感的多功能性而闻名，使其成为集成到软机器人传感任务中的潜在候选者。然而，传统的微流体传感器往往需要注射过程来加载样品，并且需要外部信号激励和处理硬件，如电化学站，才能实现全部功能。这些先决条件使得传感器与机器人系统的集成变得困难。因此，一个独立的软传感器，与信号激励和参数转导功能完全集成在板载，可以是很有价值的。本文通过一种新型软质微流体光波导换能器，寻求一种将位移和压力信号转换为电信号的新方法。该装置是完全独立的，使用激光光刻和紫外线辅助边界制造，在流动通道内产生稳定的两相流体，由两个相对的腔室平衡。通过在两个腔室之间产生压差，两相流体的界面移动，实现可变位移光波导功能。最后利用光伏电池将光信号转换成电压读数，实现传感目的。在实验室制作了传感器样机，并通过实验验证了关键的转导原理。初步试验结果表明，该传感器的位移灵敏度为0.6263 V/mm，压力灵敏度为0.5809×10-5 V/Pa。测定的最大不确定度为±5.74%。

{"title":"Soft microfluidic optical waveguide transducer based on pressure balanced stable two-phase flow for displacement and pressure measurement","authors":"Hao Tian, Hong Xu","doi":"10.1016/j.measurement.2026.120569","DOIUrl":"10.1016/j.measurement.2026.120569","url":null,"abstract":"<div><div>Microfluidic sensors are known for accuracy in fluid control and versatility in chemical component sensing, making them potential candidates for integration into soft robotic sensing tasks. However, traditional microfluidic sensors often need injection process to load the sample, and requires external signal excitation and processing hardware, such as an electrochemical station, to achieve full functionality. These prerequisites make the sensor integration with a robotic system difficult. Therefore, a standalone soft sensor, with signal excitation and parameter transduction functions fully integrated onboard, can be of great value. This paper seeks a new way to convert the displacement and pressure signals into electrical signals via a novel soft microfluidic optical waveguide transducer. The proposed device is fully self-contained, manufactured using laser lithography and UV-assisted bounding, creating a stable two-phase fluid within a flow channel, balanced by two opposing chambers. By creating a pressure differential across the two chambers, the interface of the two-phase fluid travels, achieving variable displacement optical waveguide function. Finally, using a photovoltaic cell, the optical signal can be converted into voltage readings, achieving the sensing goals. Prototype sensors were fabricated in lab, and the key transduction principles were validated against experiment. Initial results have shown that the proposed sensor can achieve a displacement sensitivity of 0.6263 V/mm and a pressure sensitivity of 0.5809<span><math><msup><mrow><mo>×</mo><mn>10</mn></mrow><mrow><mo>-</mo><mn>5</mn></mrow></msup></math></span> V/Pa in standalone tests. The maximum measurement uncertainty was found to be <span><math><mo>±</mo></math></span> 5.74%.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120569"},"PeriodicalIF":5.6,"publicationDate":"2026-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081563","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

Error distribution prediction of five-axis on-machine measurement for aerospace structural parts 航空结构件五轴测量误差分布预测

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-26 DOI: 10.1016/j.measurement.2026.120573

Qixin Zhuang , Neng Wan , Bin Cheng , Dao Wang , Liangliang Wang , Xueqin Kou

On-machine measurement (OMM) using a touch-trigger probe (TTP) can identify the allowance distribution on aerospace structural parts during adaptive machining. Therefore, accurate description and prediction of OMM accuracy are essential to improve the machining accuracy of parts. However, existing methods for evaluating OMM results mostly focus on the average values while neglecting their dispersion. To address this issue, a five-axis OMM error distribution prediction method for aerospace structural parts is proposed. The expectation and standard uncertainty (EASU) are used to quantify the trueness and precision of the measurement results, respectively. Furthermore, the EASU of the machine tool (MT) positioning error, probe eccentricity error and workpiece alignment error in the measurement system are evaluated. Subsequently, the propagation model of multi-source error EASU during the measurement is derived. Notably, the propagation model addresses the correlation between the error sources and between the derived variables in the propagation process, and retains the second-order Taylor series term, thereby improving the accuracy of the prediction model under nonlinear conditions. Taking the typical geometric dimensioning and tolerancing (GD&T) on aerospace structural parts as an example, an OMM error distribution prediction model for GD&T is established. Finally, both numerical simulations and experiments were conducted on an aero-engine casing and impeller to verify the effectiveness and extensibility of the proposed method.

采用触控触发探针（TTP）的机内测量（OMM）可以识别航天结构件在自适应加工过程中的余量分布。因此，准确地描述和预测OMM精度对于提高零件的加工精度至关重要。然而，现有的评价OMM结果的方法大多侧重于平均值，而忽略了其离散度。针对这一问题，提出了一种航空结构件的五轴OMM误差分布预测方法。期望不确定度和标准不确定度（EASU）分别用于量化测量结果的真实度和精密度。在此基础上，对测量系统中机床定位误差、测头偏心误差和工件对中误差的EASU进行了评估。在此基础上，推导了多源误差EASU在测量过程中的传播模型。值得注意的是，传播模型解决了传播过程中误差源之间和推导变量之间的相关性，并保留了二阶泰勒级数项，从而提高了非线性条件下预测模型的精度。以航空航天结构件典型几何尺寸公差（GD&；T）为例，建立了通用公差的OMM误差分布预测模型。最后，对某航空发动机机匣和叶轮进行了数值模拟和实验，验证了该方法的有效性和可扩展性。

{"title":"Error distribution prediction of five-axis on-machine measurement for aerospace structural parts","authors":"Qixin Zhuang , Neng Wan , Bin Cheng , Dao Wang , Liangliang Wang , Xueqin Kou","doi":"10.1016/j.measurement.2026.120573","DOIUrl":"10.1016/j.measurement.2026.120573","url":null,"abstract":"<div><div>On-machine measurement (OMM) using a touch-trigger probe (TTP) can identify the allowance distribution on aerospace structural parts during adaptive machining. Therefore, accurate description and prediction of OMM accuracy are essential to improve the machining accuracy of parts. However, existing methods for evaluating OMM results mostly focus on the average values while neglecting their dispersion. To address this issue, a five-axis OMM error distribution prediction method for aerospace structural parts is proposed. The expectation and standard uncertainty (EASU) are used to quantify the trueness and precision of the measurement results, respectively. Furthermore, the EASU of the machine tool (MT) positioning error, probe eccentricity error and workpiece alignment error in the measurement system are evaluated. Subsequently, the propagation model of multi-source error EASU during the measurement is derived. Notably, the propagation model addresses the correlation between the error sources and between the derived variables in the propagation process, and retains the second-order Taylor series term, thereby improving the accuracy of the prediction model under nonlinear conditions. Taking the typical geometric dimensioning and tolerancing (GD&T) on aerospace structural parts as an example, an OMM error distribution prediction model for GD&T is established. Finally, both numerical simulations and experiments were conducted on an aero-engine casing and impeller to verify the effectiveness and extensibility of the proposed method.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120573"},"PeriodicalIF":5.6,"publicationDate":"2026-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081027","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 rapid large-range hydrophilicity mapping system based on top-down approach 基于自顶向下方法的大范围亲水性快速制图系统

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-26 DOI: 10.1016/j.measurement.2026.120597

Shan-Ya Tsao , Wei-Jen Chen , Yi-Chen Chen , Feng-Sheng Kao , Jen-You Chu , Hsien-Shun Liao

Surface hydrophilicity mapping is primarily quantified using contact angles. The sessile drop method is a widely used contact angle measurement method. It involves observing the profile of a droplet on a surface from a side-view perspective. Side-view imaging makes it difficult to rapidly measure hydrophilicity distribution over large surface areas. This study proposes a novel hydrophilicity mapping method that combines rapid microdroplet spraying via an ultrasonic nozzle with top-down imaging using a line-scan camera. The system achieves rapid large-area hydrophilicity mapping. Experimental results indicate that the system has a scanning range of 300 mm × 250 mm, scanning speed of 30 mm/s, and resolution of 2.52 mm × 2.52 mm.

表面亲水性制图主要是用接触角来量化的。固滴法是一种应用广泛的接触角测量方法。它包括从侧面观察表面上液滴的轮廓。侧视成像使得在大表面积上快速测量亲水性分布变得困难。本研究提出了一种新的亲水性测绘方法，该方法结合了超声喷嘴的快速微滴喷涂和线扫描相机的自上而下成像。该系统实现了快速的大面积亲水性制图。实验结果表明，该系统扫描范围为300 mm × 250 mm，扫描速度为30 mm/s，分辨率为2.52 mm × 2.52 mm。

引用次数: 0

Miniature heterodyne interferometer with reduced periodic nonlinearity and enhanced thermal stability for ultraprecision measurement 微型外差干涉仪与减少周期非线性和增强热稳定性的超精密测量

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-26 DOI: 10.1016/j.measurement.2026.120587

Zhiwei Wang , Pengcheng Hu , Xin Yang , Xionglei Lin , Junhong Luan , Wei Chen , Haijin Fu , Tao Sun , Liang Yu

Heterodyne interferometers are widely used in ultraprecision measurement due to their high resolution and traceability. With the rapid development of extreme ultraviolet lithography, wafer alignment, and other fields, new requirements have emerged regarding the miniaturization and high thermal stability of interferometers. However, miniaturization is often constrained by periodic nonlinearity (PNL) and the thermal drift of optics (TDO). To address these issues, this work establishes a polarization–nonlinearity transfer model based on the Jones matrix formalism and optimizes the key polarization-dependent parameters of the optical components, thereby suppressing the “Type-I PNL” originating from polarization leakage at its source. In parallel, refractive index matching is employed to reduce parasitic interface reflections, and microwedged angles are introduced at the key optical interfaces to induce spatial mismatch of the ghost-reflection beams, effectively mitigating the “Type-II PNL” caused by their self-mixing interference. Furthermore, by implementing paired and symmetric microwedged angles, the optical-path imbalance introduced by single-sided wedge tuning is eliminated, enabling simultaneous realization of low nonlinearity, high stability, and high integration. Based on the proposed methods, a prototype interferometer was developed and experimentally evaluated. The experimental results show that the proposed interferometer achieves a PNL of approximately 59 pm, a TDO coefficient below 6 nm/K, and a temperature-drift-corrected measurement stability with a peak-to-peak variation better than 0.8 nm over 5 h. Furthermore, the direct comparison with a commercial interferometer operating in air shows a standard deviation of the difference of 5.1 nm.

外差干涉仪以其高分辨率和可追溯性在超精密测量中得到广泛应用。随着极紫外光刻、晶圆对准等领域的快速发展，对干涉仪的小型化和高热稳定性提出了新的要求。然而，小型化往往受到周期性非线性（PNL）和光学热漂移（TDO）的限制。为了解决这些问题，本文建立了基于Jones矩阵形式的偏振非线性传递模型，并优化了光学元件的关键偏振相关参数，从而从源头抑制了由偏振泄漏引起的“i型PNL”。同时，利用折射率匹配来减少寄生界面反射，在关键光接口处引入微楔角诱导鬼反射光束的空间失配，有效缓解了鬼反射光束自混合干涉引起的“ii型PNL”。此外，通过实现配对和对称的微楔角，消除了单侧楔调谐引起的光路不平衡，同时实现了低非线性、高稳定性和高集成度。在此基础上，研制了干涉仪样机并进行了实验验证。实验结果表明，该干涉仪的PNL约为59 pm， TDO系数低于6 nm/K，经温度漂移校正后的测量稳定性优于0.8 nm，在5 h内峰间变化优于0.8 nm。此外，与商用干涉仪在空中工作的直接比较，其标准差差为5.1 nm。

{"title":"Miniature heterodyne interferometer with reduced periodic nonlinearity and enhanced thermal stability for ultraprecision measurement","authors":"Zhiwei Wang , Pengcheng Hu , Xin Yang , Xionglei Lin , Junhong Luan , Wei Chen , Haijin Fu , Tao Sun , Liang Yu","doi":"10.1016/j.measurement.2026.120587","DOIUrl":"10.1016/j.measurement.2026.120587","url":null,"abstract":"<div><div>Heterodyne interferometers are widely used in ultraprecision measurement due to their high resolution and traceability. With the rapid development of extreme ultraviolet lithography, wafer alignment, and other fields, new requirements have emerged regarding the miniaturization and high thermal stability of interferometers. However, miniaturization is often constrained by periodic nonlinearity (PNL) and the thermal drift of optics (TDO). To address these issues, this work establishes a polarization–nonlinearity transfer model based on the Jones matrix formalism and optimizes the key polarization-dependent parameters of the optical components, thereby suppressing the “Type-I PNL” originating from polarization leakage at its source. In parallel, refractive index matching is employed to reduce parasitic interface reflections, and microwedged angles are introduced at the key optical interfaces to induce spatial mismatch of the ghost-reflection beams, effectively mitigating the “Type-II PNL” caused by their self-mixing interference. Furthermore, by implementing paired and symmetric microwedged angles, the optical-path imbalance introduced by single-sided wedge tuning is eliminated, enabling simultaneous realization of low nonlinearity, high stability, and high integration. Based on the proposed methods, a prototype interferometer was developed and experimentally evaluated. The experimental results show that the proposed interferometer achieves a PNL of approximately 59 pm, a TDO coefficient below 6 nm/K, and a temperature-drift-corrected measurement stability with a peak-to-peak variation better than 0.8 nm over 5 h. Furthermore, the direct comparison with a commercial interferometer operating in air shows a standard deviation of the difference of 5.1 nm.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120587"},"PeriodicalIF":5.6,"publicationDate":"2026-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081078","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

GlassCurtainCrackAdversarialDefense: A convolutional and attention-based adversarial defense network for glass curtain wall crack detection 玻璃幕墙裂缝对抗防御：一个基于卷积和注意力的对抗防御网络，用于玻璃幕墙裂缝检测

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-25 DOI: 10.1016/j.measurement.2026.120544

Jiaxi Huang, Guixiong Liu, Jing Xu

Deep learning segmentation methods have shown promise in glass curtain wall crack detection but are vulnerable to adversarial attacks. Adversarial attacks, which introduce imperceptible perturbations to the input image, can lead to false or missed detection of glass curtain wall cracks. To address this issue, we propose a novel defense model for glass curtain wall crack adversarial samples, named GlassCurtainCrackAdversarialDefense. The proposed model is a generative adversarial network that incorporates convolutional mechanisms and two types of attention mechanisms, enabling it to extract both local and global perturbation features from adversarial samples and generate high-quality purified images that meet the requirements for glass curtain wall crack segmentation. After applying various adversarial attack methods to the glass curtain wall crack dataset, we evaluated the GlassCurtainCrackAdversarialDefense model on this dataset. Experimental results demonstrate that the proposed network exhibits the best adversarial sample defense capability among multiple defense models. Due to the higher defense performance of the proposed network, false and missed detections of cracks caused by adversarial attacks can be avoided, making the glass curtain wall crack detection process more reliable and trustworthy.

深度学习分割方法在玻璃幕墙裂缝检测中显示出前景，但容易受到对抗性攻击。对抗性攻击，引入难以察觉的扰动输入图像，可能导致错误或漏检玻璃幕墙裂缝。为了解决这一问题，我们提出了一种新的玻璃幕墙裂缝对抗样本防御模型，命名为GlassCurtainCrackAdversarialDefense。该模型是一种结合卷积机制和两种注意机制的生成式对抗网络，能够从对抗样本中提取局部和全局扰动特征，并生成满足玻璃幕墙裂缝分割要求的高质量纯化图像。在对玻璃幕墙裂缝数据集应用各种对抗性攻击方法后，我们在该数据集上评估了GlassCurtainCrackAdversarialDefense模型。实验结果表明，该网络在多种防御模型中表现出最佳的对抗样本防御能力。由于本文提出的网络具有更高的防御性能，可以避免对抗性攻击造成的裂缝误检和漏检，使玻璃幕墙裂缝检测过程更加可靠可信。

{"title":"GlassCurtainCrackAdversarialDefense: A convolutional and attention-based adversarial defense network for glass curtain wall crack detection","authors":"Jiaxi Huang, Guixiong Liu, Jing Xu","doi":"10.1016/j.measurement.2026.120544","DOIUrl":"10.1016/j.measurement.2026.120544","url":null,"abstract":"<div><div>Deep learning segmentation methods have shown promise in glass curtain wall crack detection but are vulnerable to adversarial attacks. Adversarial attacks, which introduce imperceptible perturbations to the input image, can lead to false or missed detection of glass curtain wall cracks. To address this issue, we propose a novel defense model for glass curtain wall crack adversarial samples, named GlassCurtainCrackAdversarialDefense. The proposed model is a generative adversarial network that incorporates convolutional mechanisms and two types of attention mechanisms, enabling it to extract both local and global perturbation features from adversarial samples and generate high-quality purified images that meet the requirements for glass curtain wall crack segmentation. After applying various adversarial attack methods to the glass curtain wall crack dataset, we evaluated the GlassCurtainCrackAdversarialDefense model on this dataset. Experimental results demonstrate that the proposed network exhibits the best adversarial sample defense capability among multiple defense models. Due to the higher defense performance of the proposed network, false and missed detections of cracks caused by adversarial attacks can be avoided, making the glass curtain wall crack detection process more reliable and trustworthy.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120544"},"PeriodicalIF":5.6,"publicationDate":"2026-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081576","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 novel interpretable deep autoencoder for health indicators construction and similarity-based RUL prediction 一种新的可解释深度自编码器，用于健康指标构建和基于相似性的RUL预测

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-25 DOI: 10.1016/j.measurement.2026.120571

Longya Liu , Yang Yu , Xin Guan , Wei Jing , Xiaoxiong Wu

In Prognostics and Health Management (PHM), constructing high-quality Health Indicators (HIs) from Condition Monitoring (CM) data and accurately predicting Remaining Useful Life (RUL) are critical yet challenging tasks. Existing methods often face limitations: HI construction typically requires extensive labeled data or assumes the availability of pristine healthy-state data, while similarity-based RUL is sensitive to scarce reference run-to-failure instances and the risk of selecting non-representative degradation patterns. To address these challenges, this paper proposes an Unsupervised and Interpretable Hybrid Autoencoder (UIHAE) for robust HI construction. Grounded in causal inference and enhanced by physical constraints, the UIHAE integrates a multi-head self-attention mechanism to capture long-range dependencies and a novel gated fusion unit to adaptively disentangle the confounding effects of operational conditions. Building upon the high-quality HIs generated by the UIHAE, we further introduce a dual-perspective similarity-based RUL prediction method. This method mitigates the limitations of scarce reference data by combining a global trend metric (cosine similarity) and a local shape metric (Derivative Dynamic Time Warping, DDTW) within a composite similarity measure. Comprehensive experiments on the N-CMAPSS dataset demonstrate that the proposed framework significantly outperforms existing state-of-the-art methods in both the fidelity of constructed HIs and the accuracy of RUL predictions, showcasing its strong potential for practical industrial applications.

在预后和健康管理（PHM）中，从状态监测（CM）数据构建高质量的健康指标（HIs）并准确预测剩余使用寿命（RUL）是一项关键而具有挑战性的任务。现有方法经常面临限制：HI构建通常需要大量标记数据或假设原始健康状态数据的可用性，而基于相似性的RUL对稀缺的参考运行到故障实例和选择非代表性退化模式的风险很敏感。为了解决这些挑战，本文提出了一种用于鲁棒HI构建的无监督和可解释混合自编码器（UIHAE）。基于因果推理和物理约束，UIHAE集成了一个多头自注意机制来捕获远程依赖关系，以及一个新的门控融合单元来自适应地解开操作条件的混淆效应。在uhae生成的高质量规则模型的基础上，我们进一步引入了一种基于双视角相似性的规则模型预测方法。该方法通过将全局趋势度量（余弦相似性）和局部形状度量（导数动态时间翘曲，DDTW）结合在一个复合相似性度量中，减轻了参考数据稀缺的局限性。在N-CMAPSS数据集上的综合实验表明，所提出的框架在构建HIs的保真度和RUL预测的准确性方面都明显优于现有的最先进的方法，显示了其在实际工业应用中的强大潜力。

{"title":"A novel interpretable deep autoencoder for health indicators construction and similarity-based RUL prediction","authors":"Longya Liu , Yang Yu , Xin Guan , Wei Jing , Xiaoxiong Wu","doi":"10.1016/j.measurement.2026.120571","DOIUrl":"10.1016/j.measurement.2026.120571","url":null,"abstract":"<div><div>In Prognostics and Health Management (PHM), constructing high-quality Health Indicators (HIs) from Condition Monitoring (CM) data and accurately predicting Remaining Useful Life (RUL) are critical yet challenging tasks. Existing methods often face limitations: HI construction typically requires extensive labeled data or assumes the availability of pristine healthy-state data, while similarity-based RUL is sensitive to scarce reference run-to-failure instances and the risk of selecting non-representative degradation patterns. To address these challenges, this paper proposes an Unsupervised and Interpretable Hybrid Autoencoder (UIHAE) for robust HI construction. Grounded in causal inference and enhanced by physical constraints, the UIHAE integrates a multi-head self-attention mechanism to capture long-range dependencies and a novel gated fusion unit to adaptively disentangle the confounding effects of operational conditions. Building upon the high-quality HIs generated by the UIHAE, we further introduce a dual-perspective similarity-based RUL prediction method. This method mitigates the limitations of scarce reference data by combining a global trend metric (cosine similarity) and a local shape metric (Derivative Dynamic Time Warping, DDTW) within a composite similarity measure. Comprehensive experiments on the N-CMAPSS dataset demonstrate that the proposed framework significantly outperforms existing state-of-the-art methods in both the fidelity of constructed HIs and the accuracy of RUL predictions, showcasing its strong potential for practical industrial applications.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"268 ","pages":"Article 120571"},"PeriodicalIF":5.6,"publicationDate":"2026-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146096175","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

Laser scanning infrared thermography for metal microcracks inspection based on thermal signal spatial quasi-static reconstruction 基于热信号空间准静态重构的激光扫描红外热像检测金属微裂纹

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-25 DOI: 10.1016/j.measurement.2026.120448

Zhuoyan Yue , Fei Wang , Peng Yin , Rongcheng Li , Feng Yang , Jianghao Zhao , Mingyu Gao , Stefano Sfarra , V.S. Ghali , G.T. Vesala , R. Mulaveesala , Honghao Yue , Junyan Liu

Active infrared thermography has been demonstrated to offer a wide range of applications in the domain of non-destructive testing, thanks to its multiple advantages. However, when applied to the detection and identification of minute defects on the surfaces of metals and other highly thermally conductive materials, the performance of conventional active infrared thermography is suboptimal. In this paper, a quasi-static reconstruction method based on spatial domain feature reconstruction is proposed to solve this problem. Combined with miniaturized detection equipment, this method enables reliable detection of cracks as narrow as 2 µm on titanium alloy surfaces, offering a novel approach for detecting microcracks on alloy surfaces. Based on the three-dimensional simulation model, the thermal response characteristics caused by crack defect are analyzed to provide theoretical support for the quasi-static reconstruction method proposed for different scanning methods. To verify the reliability of this method, a TC4 specimen containing cracks of various widths is prepared and a mobile laser scanning infrared thermography system is established, which demonstrates the potential for miniaturization and integration, broadening its applicability in industrial in-situ inspection endoscopic inspection.. Comparative experiments demonstrate that an appropriate combination of scanning method and reconstruction method can effectively suppress background signal interference caused by surface reflections, allowing for the reliable detection and identification of cracks with widths down to 2 µm. Analysis indicates that the signal-to-noise ratio(SNR) is positively correlated with crack width, and feature extraction algorithms demonstrate a significant contribution to the improvement of SNR for cracks with a width of 2 µm.

由于其多种优势，主动红外热成像技术已被证明在无损检测领域具有广泛的应用。然而，当应用于检测和识别金属和其他高导热材料表面的微小缺陷时，传统的主动红外热像仪的性能并不理想。本文提出了一种基于空间域特征重构的准静态重构方法来解决这一问题。结合小型化的检测设备，该方法可以可靠地检测到钛合金表面窄至2 μ m的裂纹，为检测合金表面微裂纹提供了一种新的方法。基于三维仿真模型，分析了裂纹缺陷引起的热响应特征，为针对不同扫描方式提出的准静态重构方法提供理论支持。为验证该方法的可靠性，制备了含不同宽度裂纹的TC4试样，建立了移动激光扫描红外热成像系统，展示了其小型化和集成化的潜力，拓宽了其在工业原位检测内窥镜检测中的适用性。对比实验表明，扫描方法与重建方法的适当结合，可以有效抑制表面反射引起的背景信号干扰，从而可靠地检测和识别宽度低至2µm的裂纹。分析表明，信噪比（SNR）与裂缝宽度呈正相关，特征提取算法对裂缝宽度为2µm时信噪比的提高有显著贡献。

{"title":"Laser scanning infrared thermography for metal microcracks inspection based on thermal signal spatial quasi-static reconstruction","authors":"Zhuoyan Yue , Fei Wang , Peng Yin , Rongcheng Li , Feng Yang , Jianghao Zhao , Mingyu Gao , Stefano Sfarra , V.S. Ghali , G.T. Vesala , R. Mulaveesala , Honghao Yue , Junyan Liu","doi":"10.1016/j.measurement.2026.120448","DOIUrl":"10.1016/j.measurement.2026.120448","url":null,"abstract":"<div><div>Active infrared thermography has been demonstrated to offer a wide range of applications in the domain of non-destructive testing, thanks to its multiple advantages. However, when applied to the detection and identification of minute defects on the surfaces of metals and other highly thermally conductive materials, the performance of conventional active infrared thermography is suboptimal. In this paper, a quasi-static reconstruction method based on spatial domain feature reconstruction is proposed to solve this problem. Combined with miniaturized detection equipment, this method enables reliable detection of cracks as narrow as 2 µm on titanium alloy surfaces, offering a novel approach for detecting microcracks on alloy surfaces. Based on the three-dimensional simulation model, the thermal response characteristics caused by crack defect are analyzed to provide theoretical support for the quasi-static reconstruction method proposed for different scanning methods. To verify the reliability of this method, a TC4 specimen containing cracks of various widths is prepared and a mobile laser scanning infrared thermography system is established, which demonstrates the potential for miniaturization and integration, broadening its applicability in industrial in-situ inspection endoscopic inspection.. Comparative experiments demonstrate that an appropriate combination of scanning method and reconstruction method can effectively suppress background signal interference caused by surface reflections, allowing for the reliable detection and identification of cracks with widths down to 2 µm. Analysis indicates that the signal-to-noise ratio(SNR) is positively correlated with crack width, and feature extraction algorithms demonstrate a significant contribution to the improvement of SNR for cracks with a width of 2 µm.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120448"},"PeriodicalIF":5.6,"publicationDate":"2026-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146080770","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

The path to high-performance: the manufacturing, measurement, and trimming for hemispherical resonators 通往高性能之路：半球形谐振器的制造、测量和修整

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-25 DOI: 10.1016/j.measurement.2026.120582

Wenhui Chen , Tao Lai , Yufang Zhou , Qinghang Liu , Wenxiang Peng , Fulei Chen , Shanyong Chen , Junfeng Liu

The hemispherical resonant gyroscope (HRG) is renowned as the most precise solid-state gyroscope, characterized by a long lifespan and robust overload resistance. Its exceptional temperature adaptability and unique shutdown radiation resistance render it indispensable for inertial navigation systems operating in extreme environments. At the heart of the HRG is the hemispherical resonator (HR), a component notable for its high steepness and irregular curved surface geometry. The complex machining of the HR, coupled with potential uneven mass distribution due to manufacturing errors, can significantly impact the performance and output accuracy of the HRG. Consequently, the development of high-performance manufacturing methods for the HR has been a prominent focus within the field. This review examines state-of-the-art ultra-precision manufacturing techniques, including both processing and trimming procedures, alongside advanced measurement methods pertaining to geometric parameters and vibration performance metrics. Key challenges identified include achieving the necessary precision during manufacturing processes and establishing standardized measurement and evaluation protocols. By synthesizing current research, this review elucidates the correlations between manufacturing parameters and device performance while highlighting existing technological gaps. The findings offer valuable insights for optimizing resonator production and advancing HRG technology, thereby serving as a reference for future research aimed at enhancing the precision of next-generation inertial navigation systems.

半球形谐振陀螺仪（HRG）是公认的最精确的固态陀螺仪，具有寿命长，抗过载能力强的特点。其卓越的温度适应性和独特的停机抗辐射性能使其成为在极端环境下运行的惯性导航系统不可或缺的部件。HRG的核心是半球形谐振器（HR），这是一个以其高陡度和不规则曲面几何形状而闻名的部件。由于加工过程复杂，再加上制造误差可能导致的质量分布不均匀，会严重影响液位计的性能和输出精度。因此，高性能人力资源制造方法的开发一直是该领域的一个突出焦点。这篇综述研究了最先进的超精密制造技术，包括加工和修剪程序，以及有关几何参数和振动性能指标的先进测量方法。确定的主要挑战包括在制造过程中实现必要的精度，以及建立标准化的测量和评估协议。通过综合目前的研究，本文阐述了制造参数与设备性能之间的相关性，同时强调了现有的技术差距。研究结果为优化谐振器生产和推进HRG技术提供了有价值的见解，从而为未来旨在提高下一代惯性导航系统精度的研究提供了参考。

{"title":"The path to high-performance: the manufacturing, measurement, and trimming for hemispherical resonators","authors":"Wenhui Chen , Tao Lai , Yufang Zhou , Qinghang Liu , Wenxiang Peng , Fulei Chen , Shanyong Chen , Junfeng Liu","doi":"10.1016/j.measurement.2026.120582","DOIUrl":"10.1016/j.measurement.2026.120582","url":null,"abstract":"<div><div>The hemispherical resonant gyroscope (HRG) is renowned as the most precise solid-state gyroscope, characterized by a long lifespan and robust overload resistance. Its exceptional temperature adaptability and unique shutdown radiation resistance render it indispensable for inertial navigation systems operating in extreme environments. At the heart of the HRG is the hemispherical resonator (HR), a component notable for its high steepness and irregular curved surface geometry. The complex machining of the HR, coupled with potential uneven mass distribution due to manufacturing errors, can significantly impact the performance and output accuracy of the HRG. Consequently, the development of high-performance manufacturing methods for the HR has been a prominent focus within the field. This review examines state-of-the-art ultra-precision manufacturing techniques, including both processing and trimming procedures, alongside advanced measurement methods pertaining to geometric parameters and vibration performance metrics. Key challenges identified include achieving the necessary precision during manufacturing processes and establishing standardized measurement and evaluation protocols. By synthesizing current research, this review elucidates the correlations between manufacturing parameters and device performance while highlighting existing technological gaps. The findings offer valuable insights for optimizing resonator production and advancing HRG technology, thereby serving as a reference for future research aimed at enhancing the precision of next-generation inertial navigation systems.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120582"},"PeriodicalIF":5.6,"publicationDate":"2026-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081400","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

Stratified flow detection in horizontal wells based on distributed fiber-optic temperature sensing 基于分布式光纤温度传感的水平井分层流检测

IF 5.6 2区工程技术 Q1 ENGINEERING, MULTIDISCIPLINARY

Measurement

Pub Date : 2026-01-25 DOI: 10.1016/j.measurement.2026.120563

Taiji Dong , Zhenzong Ma , Xu Liu , Chunlei Jiang , Wei Li , Yangyang Li , Yuxin Xiao

Stratified flow detection is one of the key technologies for achieving refined injection and regulation in oil and gas fields. This paper proposes a novel stratified flow detection method based on distributed fiber Raman scattering temperature sensing. A multi-physics coupling model of the fiber temperature field and fluid flow state is established to simulate the complex temperature distribution and flow characteristics in horizontal wells, with experimental verification conducted under a controlled downhole temperature gradient environment. The experimental results indicate that, under single water injection port conditions, the temperature difference change rates are 0.343 ℃/min, 0.325 ℃/min, and 0.335 ℃/min, respectively. The flow rate is highest at water injection port 1. Under combined operation of multiple water injection ports, The combination of ports 1 and 3 yields the maximum total flow. Moreover, further investigation indicates that under multi-inlet operating conditions with complex simultaneous injection, the temperature-difference response can still provide a stable linear characterization of the flow rate at each port. This method enables real-time acquisition and inversion of wellbore temperature signals, demonstrating high detection accuracy and engineering practicality. This study provides reliable technical pathways and theoretical support for flow monitoring during stratified injection and production in oilfields, offering broad application prospects.

分层流量检测是油气田实现精细化注调的关键技术之一。提出了一种基于分布式光纤拉曼散射温度传感的分层流量检测方法。建立了纤维温度场与流体流动状态的多物理场耦合模型，模拟了水平井中复杂的温度分布和流动特性，并在受控的井下温度梯度环境下进行了实验验证。实验结果表明，在单一注水口条件下，温差变化率分别为0.343℃/min、0.325℃/min和0.335℃/min。注水口1的流量最大。多个注水口联合作业时，1、3口组合总流量最大。此外，进一步研究表明，在复杂同时喷射的多入口工况下，温差响应仍然可以提供每个端口流量的稳定线性表征。该方法能够实时采集和反演井筒温度信号，具有较高的探测精度和工程实用性。该研究为油田分层注采过程中的流动监测提供了可靠的技术途径和理论支持，具有广阔的应用前景。

{"title":"Stratified flow detection in horizontal wells based on distributed fiber-optic temperature sensing","authors":"Taiji Dong , Zhenzong Ma , Xu Liu , Chunlei Jiang , Wei Li , Yangyang Li , Yuxin Xiao","doi":"10.1016/j.measurement.2026.120563","DOIUrl":"10.1016/j.measurement.2026.120563","url":null,"abstract":"<div><div>Stratified flow detection is one of the key technologies for achieving refined injection and regulation in oil and gas fields. This paper proposes a novel stratified flow detection method based on distributed fiber Raman scattering temperature sensing. A multi-physics coupling model of the fiber temperature field and fluid flow state is established to simulate the complex temperature distribution and flow characteristics in horizontal wells, with experimental verification conducted under a controlled downhole temperature gradient environment. The experimental results indicate that, under single water injection port conditions, the temperature difference change rates are 0.343 ℃/min, 0.325 ℃/min, and 0.335 ℃/min, respectively. The flow rate is highest at water injection port 1. Under combined operation of multiple water injection ports, The combination of ports 1 and 3 yields the maximum total flow. Moreover, further investigation indicates that under multi-inlet operating conditions with complex simultaneous injection, the temperature-difference response can still provide a stable linear characterization of the flow rate at each port. This method enables real-time acquisition and inversion of wellbore temperature signals, demonstrating high detection accuracy and engineering practicality. This study provides reliable technical pathways and theoretical support for flow monitoring during stratified injection and production in oilfields, offering broad application prospects.</div></div>","PeriodicalId":18349,"journal":{"name":"Measurement","volume":"267 ","pages":"Article 120563"},"PeriodicalIF":5.6,"publicationDate":"2026-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081562","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