Flow Measurement and Instrumentation最新文献_第2页

Enhancing rainfall simulator Efficiency: A search for new insights into performance parameters 提高降雨模拟器的效率：对性能参数的新见解的探索

IF 2.7 3区工程技术 Q2 ENGINEERING, MECHANICAL

Flow Measurement and Instrumentation

Pub Date : 2026-01-17 DOI: 10.1016/j.flowmeasinst.2026.103203

Abdullah Emin Demircioglu , Kadir Gezici , Erdal Kesgin

Rainfall simulator (RS) is an indispensable tool for researchers conducting studies in hydrology, geomorphology, and erosion; however, performance evaluation varies due to differences in the methods used by each researcher. This study applies the concept of spatial variability within rainfall parameter limits and aims to contribute to a better calibration method for RS. Rainfall parameters such as intensity, homogeneity, and raindrop size were examined using different measurement strategies. A pressurized RS was tested in a controlled laboratory flume using nominal intensities of 40-, 70-, and 100-mmh-¹ over a 2.4 m × 1.2 m (3.2 m²) test channel. The channel was divided into nine sub-areas, and the number and configuration of collection cups was varied above and below the channel using between 108 and 315 cups, with areal sampling densities of 16–48 %. Point rainfall intensities ranged from 15 to 185 mmh-¹ within single events, depending on spatial position and cup configuration, which demonstrated that using a single representative parameter value will not allow for capturing the strong local variability that is present in RS. Christiansen Uniformity (CU) coefficients had clear regional disparities, with sub-area CU coefficients ranging between 66 and 96 %; under the best configuration, local CU coefficients reach 96 %, but lower density configurations had sub-area CU coefficient values as low as 66 %. With an increase in cup density, spread of measured intensities was reduced by more than one-third and produced smoother spatial patterns of uniformity while showing only minor corresponding shifts in CU values (typically on the order of ±1 % for a given nominal intensity). No systematic trend with cup number was displayed. The raindrop size characteristics only demonstrated a small but consistent increase in median raindrop diameter (D50) in response to increasing intensity, moving from 1.39 mm at 40 mmh-¹ to approximately 1.60 mm at 100 mmh-¹ with greater spatial variability being displayed at the higher intensity level. The results of this study emphasize the need for the incorporation of spatially sensitive calibration protocols for RS systems. The method outlined provides a more robust and comparable framework for evaluating rainfall simulator performance and thereby increases the reliability of simulation-based studies in environmental research.

降雨模拟器是水文、地貌学和侵蚀学研究人员不可缺少的工具；然而，由于每个研究人员使用的方法不同，绩效评估也有所不同。本研究在降雨参数范围内引入了空间变异性的概念，旨在为RS提供更好的校准方法。采用不同的测量策略对降雨参数（如强度、均匀性和雨滴大小）进行了检验。加压RS在受控实验室水槽中进行测试，在2.4 m × 1.2 m （3.2 m2）的测试通道上使用40、70和100 mmh-1的标称强度。该通道被划分为9个子区域，收集杯的数量和配置在通道上下变化，使用108 - 315杯，面采样密度为16 - 48%。单次事件的点降雨强度范围为15 ~ 185 mmh-1，取决于空间位置和杯形，这表明使用单一代表性参数值无法捕捉RS中存在的强烈的局部变率。克里斯蒂安森均匀度（CU）系数具有明显的区域差异，子区域CU系数在66% ~ 96%之间；最佳配置下，局部CU系数可达96%，较低密度配置下，分区CU系数仅为66%。随着杯形密度的增加，测量强度的分布减少了三分之一以上，并产生了更平滑的均匀性空间模式，而CU值仅发生了较小的相应变化（对于给定的标称强度，通常在±1%左右）。杯数变化无系统趋势。雨滴大小特征显示，雨滴中值直径（D50）随强度的增加而增加，从40 mmh-1时的1.39 mm增加到100 mmh-1时的约1.60 mm，并且在高强度水平下表现出更大的空间变异性。本研究的结果强调需要纳入空间敏感的RS系统校准协议。概述的方法为评估降雨模拟器的性能提供了一个更稳健和可比较的框架，从而提高了环境研究中基于模拟的研究的可靠性。

{"title":"Enhancing rainfall simulator Efficiency: A search for new insights into performance parameters","authors":"Abdullah Emin Demircioglu , Kadir Gezici , Erdal Kesgin","doi":"10.1016/j.flowmeasinst.2026.103203","DOIUrl":"10.1016/j.flowmeasinst.2026.103203","url":null,"abstract":"<div><div>Rainfall simulator (RS) is an indispensable tool for researchers conducting studies in hydrology, geomorphology, and erosion; however, performance evaluation varies due to differences in the methods used by each researcher. This study applies the concept of spatial variability within rainfall parameter limits and aims to contribute to a better calibration method for RS. Rainfall parameters such as intensity, homogeneity, and raindrop size were examined using different measurement strategies. A pressurized RS was tested in a controlled laboratory flume using nominal intensities of 40-, 70-, and 100-mmh-<sup>1</sup> over a 2.4 m × 1.2 m (3.2 m<sup>2</sup>) test channel. The channel was divided into nine sub-areas, and the number and configuration of collection cups was varied above and below the channel using between 108 and 315 cups, with areal sampling densities of 16–48 %. Point rainfall intensities ranged from 15 to 185 mmh-<sup>1</sup> within single events, depending on spatial position and cup configuration, which demonstrated that using a single representative parameter value will not allow for capturing the strong local variability that is present in RS. Christiansen Uniformity (CU) coefficients had clear regional disparities, with sub-area CU coefficients ranging between 66 and 96 %; under the best configuration, local CU coefficients reach 96 %, but lower density configurations had sub-area CU coefficient values as low as 66 %. With an increase in cup density, spread of measured intensities was reduced by more than one-third and produced smoother spatial patterns of uniformity while showing only minor corresponding shifts in CU values (typically on the order of ±1 % for a given nominal intensity). No systematic trend with cup number was displayed. The raindrop size characteristics only demonstrated a small but consistent increase in median raindrop diameter (D50) in response to increasing intensity, moving from 1.39 mm at 40 mmh-<sup>1</sup> to approximately 1.60 mm at 100 mmh-<sup>1</sup> with greater spatial variability being displayed at the higher intensity level. The results of this study emphasize the need for the incorporation of spatially sensitive calibration protocols for RS systems. The method outlined provides a more robust and comparable framework for evaluating rainfall simulator performance and thereby increases the reliability of simulation-based studies in environmental research.</div></div>","PeriodicalId":50440,"journal":{"name":"Flow Measurement and Instrumentation","volume":"108 ","pages":"Article 103203"},"PeriodicalIF":2.7,"publicationDate":"2026-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146022799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Analysis of 70 MPa two-stage series high-pressure reducing valve for hydrogen fuel cell vehicles 氢燃料电池汽车用70 MPa两级串联高压减压阀分析

IF 2.7 3区工程技术 Q2 ENGINEERING, MECHANICAL

Flow Measurement and Instrumentation

Pub Date : 2026-01-16 DOI: 10.1016/j.flowmeasinst.2026.103206

Ruiyao Tao , Yaobao Yin , Huiyun Xu , Li Hu , Naiping Gao , Pengfei He

The two-stage series high-pressure reducing valve (TSHPRV) is regarded as a key component for ensuring the stable operation of 70 MPa hydrogen supply systems in hydrogen fuel cell vehicles. Under extreme pressure-drop gradients and complex multistage coupled flows, conventional analyses based on the ideal-gas assumption become inadequate. A nonlinear dynamic model of the TSHPRV is established by incorporating a real-gas equation of state together with mass and energy conservation, allowing the dominant pressure drop across the first-stage orifice and the precise regulation by the second stage to be accurately represented. It is shown that the outlet pressure can be maintained within ±0.02 MPa over wide ranges of inlet pressure (70-5 MPa) and load variation (0–2.6 g/s). The first-stage orifice flow is observed to transition from subsonic to supersonic with changing operating conditions. The dynamic regulation process is divided into three stages, during which the co-regulation stage reveals dynamic coupling between the two spools and the inherent adaptive mechanism. Parametric analysis indicates that the intermediate chamber volume and spring stiffness considerably affect the dynamic characteristics, while inlet-pressure variations induce adaptive adjustments of the orifice openings and consequently influence internal flow and energy dissipation. Experimental results confirm that the deviation between predicted and measured values remains within 3.2%, demonstrating the accuracy of the model and providing theoretical guidance for analyzing the regulation mechanism of 70 MPa TSHPRVs.

两级串联高压减压阀（TSHPRV）是保证氢燃料电池汽车70 MPa供氢系统稳定运行的关键部件。在极端压降梯度和复杂多级耦合流动的情况下，传统的基于理想气体假设的分析是不充分的。将实气状态方程与质量守恒和能量守恒相结合，建立了TSHPRV的非线性动力学模型，可以准确地表示第一级孔口的主压降和第二级的精确调节。结果表明，在进口压力（70 ~ 5mpa）和负载变化（0 ~ 2.6 g/s）范围内，出口压力均可保持在±0.02 MPa。观察到随着工作条件的变化，第一级孔板流动由亚音速过渡到超音速。动态调节过程分为三个阶段，其中共调节阶段揭示了两轴之间的动态耦合和内在的自适应机制。参数分析表明，中间腔室体积和弹簧刚度对动态特性影响较大，而进口压力的变化会引起孔口开口的自适应调整，从而影响内部流动和能量耗散。实验结果证实，预测值与实测值的偏差在3.2%以内，验证了模型的准确性，为分析70 MPa TSHPRVs的调节机理提供了理论指导。

{"title":"Analysis of 70 MPa two-stage series high-pressure reducing valve for hydrogen fuel cell vehicles","authors":"Ruiyao Tao , Yaobao Yin , Huiyun Xu , Li Hu , Naiping Gao , Pengfei He","doi":"10.1016/j.flowmeasinst.2026.103206","DOIUrl":"10.1016/j.flowmeasinst.2026.103206","url":null,"abstract":"<div><div>The two-stage series high-pressure reducing valve (TSHPRV) is regarded as a key component for ensuring the stable operation of 70 MPa hydrogen supply systems in hydrogen fuel cell vehicles. Under extreme pressure-drop gradients and complex multistage coupled flows, conventional analyses based on the ideal-gas assumption become inadequate. A nonlinear dynamic model of the TSHPRV is established by incorporating a real-gas equation of state together with mass and energy conservation, allowing the dominant pressure drop across the first-stage orifice and the precise regulation by the second stage to be accurately represented. It is shown that the outlet pressure can be maintained within ±0.02 MPa over wide ranges of inlet pressure (70-5 MPa) and load variation (0–2.6 g/s). The first-stage orifice flow is observed to transition from subsonic to supersonic with changing operating conditions. The dynamic regulation process is divided into three stages, during which the co-regulation stage reveals dynamic coupling between the two spools and the inherent adaptive mechanism. Parametric analysis indicates that the intermediate chamber volume and spring stiffness considerably affect the dynamic characteristics, while inlet-pressure variations induce adaptive adjustments of the orifice openings and consequently influence internal flow and energy dissipation. Experimental results confirm that the deviation between predicted and measured values remains within 3.2%, demonstrating the accuracy of the model and providing theoretical guidance for analyzing the regulation mechanism of 70 MPa TSHPRVs.</div></div>","PeriodicalId":50440,"journal":{"name":"Flow Measurement and Instrumentation","volume":"108 ","pages":"Article 103206"},"PeriodicalIF":2.7,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145976446","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Fibre-optic sensor design and strain-response analysis to wall pressure fluctuations in turbulent partially-filled pipe flows 光纤传感器设计及紊流部分填充管道壁面压力波动的应变响应分析

IF 2.7 3区工程技术 Q2 ENGINEERING, MECHANICAL

Flow Measurement and Instrumentation

Pub Date : 2026-01-16 DOI: 10.1016/j.flowmeasinst.2026.103208

Jah Shamas, Anton Krynkin, Gavin Sailor, Melika Gul, Simon Tait, Kirill Horoshenkov

Many studies exist analysing dynamics of pressure-driven flows, yet the number of studies quantifying hydrodynamic relations in partially-filled gravity-driven pipe flows remains limited. This work presents strain data for two uniform, steady flow regimes from a novel streamwise, flush-mounted fibre-optic sensor embedded in the interior wall of a partially-filled pipe. The sensor responds to turbulence-induced pressure fluctuations at the wall, transducing them into measurable strain via a custom plate-gel structure. This sensor represents the first of its kind for non-invasive response to wall pressure fluctuations in partially-filled pipes. Cross-correlation analysis recovered bulk flow velocity estimates within 5% of the true velocity. A Metropolis–Hastings-based automated filtering process identified the frequency range retaining bulk flow properties while removing noisy measurements, further refining velocity estimates. The identified streamwise correlation pattern agreed with experimentally observed wall pressure correlations. Results suggest streamwise correlations decay at a rate proportional to the hydraulic radius in each regime, with only a 7% discrepancy between regimes. Furthermore, spectral analysis indicates the sensor’s sensitivity to coherent large-scale turbulent structures from wall data alone, a capability not previously demonstrated experimentally, suggesting potential for monitoring flow structures beyond bulk conditions. Out-of-water tests produced uninformative results, confirming that in-flow placement captures genuine flow dynamics. These findings demonstrate a novel sensing approach to monitoring partially-filled pipe flows, providing both bulk property estimation and insights into multi-scale flow structures from data collection at the wall, making way for a new form of sensor to be deployed for monitoring these flows in varying industrial contexts.

已有许多研究分析了压力驱动流动的动力学特性，但对部分填充的重力驱动管道流动的水动力关系进行量化研究的数量仍然有限。这项工作提供了两种均匀、稳定流动状态的应变数据，这些数据来自一种新型的顺流式光纤传感器，该传感器嵌入在部分填充的管道内壁中。传感器响应湍流引起的壁面压力波动，通过定制的板-凝胶结构将其转换为可测量的应变。该传感器是首个对部分填充管道壁面压力波动进行非侵入性响应的传感器。互相关分析恢复的整体流速估计在真实流速的5%以内。基于metropolis - hastings的自动滤波过程确定了频率范围，保留了整体流动特性，同时去除噪声测量，进一步改进了速度估计。确定的流向相关模式与实验观察到的壁面压力相关一致。结果表明，在每个制度下，流向相关性以与水力半径成正比的速率衰减，制度之间只有7%的差异。此外，光谱分析表明，传感器仅从壁面数据就能对连贯的大尺度湍流结构敏感，这是以前没有在实验中证明的能力，这表明了监测超出体积条件的流动结构的潜力。出水测试的结果没有提供任何信息，但证实了流内放置捕捉到了真实的流动动态。这些发现展示了一种新的传感方法来监测部分填充的管道流动，提供了体积特性估计和从壁面数据收集的多尺度流动结构的见解，为在不同工业环境中监测这些流动部署一种新型传感器铺平了道路。

{"title":"Fibre-optic sensor design and strain-response analysis to wall pressure fluctuations in turbulent partially-filled pipe flows","authors":"Jah Shamas, Anton Krynkin, Gavin Sailor, Melika Gul, Simon Tait, Kirill Horoshenkov","doi":"10.1016/j.flowmeasinst.2026.103208","DOIUrl":"10.1016/j.flowmeasinst.2026.103208","url":null,"abstract":"<div><div>Many studies exist analysing dynamics of pressure-driven flows, yet the number of studies quantifying hydrodynamic relations in partially-filled gravity-driven pipe flows remains limited. This work presents strain data for two uniform, steady flow regimes from a novel streamwise, flush-mounted fibre-optic sensor embedded in the interior wall of a partially-filled pipe. The sensor responds to turbulence-induced pressure fluctuations at the wall, transducing them into measurable strain via a custom plate-gel structure. This sensor represents the first of its kind for non-invasive response to wall pressure fluctuations in partially-filled pipes. Cross-correlation analysis recovered bulk flow velocity estimates within 5% of the true velocity. A Metropolis–Hastings-based automated filtering process identified the frequency range retaining bulk flow properties while removing noisy measurements, further refining velocity estimates. The identified streamwise correlation pattern agreed with experimentally observed wall pressure correlations. Results suggest streamwise correlations decay at a rate proportional to the hydraulic radius in each regime, with only a 7% discrepancy between regimes. Furthermore, spectral analysis indicates the sensor’s sensitivity to coherent large-scale turbulent structures from wall data alone, a capability not previously demonstrated experimentally, suggesting potential for monitoring flow structures beyond bulk conditions. Out-of-water tests produced uninformative results, confirming that in-flow placement captures genuine flow dynamics. These findings demonstrate a novel sensing approach to monitoring partially-filled pipe flows, providing both bulk property estimation and insights into multi-scale flow structures from data collection at the wall, making way for a new form of sensor to be deployed for monitoring these flows in varying industrial contexts.</div></div>","PeriodicalId":50440,"journal":{"name":"Flow Measurement and Instrumentation","volume":"108 ","pages":"Article 103208"},"PeriodicalIF":2.7,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146022796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 self-priming centrifugal pump and the influence of cooling channels on its hydrodynamic performance 一种新型自吸式离心泵及其冷却通道对其流体动力性能的影响

IF 2.7 3区工程技术 Q2 ENGINEERING, MECHANICAL

Flow Measurement and Instrumentation

Pub Date : 2026-01-13 DOI: 10.1016/j.flowmeasinst.2026.103205

Xingxin Liang , Bingqian Wang , Hanqin Shen , Yiming Qiao , Ruyi Wang , Feng Wang , Beibei Li

To address the challenges of traditional marine self-priming centrifugal pumps, such as their large size, heavy weight, high noise levels, and excessive space occupation in ship cabins, a novel design for a self-priming centrifugal pump, namely an Integrated Motor Self-priming Centrifugal Pump (IMSCP) is proposed. This study employs CFD simulations and experimental validation to investigate the effects of cooling channels, specifically the shaft hole diameter and the air gap thickness, on a 37 kW IMSCP's head, efficiency, and axial force. The results show that as the shaft hole diameter increases, the head, efficiency, and axial force all decrease; however, the reduction in axial force is the most prominent. Specifically, as the diameter increases from 2 mm to 8 mm, the axial force drops most notably from 6739 N to 2463 N, representing a decrease of 63.4 %. However, beyond 8 mm, the head and efficiency decline rapidly. As the air gap thickness increases, the axial force also gradually decreases, but the variations in head, efficiency, and shaft power are relatively minor. Comparatively, the influence of the shaft hole diameter on the overall hydrodynamic performance is more significant than that of the air gap. A prototype of the 37 kW IMSCP pump unit was developed, and the accuracy of the simulation results was verified through experiments, with an error margin within 10 %. The findings provide critical insights for the optimal design of IMSCPs, supporting their application and performance enhancement in industrial fluid transportation.

针对传统船用自吸离心泵体积大、重量重、噪声高、占用舱室空间大等问题，提出了一种新型自吸离心泵——集成电机自吸离心泵（IMSCP）。本研究采用CFD模拟和实验验证，研究了冷却通道（特别是轴孔直径和气隙厚度）对37 kW IMSCP扬程、效率和轴向力的影响。结果表明：随着轴孔直径的增大，压头、效率和轴向力均减小；然而，轴向力的减小是最显著的。当直径从2 mm增加到8 mm时，轴向力从6739 N减小到2463 N，减小幅度最大，为63.4%。但超过8 mm，扬程和效率迅速下降。随着气隙厚度的增加，轴向力也逐渐减小，但扬程、效率和轴功率的变化相对较小。相比之下，井眼直径对整体水动力性能的影响比气隙的影响更显著。研制了37 kW IMSCP泵机组样机，并通过实验验证了仿真结果的准确性，误差范围在10%以内。研究结果为IMSCPs的优化设计提供了重要见解，支持其在工业流体输送中的应用和性能提高。

{"title":"A novel self-priming centrifugal pump and the influence of cooling channels on its hydrodynamic performance","authors":"Xingxin Liang , Bingqian Wang , Hanqin Shen , Yiming Qiao , Ruyi Wang , Feng Wang , Beibei Li","doi":"10.1016/j.flowmeasinst.2026.103205","DOIUrl":"10.1016/j.flowmeasinst.2026.103205","url":null,"abstract":"<div><div>To address the challenges of traditional marine self-priming centrifugal pumps, such as their large size, heavy weight, high noise levels, and excessive space occupation in ship cabins, a novel design for a self-priming centrifugal pump, namely an Integrated Motor Self-priming Centrifugal Pump (IMSCP) is proposed. This study employs CFD simulations and experimental validation to investigate the effects of cooling channels, specifically the shaft hole diameter and the air gap thickness, on a 37 kW IMSCP's head, efficiency, and axial force. The results show that as the shaft hole diameter increases, the head, efficiency, and axial force all decrease; however, the reduction in axial force is the most prominent. Specifically, as the diameter increases from 2 mm to 8 mm, the axial force drops most notably from 6739 N to 2463 N, representing a decrease of 63.4 %. However, beyond 8 mm, the head and efficiency decline rapidly. As the air gap thickness increases, the axial force also gradually decreases, but the variations in head, efficiency, and shaft power are relatively minor. Comparatively, the influence of the shaft hole diameter on the overall hydrodynamic performance is more significant than that of the air gap. A prototype of the 37 kW IMSCP pump unit was developed, and the accuracy of the simulation results was verified through experiments, with an error margin within 10 %. The findings provide critical insights for the optimal design of IMSCPs, supporting their application and performance enhancement in industrial fluid transportation.</div></div>","PeriodicalId":50440,"journal":{"name":"Flow Measurement and Instrumentation","volume":"108 ","pages":"Article 103205"},"PeriodicalIF":2.7,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146022794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 review of deep learning-based bubble recognition methods 基于深度学习的气泡识别方法综述

IF 2.7 3区工程技术 Q2 ENGINEERING, MECHANICAL

Flow Measurement and Instrumentation

Pub Date : 2026-01-12 DOI: 10.1016/j.flowmeasinst.2026.103204

Yun Ren , Yuchao Wang , Bo Yang , Shuihua Zheng , Min Chai

Bubble recognition is essential for the quantitative analysis of gas–liquid two-phase flows and underpins a wide range of applications in fluid dynamics and chemical engineering. Traditional image-processing-based methods perform well under controlled conditions but exhibit limited robustness when confronted with complex bubble morphologies, dense overlap, and non-uniform illumination. Recent advances in deep learning have enabled automated feature learning and significantly improved recognition performance under challenging imaging environments. This review summarizes recent progress in deep learning-based bubble recognition, focusing on bubble detection, semantic and instance segmentation, and boundary-preserving shape reconstruction. Existing methods are organized according to their dominant application scenarios, including real-time multi-scale detection, instance separation in dense bubbly flows, and geometry-aware reconstruction for quantitative bubble characterization. In addition, strategies for alleviating data scarcity, such as synthetic data generation and physically driven are discussed with respect to their role in improving model generalization. This review aims to clarify current methodological trends, highlight remaining challenges, and provide guidance for future development and practical deployment of deep learning-based bubble recognition techniques.

气泡识别对于气液两相流的定量分析至关重要，在流体动力学和化学工程中有着广泛的应用。传统的基于图像处理的方法在受控条件下表现良好，但在面对复杂气泡形态、密集重叠和非均匀光照时，鲁棒性有限。深度学习的最新进展使自动化特征学习成为可能，并显著提高了具有挑战性的成像环境下的识别性能。本文综述了基于深度学习的气泡识别的最新进展，重点是气泡检测、语义和实例分割以及保持边界的形状重建。现有方法根据其主要应用场景进行组织，包括实时多尺度检测、密集气泡流中的实例分离以及用于定量气泡表征的几何感知重构。此外，还讨论了缓解数据稀缺性的策略，如合成数据生成和物理驱动，以及它们在提高模型泛化方面的作用。本综述旨在澄清当前的方法趋势，突出仍然存在的挑战，并为基于深度学习的气泡识别技术的未来发展和实际部署提供指导。

{"title":"A review of deep learning-based bubble recognition methods","authors":"Yun Ren , Yuchao Wang , Bo Yang , Shuihua Zheng , Min Chai","doi":"10.1016/j.flowmeasinst.2026.103204","DOIUrl":"10.1016/j.flowmeasinst.2026.103204","url":null,"abstract":"<div><div>Bubble recognition is essential for the quantitative analysis of gas–liquid two-phase flows and underpins a wide range of applications in fluid dynamics and chemical engineering. Traditional image-processing-based methods perform well under controlled conditions but exhibit limited robustness when confronted with complex bubble morphologies, dense overlap, and non-uniform illumination. Recent advances in deep learning have enabled automated feature learning and significantly improved recognition performance under challenging imaging environments. This review summarizes recent progress in deep learning-based bubble recognition, focusing on bubble detection, semantic and instance segmentation, and boundary-preserving shape reconstruction. Existing methods are organized according to their dominant application scenarios, including real-time multi-scale detection, instance separation in dense bubbly flows, and geometry-aware reconstruction for quantitative bubble characterization. In addition, strategies for alleviating data scarcity, such as synthetic data generation and physically driven are discussed with respect to their role in improving model generalization. This review aims to clarify current methodological trends, highlight remaining challenges, and provide guidance for future development and practical deployment of deep learning-based bubble recognition techniques.</div></div>","PeriodicalId":50440,"journal":{"name":"Flow Measurement and Instrumentation","volume":"108 ","pages":"Article 103204"},"PeriodicalIF":2.7,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145976531","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

From sparse data to accurate flux: A hybrid modeling strategy for metering greenhouse gas emissions from dispersed industrial sources 从稀疏数据到精确通量：用于计量分散工业源温室气体排放的混合建模策略

IF 2.7 3区工程技术 Q2 ENGINEERING, MECHANICAL

Flow Measurement and Instrumentation

Pub Date : 2026-01-12 DOI: 10.1016/j.flowmeasinst.2025.103180

Ning Ding , Ye Wei , Yanheng Xi , Wenting Jiang , Jun Su , Ruiqi Zhang , Sixiang Yang , Tek Tjing Lie

Accurate carbon emission data is the foundation of greenhouse effect mitigation. Compared with carbon accounting, the quantitative data obtained through gas detection technology to monitor greenhouse gases (GHGs) directly emitted into the atmosphere seems more convincing. Continuous emission monitoring system (CEMS) quantify GHG emissions from fixed emission sources. However, the quantification of emissions in open scenarios cannot be achieved due to the difficulty in measuring key parameters. To solve this problem, we propose a carbon emission metering technology. Based on real meteorological data, machine learning and weight analysis were used to equivalently express immeasurable meteorological factors. Key parameters for metering were obtained through multi-physics coupling simulation. Taking Fuyang Industry in China as an example, using the real meteorological data in 2022 and 2023, the carbon emission metering results in February and August were 641.5 Kg and 277.6 Kg, respectively. This method could achieve real-time metering in open scenarios with sufficient computing power.

准确的碳排放数据是减缓温室效应的基础。与碳核算相比，通过气体检测技术获得的直接排放到大气中的温室气体（GHGs）的定量数据似乎更有说服力。连续排放监测系统（CEMS）对固定排放源的温室气体排放进行量化。然而，由于难以测量关键参数，无法实现开放情景下排放的量化。为了解决这一问题，我们提出了一种碳排放计量技术。以真实气象数据为基础，利用机器学习和权重分析等价表达不可测量的气象因子。通过多物理场耦合仿真，获得了测量的关键参数。以中国阜阳工业为例，利用2022年和2023年的真实气象数据，2月和8月的碳排放计量结果分别为641.5 Kg和277.6 Kg。该方法可以实现开放场景下的实时计量，具有足够的计算能力。

{"title":"From sparse data to accurate flux: A hybrid modeling strategy for metering greenhouse gas emissions from dispersed industrial sources","authors":"Ning Ding , Ye Wei , Yanheng Xi , Wenting Jiang , Jun Su , Ruiqi Zhang , Sixiang Yang , Tek Tjing Lie","doi":"10.1016/j.flowmeasinst.2025.103180","DOIUrl":"10.1016/j.flowmeasinst.2025.103180","url":null,"abstract":"<div><div>Accurate carbon emission data is the foundation of greenhouse effect mitigation. Compared with carbon accounting, the quantitative data obtained through gas detection technology to monitor greenhouse gases (GHGs) directly emitted into the atmosphere seems more convincing. Continuous emission monitoring system (CEMS) quantify GHG emissions from fixed emission sources. However, the quantification of emissions in open scenarios cannot be achieved due to the difficulty in measuring key parameters. To solve this problem, we propose a carbon emission metering technology. Based on real meteorological data, machine learning and weight analysis were used to equivalently express immeasurable meteorological factors. Key parameters for metering were obtained through multi-physics coupling simulation. Taking Fuyang Industry in China as an example, using the real meteorological data in 2022 and 2023, the carbon emission metering results in February and August were 641.5 Kg and 277.6 Kg, respectively. This method could achieve real-time metering in open scenarios with sufficient computing power.</div></div>","PeriodicalId":50440,"journal":{"name":"Flow Measurement and Instrumentation","volume":"108 ","pages":"Article 103180"},"PeriodicalIF":2.7,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146022797","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 flow and heat transfer characteristics of improved aero-engine labyrinth seal for suppressing tooth tip collision 用于抑制齿尖碰撞的改进型航空发动机迷宫密封流动与传热特性研究

IF 2.7 3区工程技术 Q2 ENGINEERING, MECHANICAL

Flow Measurement and Instrumentation

Pub Date : 2026-01-12 DOI: 10.1016/j.flowmeasinst.2026.103197

Guozhe Ren, Songhan Guo, Dan Sun, Yu Li, Zhining Zhang

Aiming at the typical labyrinth seal structure in aero-engine, established labyrinth seal model for rotor-stator cavity based on fluid-solid-thermal coupled. Considered the influence of centrifugal deformation and thermal deformation on flow and heat transfer characteristics of labyrinth seal. Through numerical research, it was found that conventional benchmark type labyrinth seal is prone to collision and wear during operation, which affects the normal operation of labyrinth seal. Therefore, the improved design of labyrinth seal disc was proposed to obtain the new improved labyrinth seal structure. The results show that improved structure D reduces the sensitivity of mass flow to clearance changes and suppresses the influence of wind resistance temperature rise when initial clearance is 0.3 mm. The calculation results tend to be stable, and final hot clearance is 0.112 mm, which is 62.66 % higher than initial clearance change, without collision and wear. Its working characteristic stability is stronger than benchmark type labyrinth seal. Reducing the length of bolts can reduce the relative change in hot mass flow compared to cold mass flow, weaken the influence of wind resistance temperature rise, and improve the stability of working characteristics. However, when sealing clearance is too small, collision and wear may still occur.

针对航空发动机典型的篦齿密封结构，建立了基于流固热耦合的动静腔篦齿密封模型。考虑了离心变形和热变形对迷宫式密封流动和传热特性的影响。通过数值研究发现，常规基准型迷宫式密封在工作过程中容易发生碰撞磨损，影响迷宫式密封的正常工作。为此，提出了对篦齿密封盘的改进设计，得到了新的改进篦齿密封结构。结果表明：当初始间隙为0.3 mm时，改进后的D结构降低了质量流量对间隙变化的敏感性，抑制了风阻温升的影响；计算结果趋于稳定，最终热间隙为0.112 mm，比初始间隙变化高62.66%，无碰撞磨损。其工作特性稳定性比基准型迷宫密封强。减小螺栓长度可以减小热质量流相对于冷质量流的相对变化，减弱风阻温升的影响，提高工作特性的稳定性。但是，当密封间隙太小时，仍然可能发生碰撞和磨损。

{"title":"Research on flow and heat transfer characteristics of improved aero-engine labyrinth seal for suppressing tooth tip collision","authors":"Guozhe Ren, Songhan Guo, Dan Sun, Yu Li, Zhining Zhang","doi":"10.1016/j.flowmeasinst.2026.103197","DOIUrl":"10.1016/j.flowmeasinst.2026.103197","url":null,"abstract":"<div><div>Aiming at the typical labyrinth seal structure in aero-engine, established labyrinth seal model for rotor-stator cavity based on fluid-solid-thermal coupled. Considered the influence of centrifugal deformation and thermal deformation on flow and heat transfer characteristics of labyrinth seal. Through numerical research, it was found that conventional benchmark type labyrinth seal is prone to collision and wear during operation, which affects the normal operation of labyrinth seal. Therefore, the improved design of labyrinth seal disc was proposed to obtain the new improved labyrinth seal structure. The results show that improved structure D reduces the sensitivity of mass flow to clearance changes and suppresses the influence of wind resistance temperature rise when initial clearance is 0.3 mm. The calculation results tend to be stable, and final hot clearance is 0.112 mm, which is 62.66 % higher than initial clearance change, without collision and wear. Its working characteristic stability is stronger than benchmark type labyrinth seal. Reducing the length of bolts can reduce the relative change in hot mass flow compared to cold mass flow, weaken the influence of wind resistance temperature rise, and improve the stability of working characteristics. However, when sealing clearance is too small, collision and wear may still occur.</div></div>","PeriodicalId":50440,"journal":{"name":"Flow Measurement and Instrumentation","volume":"109 ","pages":"Article 103197"},"PeriodicalIF":2.7,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039873","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Vibrothermal signal-based fault detection in 3D-printed polymer pump impellers 基于热振信号的3d打印聚合物泵叶轮故障检测

IF 2.7 3区工程技术 Q2 ENGINEERING, MECHANICAL

Flow Measurement and Instrumentation

Pub Date : 2026-01-11 DOI: 10.1016/j.flowmeasinst.2026.103201

Ankit Saxena, Sushila Rani

Given the advancements in rapid prototyping and the increasing adoption of polymer materials in the fluid pump industry, effective fault diagnosis of centrifugal pump (CP) impellers is crucial. This study presents a novel hybrid Convolutional Neural Network with a Random Forest (CNN-RF) framework for fault detection in a 3D-printed PA6 polymer impeller, addressing the unique vibration characteristics of polymers compared to metal impellers. This study extends prior research by incorporating a comparative analysis of multinomial logistic regression, random forest, decision tree, support vector machine (SVM), and gradient boosting machine (GBM), demonstrating their strengths and trade-offs in predictive maintenance applications. An experimental setup employing vibration and temperature sensors served to gather data under three impeller conditions: healthy, one missing blade, and impeller with deformed disk. Advanced signal processing techniques facilitated feature extraction, and precision, recall, F1 index, area under the ROC curve, mean absolute error (MAE), and mean squared error (MSE) assessed model performance. Furthermore, stratified K-fold cross-validation, calibration curves, confusion matrices, and PCA served to ensure model robustness. Results demonstrate that Hybrid CNN-RF achieves the highest accuracy (99.88 %), outperforming conventional classifiers, while Random Forest and GBM also show strong predictive performance. The study further highlights the correlation between impeller faults and motor temperature variations, reinforcing the importance of thermal monitoring in fault diagnosis. This research advances predictive maintenance strategies by introducing a scalable and highly reliable machine learning-driven framework, bridging a gap in the diagnostics of polymer-based pump components.

随着快速成型技术的进步和聚合物材料在流体泵工业中的应用越来越广泛，离心泵叶轮的有效故障诊断至关重要。本研究提出了一种新颖的混合卷积神经网络与随机森林（CNN-RF）框架，用于3d打印PA6聚合物叶轮的故障检测，解决了聚合物与金属叶轮相比独特的振动特性。本研究通过对多项逻辑回归、随机森林、决策树、支持向量机（SVM）和梯度增强机（GBM）的比较分析，扩展了先前的研究，展示了它们在预测性维护应用中的优势和权衡。采用振动和温度传感器的实验装置，在叶轮健康、缺叶和叶轮变形三种工况下进行数据采集。先进的信号处理技术促进了特征提取，精度、召回率、F1指数、ROC曲线下面积、平均绝对误差（MAE）和均方误差（MSE）评估了模型的性能。此外，分层K-fold交叉验证、校准曲线、混淆矩阵和主成分分析有助于确保模型的稳健性。结果表明，Hybrid CNN-RF达到了最高的准确率（99.88%），优于传统分类器，而Random Forest和GBM也表现出了较强的预测性能。研究进一步强调了叶轮故障与电机温度变化之间的相关性，强化了热监测在故障诊断中的重要性。该研究通过引入可扩展且高度可靠的机器学习驱动框架，推进了预测性维护策略，弥合了聚合物泵组件诊断方面的空白。

{"title":"Vibrothermal signal-based fault detection in 3D-printed polymer pump impellers","authors":"Ankit Saxena, Sushila Rani","doi":"10.1016/j.flowmeasinst.2026.103201","DOIUrl":"10.1016/j.flowmeasinst.2026.103201","url":null,"abstract":"<div><div>Given the advancements in rapid prototyping and the increasing adoption of polymer materials in the fluid pump industry, effective fault diagnosis of centrifugal pump (CP) impellers is crucial. This study presents a novel hybrid Convolutional Neural Network with a Random Forest (CNN-RF) framework for fault detection in a 3D-printed PA6 polymer impeller, addressing the unique vibration characteristics of polymers compared to metal impellers. This study extends prior research by incorporating a comparative analysis of multinomial logistic regression, random forest, decision tree, support vector machine (SVM), and gradient boosting machine (GBM), demonstrating their strengths and trade-offs in predictive maintenance applications. An experimental setup employing vibration and temperature sensors served to gather data under three impeller conditions: healthy, one missing blade, and impeller with deformed disk. Advanced signal processing techniques facilitated feature extraction, and precision, recall, F1 index, area under the ROC curve, mean absolute error (MAE), and mean squared error (MSE) assessed model performance. Furthermore, stratified K-fold cross-validation, calibration curves, confusion matrices, and PCA served to ensure model robustness. Results demonstrate that Hybrid CNN-RF achieves the highest accuracy (99.88 %), outperforming conventional classifiers, while Random Forest and GBM also show strong predictive performance. The study further highlights the correlation between impeller faults and motor temperature variations, reinforcing the importance of thermal monitoring in fault diagnosis. This research advances predictive maintenance strategies by introducing a scalable and highly reliable machine learning-driven framework, bridging a gap in the diagnostics of polymer-based pump components.</div></div>","PeriodicalId":50440,"journal":{"name":"Flow Measurement and Instrumentation","volume":"108 ","pages":"Article 103201"},"PeriodicalIF":2.7,"publicationDate":"2026-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145976532","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 the pulsed spraying characteristics of non-rotating deflector sprinklers 非旋转导流喷头脉冲喷射特性研究

IF 2.7 3区工程技术 Q2 ENGINEERING, MECHANICAL

Flow Measurement and Instrumentation

Pub Date : 2026-01-11 DOI: 10.1016/j.flowmeasinst.2026.103202

Peng Li , Xinkun Wang , Wenjing Wu , Fan Yang , Weiwei Yuan

Pulsed water pressure generated by a jet three-way device was employed to drive a non-rotating deflector sprinkler for pulsed spraying, aiming to alleviate such issues as poor jet breakup, excessive application rates, and water concentration at the end of the throw at low operating pressures. Comparative experiments were conducted between pulsed water pressure and steady water pressure spraying. Flow rate, throw radius, and water distribution were measured at pressures ranging from 50 kPa to 200 kPa, and application rate and irrigation uniformity were calculated. The results showed that for both pulsed and steady water pressure, the water distribution followed a triangular pattern with the highest volume near the end of the throw. However, under pulsed pressure, the base width of the triangle was approximately twice that under steady pressure, while its height was about half, indicating a pronounced peak-shaving and valley-filling effect. Compared with steady spraying, pulsed spraying reduced flow rate by 4.2 %–9.3 %; increased throw by 0.5 %–11.4 % at medium and low pressures (but decreased it by 3.3 % at high pressure); reduced the maximum application rate by 42.3 %–47.6 % and the average application rate by 35.8 %–53.4 %; and improved the Christiansen uniformity coefficient by 0.7 %–23.2 %. These findings demonstrate that the pulsed water pressure generated by the jet three-way device significantly improves water distribution and shows strong potential for further development.

利用射流三通装置产生的脉冲水压驱动非旋转偏转喷头进行脉冲喷洒，解决了射流破碎差、施用量过大、低工作压力下抛射尾水浓度过高等问题。进行了脉冲水压与稳态水压喷雾的对比试验。在50kpa ~ 200kpa压力范围内测量了流量、抛射半径和水量分布，并计算了施用量和灌溉均匀度。结果表明：在脉动水压和稳态水压下，水的分布均呈三角形，且在落水端附近体积最大；然而，在脉冲压力下，三角形的基底宽度大约是稳定压力下的两倍，而其高度约为一半，表明明显的削峰和填谷效应。与稳定喷涂相比，脉冲喷涂使流量降低4.2% ~ 9.3%；增加投掷0.5% - 11.4%在中低压（但减少3.3%在高压）；最大应用率降低42.3% ~ 47.6%，平均应用率降低35.8% ~ 53.4%；Christiansen均匀系数提高0.7% ~ 23.2%。这些结果表明，射流三通装置产生的脉冲水压显著改善了水的分布，具有很大的发展潜力。

{"title":"Research on the pulsed spraying characteristics of non-rotating deflector sprinklers","authors":"Peng Li , Xinkun Wang , Wenjing Wu , Fan Yang , Weiwei Yuan","doi":"10.1016/j.flowmeasinst.2026.103202","DOIUrl":"10.1016/j.flowmeasinst.2026.103202","url":null,"abstract":"<div><div>Pulsed water pressure generated by a jet three-way device was employed to drive a non-rotating deflector sprinkler for pulsed spraying, aiming to alleviate such issues as poor jet breakup, excessive application rates, and water concentration at the end of the throw at low operating pressures. Comparative experiments were conducted between pulsed water pressure and steady water pressure spraying. Flow rate, throw radius, and water distribution were measured at pressures ranging from 50 kPa to 200 kPa, and application rate and irrigation uniformity were calculated. The results showed that for both pulsed and steady water pressure, the water distribution followed a triangular pattern with the highest volume near the end of the throw. However, under pulsed pressure, the base width of the triangle was approximately twice that under steady pressure, while its height was about half, indicating a pronounced peak-shaving and valley-filling effect. Compared with steady spraying, pulsed spraying reduced flow rate by 4.2 %–9.3 %; increased throw by 0.5 %–11.4 % at medium and low pressures (but decreased it by 3.3 % at high pressure); reduced the maximum application rate by 42.3 %–47.6 % and the average application rate by 35.8 %–53.4 %; and improved the Christiansen uniformity coefficient by 0.7 %–23.2 %. These findings demonstrate that the pulsed water pressure generated by the jet three-way device significantly improves water distribution and shows strong potential for further development.</div></div>","PeriodicalId":50440,"journal":{"name":"Flow Measurement and Instrumentation","volume":"108 ","pages":"Article 103202"},"PeriodicalIF":2.7,"publicationDate":"2026-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145976603","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Modeling of ultrasonic blood flow sensor and its extended application discussion 超声血流传感器的建模及其扩展应用探讨

IF 2.7 3区工程技术 Q2 ENGINEERING, MECHANICAL

Flow Measurement and Instrumentation

Pub Date : 2026-01-10 DOI: 10.1016/j.flowmeasinst.2026.103200

Dandan Zheng, Manman Wu, Ying Xu, Zimeng Zheng, Chenglin Yv, Xueyong Chen, Haojun Fan

—Extracorporeal membrane oxygenation (ECMO) serves as a critical life-support technology for patients with cardiopulmonary failure. Ultrasonic blood flow sensors, as core ECMO components, enable accurate real-time flow monitoring. This study addresses the variation in measurement characteristics of clamp-on ultrasonic flowmeters in different media by proposing a flow-acoustic coupled numerical simulation method and establishing a data-driven correction coefficient prediction model. Experimental validation was subsequently performed using a developed high-precision ultrasonic blood flow sensor prototype. First, a coupled flow-acoustic simulation model was built through collaborative modeling of CFD and acoustic modules. Experimental validation in pure water and blood-mimicking fluid showed absolute errors between simulation and experimental results within ±3 % (flow rate ≥1 L/min), confirming the accuracy of the simulation method. Second, based on simulation data from multiple media at different temperatures, an artificial neural network (ANN)-based prediction model for instrument correction coefficients was developed. The mean absolute percentage errors (MAPE) for the training and test sets reached 0.543 % and 1.028 %, respectively, outperforming regression trees and support vector machines. The proposed method effectively eliminates medium-dependent measurement errors, enhances cross-population applicability, and promotes the generalization of ultrasonic sensors to industrial and pharmaceutical fields, thereby improving cross-medium adaptation capability of ultrasonic flow sensors. Finally, the prediction model was tested using the developed prototype ultrasonic blood flow sensor. Within the 1–10 L/min flow range, the measurement error was reduced from a maximum of 16 % to within ±10 % after correction, with repeatability better than 1 %, achieving performance comparable to commercial Transonic sensors.

体外膜氧合（extracorporeal membrane oxygenation， ECMO）是心肺衰竭患者重要的生命支持技术。超声血流传感器作为ECMO的核心部件，可实现精确的实时血流监测。针对夹紧式超声波流量计在不同介质中测量特性的变化，提出了流声耦合数值模拟方法，建立了数据驱动的校正系数预测模型。随后使用开发的高精度超声血流传感器原型进行实验验证。首先，通过CFD和声学模块的协同建模，建立了流声耦合仿真模型；在纯净水和血液模拟液中进行实验验证，模拟结果与实验结果的绝对误差在±3%以内（流速≥1 L/min），证实了模拟方法的准确性。其次，基于不同温度下多种介质的模拟数据，建立了基于人工神经网络的仪器校正系数预测模型；训练集和测试集的平均绝对百分比误差（MAPE）分别达到0.543%和1.028%，优于回归树和支持向量机。该方法有效地消除了介质依赖的测量误差，增强了跨种群适用性，促进了超声波传感器在工业和医药领域的推广，从而提高了超声波流量传感器的跨介质适应能力。最后，利用研制的超声血流传感器样机对预测模型进行了验证。在1 - 10 L/min的流量范围内，校正后的测量误差从最大的16%减少到±10%，重复性优于1%，实现了与商用跨音速传感器相当的性能。

{"title":"Modeling of ultrasonic blood flow sensor and its extended application discussion","authors":"Dandan Zheng, Manman Wu, Ying Xu, Zimeng Zheng, Chenglin Yv, Xueyong Chen, Haojun Fan","doi":"10.1016/j.flowmeasinst.2026.103200","DOIUrl":"10.1016/j.flowmeasinst.2026.103200","url":null,"abstract":"<div><div>—Extracorporeal membrane oxygenation (ECMO) serves as a critical life-support technology for patients with cardiopulmonary failure. Ultrasonic blood flow sensors, as core ECMO components, enable accurate real-time flow monitoring. This study addresses the variation in measurement characteristics of clamp-on ultrasonic flowmeters in different media by proposing a flow-acoustic coupled numerical simulation method and establishing a data-driven correction coefficient prediction model. Experimental validation was subsequently performed using a developed high-precision ultrasonic blood flow sensor prototype. First, a coupled flow-acoustic simulation model was built through collaborative modeling of CFD and acoustic modules. Experimental validation in pure water and blood-mimicking fluid showed absolute errors between simulation and experimental results within ±3 % (flow rate ≥1 L/min), confirming the accuracy of the simulation method. Second, based on simulation data from multiple media at different temperatures, an artificial neural network (ANN)-based prediction model for instrument correction coefficients was developed. The mean absolute percentage errors (MAPE) for the training and test sets reached 0.543 % and 1.028 %, respectively, outperforming regression trees and support vector machines. The proposed method effectively eliminates medium-dependent measurement errors, enhances cross-population applicability, and promotes the generalization of ultrasonic sensors to industrial and pharmaceutical fields, thereby improving cross-medium adaptation capability of ultrasonic flow sensors. Finally, the prediction model was tested using the developed prototype ultrasonic blood flow sensor. Within the 1–10 L/min flow range, the measurement error was reduced from a maximum of 16 % to within ±10 % after correction, with repeatability better than 1 %, achieving performance comparable to commercial Transonic sensors.</div></div>","PeriodicalId":50440,"journal":{"name":"Flow Measurement and Instrumentation","volume":"108 ","pages":"Article 103200"},"PeriodicalIF":2.7,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145976527","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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