International Journal of Heat and Mass Transfer最新文献

A novel numerical model for detailed simulation of CO2 capture 一种新的CO2捕获详细模拟数值模型

IF 5.8 2区工程技术 Q1 ENGINEERING, MECHANICAL

International Journal of Heat and Mass Transfer

Pub Date : 2026-01-30 DOI: 10.1016/j.ijheatmasstransfer.2026.128416

Sajad Jafari , Ran Yao , Salar Zamani Salimi , Luca Brandt , Christophe Duwig

According to the Intergovernmental Panel on Climate Change, CO₂ capture using liquid absorbents is a key strategy for mitigating climate change. However, the energy footprint of the technique is still high and novel solutions are needed to design better units. To that end, this study presents a novel and detailed diffuse-interface model for interfacial mass and heat transfer coupled with chemical reactions for CO₂ capture. Two scalar transport equations describe the species evolution in each phase, coupled through a variable apparent Henry’s constant that captures non-ideal vapor–liquid equilibrium at the interface. Momentum and energy transport are modeled through single-scalar formulations, with interfacial velocity discontinuities arising from reactive phase change, handled via a Stefan condition. A conservative phase-field method closes the equations, with regularization applied to suppress numerical diffusion when tracking the interface. The model resolves key physical phenomena, including reaction kinetics, mass transfer resistance, H₂O phase change, and interfacial velocity jumps during both absorption and desorption. A sensitivity analysis shows that increasing the solvent mole fraction enhances chemical reactivity but increases diffusive resistance, inducing complex nonlinear effects on the interfacial reactive transport. The coupled CO₂ and H₂O interphase transport are captured simultaneously, with water evaporation shown to have limited impact on CO₂ uptake and on the interfacial reactive Stefan velocity for isolated droplets. Additionally, multi-droplet simulations demonstrate that the droplet number and an imposed gas-phase mean flow significantly affect absorption rates and spatial asymmetry through droplet-droplet interactions and convective transport. The findings offer critical insights into interfacial CO₂ transport in reactive, two-phase systems and supports the need for advanced numerical studies like the present one, given the lack of droplet-scale data and the limited applicability of bulk-scale experiments to localized transient interfacial processes.

根据政府间气候变化专门委员会的说法，利用液体吸收剂捕获二氧化碳是减缓气候变化的关键战略。然而，该技术的能源足迹仍然很高，需要新的解决方案来设计更好的装置。为此，本研究提出了一种新颖而详细的扩散界面模型，用于界面质量和热量传递以及二氧化碳捕获的化学反应。两个标量输运方程描述了每个阶段的物种演化，通过一个可变的表观亨利常数耦合，该常数捕获了界面处的非理想汽液平衡。动量和能量输运通过单标量公式建模，反应相变化引起的界面速度不连续通过Stefan条件处理。采用保守相场法封闭方程，并采用正则化方法抑制界面跟踪时的数值扩散。该模型解决了吸附和脱附过程中的关键物理现象，包括反应动力学、传质阻力、H2O相变和界面速度跳变。灵敏度分析表明，溶剂摩尔分数的增加提高了化学反应活性，但增加了扩散阻力，对界面反应输运产生复杂的非线性效应。同时捕获耦合的CO2和H2O界面传输，水蒸发对CO2吸收和分离液滴的界面反应速度的影响有限。此外，多液滴模拟表明，液滴数和施加的气相平均流量通过液滴-液滴相互作用和对流输送显著影响吸收率和空间不对称性。由于缺乏液滴尺度的数据和有限的批量实验对局部瞬态界面过程的适用性，这些发现为反应性两相系统的界面CO2传输提供了重要的见解，并支持了像目前这样的高级数值研究的需要。

{"title":"A novel numerical model for detailed simulation of CO2 capture","authors":"Sajad Jafari , Ran Yao , Salar Zamani Salimi , Luca Brandt , Christophe Duwig","doi":"10.1016/j.ijheatmasstransfer.2026.128416","DOIUrl":"10.1016/j.ijheatmasstransfer.2026.128416","url":null,"abstract":"<div><div>According to the Intergovernmental Panel on Climate Change, CO<sub>2</sub> capture using liquid absorbents is a key strategy for mitigating climate change. However, the energy footprint of the technique is still high and novel solutions are needed to design better units. To that end, this study presents a novel and detailed diffuse-interface model for interfacial mass and heat transfer coupled with chemical reactions for CO<sub>2</sub> capture. Two scalar transport equations describe the species evolution in each phase, coupled through a variable apparent Henry’s constant that captures non-ideal vapor–liquid equilibrium at the interface. Momentum and energy transport are modeled through single-scalar formulations, with interfacial velocity discontinuities arising from reactive phase change, handled via a Stefan condition. A conservative phase-field method closes the equations, with regularization applied to suppress numerical diffusion when tracking the interface. The model resolves key physical phenomena, including reaction kinetics, mass transfer resistance, H<sub>2</sub>O phase change, and interfacial velocity jumps during both absorption and desorption. A sensitivity analysis shows that increasing the solvent mole fraction enhances chemical reactivity but increases diffusive resistance, inducing complex nonlinear effects on the interfacial reactive transport. The coupled CO<sub>2</sub> and H<sub>2</sub>O interphase transport are captured simultaneously, with water evaporation shown to have limited impact on CO<sub>2</sub> uptake and on the interfacial reactive Stefan velocity for isolated droplets. Additionally, multi-droplet simulations demonstrate that the droplet number and an imposed gas-phase mean flow significantly affect absorption rates and spatial asymmetry through droplet-droplet interactions and convective transport. The findings offer critical insights into interfacial CO<sub>2</sub> transport in reactive, two-phase systems and supports the need for advanced numerical studies like the present one, given the lack of droplet-scale data and the limited applicability of bulk-scale experiments to localized transient interfacial processes.</div></div>","PeriodicalId":336,"journal":{"name":"International Journal of Heat and Mass Transfer","volume":"260 ","pages":"Article 128416"},"PeriodicalIF":5.8,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146075862","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

Enhanced boiling heat transfer in silicon-based heat sinks with hybrid jet/gradient-density-pin-fin-microchannel 混合射流/梯度-密度-针鳍-微通道硅基散热器中沸腾传热的增强

IF 5.8 2区工程技术 Q1 ENGINEERING, MECHANICAL

International Journal of Heat and Mass Transfer

Pub Date : 2026-01-30 DOI: 10.1016/j.ijheatmasstransfer.2026.128423

Xia Hua, Huiying Wu, Jinya Liu, Zhenyu Liu

To enhance boiling heat transfer for ultra-high heat flux dissipation in chips, silicon-based heat sinks with hybrid jet/gradient-density-pin-fin-microchannel are proposed. Specifically, hybrid jet/uniform-pin-fin-microchannel (JUPM, serving as a control case without gradient density design), hybrid jet/dense-to-sparse-pin-fin-microchannel (JDPM), and hybrid jet/sparse-to-dense-pin-fin-microchannel (JSPM) are fabricated. By combining a data acquisition system with a high-speed microscope camera, the heat transfer performance, boiling flow patterns, pressure drop and coefficient of performance (COP) of deionized water in JUPM, JDPM, JSPM across a range of jet velocities (V_j = 1.5, 2, 2.5 m/s) and inlet subcoolings (ΔT_sub = 30, 50, 70°C) are experimentally investigated. The results are further benchmarked against those from the heat sink with hybrid jet/continuous-microchannel (JCM), revealing that: 1) JSPM, JDPM and JUPM all improve the critical heat flux (CHF) due to the enhanced liquid supply to the heating surface (caused by the suppressed reverse flow and capillary-driven liquid bridges). Moreover, JSPM further increases the CHF compared to JDPM and JUPM due to the presence of more liquid bridges downstream. Particularly, JSPM achieves the highest CHF of 1464 W/cm² when V_j = 2.5 m/s and ΔT_sub = 70 °C; 2) JSPM, JDPM and JUPM all increase the HTC due to more nucleation sites in the early boiling stage, the highly-efficient thin film evaporation in the middle boiling stage, and the enhanced liquid supply to the heating surface in the late boiling stage. Moreover, JSPM has a higher HTC than JDPM and JUPM due to the further enhanced liquid supply; 3) JUPM, JDPM and JSPM all reduce the pressure drop due to the suppression of reverse flow and violent vapor generation from local dry out. Although they have comparable pressure drop, JSPM yields a slightly lower value in the late boiling stage; 4) JSPM, JUPM and JDPM all have better thermal-hydraulic performance (i.e., higher COP). Moreover, JSPM achieves superior performance compared to JUPM and JDPM, with a maximum COP of 13197 when V_j = 1.5 m/s and ΔT_sub = 70 °C.

为了在芯片中实现超高热流密度耗散，提出了一种混合射流/梯度密度-针鳍-微通道的硅基散热器。具体而言，制备了混合射流/均匀针鳍微通道（JUPM，作为无梯度密度设计的控制案例）、混合射流/密集到稀疏针鳍微通道（JDPM）和混合射流/稀疏到密集针鳍微通道（JSPM）。通过数据采集系统与高速显微镜相机相结合，实验研究了JUPM、JDPM和JSPM中去离子水在射流速度（Vj = 1.5、2、2.5 m/s）和入口过冷（ΔTsub = 30、50、70°C）范围内的传热性能、沸腾流型、压降和性能系数（COP）。结果表明：1)JSPM、JDPM和JUPM均提高了受热面的临界热流密度（CHF），这是由于抑制了逆流和毛细管驱动的液体桥导致的液体供应增加。此外，由于下游存在更多的液体桥，JSPM比JDPM和JUPM进一步增加了CHF。特别是当Vj = 2.5 m/s， ΔTsub = 70℃时，JSPM的CHF最高，达到1464 W/cm2；2) JSPM、JDPM和JUPM均因沸腾前期成核位点较多、沸腾中期薄膜蒸发效率高、沸腾后期受热面供液量增加而提高了HTC。此外，JSPM比JDPM和JUPM有更高的HTC，这是由于液体供应进一步增加；3) JUPM、JDPM和JSPM均通过抑制逆流和局部干透产生的剧烈蒸汽而降低压降。虽然它们具有相当的压降，但JSPM在沸腾后期产生的压降略低；4) JSPM、JUPM和JDPM均具有较好的热工性能（COP较高）。JSPM在Vj = 1.5 m/s、ΔTsub = 70°C时的COP最大值为13197，性能优于JUPM和JDPM。

{"title":"Enhanced boiling heat transfer in silicon-based heat sinks with hybrid jet/gradient-density-pin-fin-microchannel","authors":"Xia Hua, Huiying Wu, Jinya Liu, Zhenyu Liu","doi":"10.1016/j.ijheatmasstransfer.2026.128423","DOIUrl":"10.1016/j.ijheatmasstransfer.2026.128423","url":null,"abstract":"<div><div>To enhance boiling heat transfer for ultra-high heat flux dissipation in chips, silicon-based heat sinks with hybrid jet/gradient-density-pin-fin-microchannel are proposed. Specifically, hybrid jet/uniform-pin-fin-microchannel (JUPM, serving as a control case without gradient density design), hybrid jet/dense-to-sparse-pin-fin-microchannel (JDPM), and hybrid jet/sparse-to-dense-pin-fin-microchannel (JSPM) are fabricated. By combining a data acquisition system with a high-speed microscope camera, the heat transfer performance, boiling flow patterns, pressure drop and coefficient of performance (COP) of deionized water in JUPM, JDPM, JSPM across a range of jet velocities (<em>V</em><sub>j</sub> = 1.5, 2, 2.5 m/s) and inlet subcoolings (Δ<em>T</em><sub>sub</sub> = 30, 50, 70°C) are experimentally investigated. The results are further benchmarked against those from the heat sink with hybrid jet/continuous-microchannel (JCM), revealing that: 1) JSPM, JDPM and JUPM all improve the critical heat flux (CHF) due to the enhanced liquid supply to the heating surface (caused by the suppressed reverse flow and capillary-driven liquid bridges). Moreover, JSPM further increases the CHF compared to JDPM and JUPM due to the presence of more liquid bridges downstream. Particularly, JSPM achieves the highest CHF of 1464 W/cm<sup>2</sup> when <em>V</em><sub>j</sub> = 2.5 m/s and Δ<em>T</em><sub>sub</sub> = 70 °C; 2) JSPM, JDPM and JUPM all increase the HTC due to more nucleation sites in the early boiling stage, the highly-efficient thin film evaporation in the middle boiling stage, and the enhanced liquid supply to the heating surface in the late boiling stage. Moreover, JSPM has a higher HTC than JDPM and JUPM due to the further enhanced liquid supply; 3) JUPM, JDPM and JSPM all reduce the pressure drop due to the suppression of reverse flow and violent vapor generation from local dry out. Although they have comparable pressure drop, JSPM yields a slightly lower value in the late boiling stage; 4) JSPM, JUPM and JDPM all have better thermal-hydraulic performance (i.e., higher COP). Moreover, JSPM achieves superior performance compared to JUPM and JDPM, with a maximum COP of 13197 when <em>V</em><sub>j</sub> = 1.5 m/s and Δ<em>T</em><sub>sub</sub> = 70 °C.</div></div>","PeriodicalId":336,"journal":{"name":"International Journal of Heat and Mass Transfer","volume":"260 ","pages":"Article 128423"},"PeriodicalIF":5.8,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146076283","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

Identification of thermal conductivities for nonhomogeneous materials based on the hybrid machine learning model 基于混合机器学习模型的非均匀材料导热系数辨识

IF 5.8 2区工程技术 Q1 ENGINEERING, MECHANICAL

International Journal of Heat and Mass Transfer

Pub Date : 2026-01-30 DOI: 10.1016/j.ijheatmasstransfer.2026.128446

Haolong Chen , Yuanlin Yi , Zhaotao Liu , Zhanli Liu , Huanlin Zhou

A hybrid machine learning model, combining the level set, the discrete cosine transform (DCT), with the convolutional neural network (CNN), is proposed for identifying thermal conductivities for nonhomogeneous materials in heat conduction problems without prior knowledge. The finite element method is used to analyze the heat conduction in nonhomogeneous materials. The level set approach can segment the temperature RGB image and avoid tracking the evolution process of the closed curve. The DCT is used to extract the main features and reduce dimensionality both in the temperature field and the thermal conductivity. The CNN is developed to establish the relationship between the input (main features of temperature field) and the output (main features of thermal conductivity). Finally, the unknown thermal conductivities of the nonhomogeneous material can be acquired by the inverse DCT. The impacts of varying main feature sizes, measurement errors, and training datasets on the outcomes are systematically analyzed to validate the proposed methodology. As the main feature size increases and measurement error decreases, the estimated results become more accurate. The study provides a robust method for reverse identification of thermal conductivity for nonhomogeneous materials, with potential applications in spanning real-time elastography and high-throughput non-destructive evaluation techniques.

提出了一种结合水平集、离散余弦变换（DCT）和卷积神经网络（CNN）的混合机器学习模型，用于在没有先验知识的情况下识别热传导问题中非均匀材料的导热系数。采用有限元法对非均匀材料的热传导进行了分析。水平集方法可以分割温度RGB图像，避免跟踪封闭曲线的演化过程。DCT用于提取温度场和导热系数的主要特征并进行降维。开发CNN是为了建立输入（温度场的主要特征）和输出（导热系数的主要特征）之间的关系。最后，非均匀材料的未知热导率可以通过反离散余弦变换得到。系统地分析了不同主要特征大小、测量误差和训练数据集对结果的影响，以验证所提出的方法。随着主特征尺寸的增大和测量误差的减小，估计结果越来越准确。该研究为非均匀材料的热导率反向识别提供了一种强大的方法，在跨越实时弹性成像和高通量无损评估技术方面具有潜在的应用前景。

{"title":"Identification of thermal conductivities for nonhomogeneous materials based on the hybrid machine learning model","authors":"Haolong Chen , Yuanlin Yi , Zhaotao Liu , Zhanli Liu , Huanlin Zhou","doi":"10.1016/j.ijheatmasstransfer.2026.128446","DOIUrl":"10.1016/j.ijheatmasstransfer.2026.128446","url":null,"abstract":"<div><div>A hybrid machine learning model, combining the level set, the discrete cosine transform (DCT), with the convolutional neural network (CNN), is proposed for identifying thermal conductivities for nonhomogeneous materials in heat conduction problems without prior knowledge. The finite element method is used to analyze the heat conduction in nonhomogeneous materials. The level set approach can segment the temperature RGB image and avoid tracking the evolution process of the closed curve. The DCT is used to extract the main features and reduce dimensionality both in the temperature field and the thermal conductivity. The CNN is developed to establish the relationship between the input (main features of temperature field) and the output (main features of thermal conductivity). Finally, the unknown thermal conductivities of the nonhomogeneous material can be acquired by the inverse DCT. The impacts of varying main feature sizes, measurement errors, and training datasets on the outcomes are systematically analyzed to validate the proposed methodology. As the main feature size increases and measurement error decreases, the estimated results become more accurate. The study provides a robust method for reverse identification of thermal conductivity for nonhomogeneous materials, with potential applications in spanning real-time elastography and high-throughput non-destructive evaluation techniques.</div></div>","PeriodicalId":336,"journal":{"name":"International Journal of Heat and Mass Transfer","volume":"260 ","pages":"Article 128446"},"PeriodicalIF":5.8,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146075852","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 synergistic influence of PA-based composite phase change materials and heat exchange structures on heat storage performance pa基复合相变材料与换热结构对储热性能的协同影响

IF 5.8 2区工程技术 Q1 ENGINEERING, MECHANICAL

International Journal of Heat and Mass Transfer

Pub Date : 2026-01-30 DOI: 10.1016/j.ijheatmasstransfer.2026.128421

Yize Liu , Xin Liu , Guiju Xing , Zhaoyi Huo , Lili Niu , Haobo Zhang , Yicheng Liu

Low thermal conductivity is widely acknowledged as a major limitation of phase change materials (PCMs), as it markedly affects phase-change heat transfer and system-level performance and thereby restricts practical deployment. Most previous studies have treated fin enhancement and conductive additives separately, whereas a quantitative and systematic analysis of the coupled synergistic mechanism between staggered fins and high loadings of expanded graphite (EG) is still lacking. In this work, experiments combined with validated three-dimensional numerical simulations were conducted under the same test unit and identical boundary conditions to examine 16 cases formed by four heat-transfer structures and four PA-based composite PCMs with EG additions of 0–6 wt%. The results show that EG builds a conductive network, increasing the thermal conductivity to 2.103 W·m⁻¹·K⁻¹ at 6 wt% EG (7.86 times that of PA), but reducing the total stored heat by 10.3%–10.4%. Fin addition accelerates heat transfer: parallel or staggered fins reduce the time for the geometric center of the storage unit to reach 50°C from 27 min to 13 min (a 52% reduction). A pronounced synergy is observed when the composite PCM is coupled with the fin structure. The staggered-fin case with 6 wt% EG yields a synergy index of S = 1.57, further shortening the time to reach 50°C to 2.3 min (a 91.5% reduction), while increasing the average charging power by 323.3% relative to Case 1. These results quantitatively demonstrate the synergistic enhancement provided by staggered fins combined with high EG loading, showing that charging power and thermal response capacity can be substantially improved at the cost of only a modest decrease in stored energy and providing guidance for the design of high-performance phase-change heat exchangers.

低导热系数被广泛认为是相变材料（PCMs）的主要限制，因为它显著影响相变传热和系统级性能，从而限制了实际应用。以往的研究大多将翅片增强和导电添加剂分开处理，而对交错翅片与膨胀石墨（EG）高负荷耦合增效机理的定量和系统分析仍然缺乏。在相同的测试单元和相同的边界条件下，实验结合验证的三维数值模拟，研究了由四种传热结构和四种pa基复合PCMs （EG添加量为0-6 wt%）形成的16种情况。结果表明，EG建立了一个传导网络，在6 wt% EG（是PA的7.86倍）的情况下，将导热系数提高到2.103 W·m（⁻¹·K），但减少了10.3%-10.4%的总储存热量。增加翅片加速传热：平行或交错的翅片减少了存储单元几何中心从27分钟达到50°C的时间（减少52%）。当复合PCM与鳍结构耦合时，观察到明显的协同作用。6 wt% EG的错开鳍案例的协同指数为S = 1.57，进一步缩短了达到50°C的时间至2.3分钟（减少91.5%），同时相对于案例1，平均充电功率提高了323.3%。这些结果定量地证明了交错翅片与高EG负荷相结合所提供的协同增强作用，表明充电功率和热响应能力可以大幅提高，而存储能量只会适度减少，这为高性能相变换热器的设计提供了指导。

{"title":"The synergistic influence of PA-based composite phase change materials and heat exchange structures on heat storage performance","authors":"Yize Liu , Xin Liu , Guiju Xing , Zhaoyi Huo , Lili Niu , Haobo Zhang , Yicheng Liu","doi":"10.1016/j.ijheatmasstransfer.2026.128421","DOIUrl":"10.1016/j.ijheatmasstransfer.2026.128421","url":null,"abstract":"<div><div>Low thermal conductivity is widely acknowledged as a major limitation of phase change materials (PCMs), as it markedly affects phase-change heat transfer and system-level performance and thereby restricts practical deployment. Most previous studies have treated fin enhancement and conductive additives separately, whereas a quantitative and systematic analysis of the coupled synergistic mechanism between staggered fins and high loadings of expanded graphite (EG) is still lacking. In this work, experiments combined with validated three-dimensional numerical simulations were conducted under the same test unit and identical boundary conditions to examine 16 cases formed by four heat-transfer structures and four PA-based composite PCMs with EG additions of 0–6 wt%. The results show that EG builds a conductive network, increasing the thermal conductivity to 2.103 W·m⁻¹·K⁻¹ at 6 wt% EG (7.86 times that of PA), but reducing the total stored heat by 10.3%–10.4%. Fin addition accelerates heat transfer: parallel or staggered fins reduce the time for the geometric center of the storage unit to reach 50°C from 27 min to 13 min (a 52% reduction). A pronounced synergy is observed when the composite PCM is coupled with the fin structure. The staggered-fin case with 6 wt% EG yields a synergy index of S = 1.57, further shortening the time to reach 50°C to 2.3 min (a 91.5% reduction), while increasing the average charging power by 323.3% relative to Case 1. These results quantitatively demonstrate the synergistic enhancement provided by staggered fins combined with high EG loading, showing that charging power and thermal response capacity can be substantially improved at the cost of only a modest decrease in stored energy and providing guidance for the design of high-performance phase-change heat exchangers.</div></div>","PeriodicalId":336,"journal":{"name":"International Journal of Heat and Mass Transfer","volume":"260 ","pages":"Article 128421"},"PeriodicalIF":5.8,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146075859","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

Bonding heterogeneity and unique transport channel leading to anomalous thermal conductivity in BaMN2 (M = Ti, Zr, Hf) BaMN2 （M = Ti, Zr, Hf）中化学键非均质性和独特输运通道导致异常导热性

IF 5.8 2区工程技术 Q1 ENGINEERING, MECHANICAL

International Journal of Heat and Mass Transfer

Pub Date : 2026-01-30 DOI: 10.1016/j.ijheatmasstransfer.2026.128451

Pin-Zhen Jia , Xiao-Gen Deng , Hong-Yu Chen , Wu-Xing Zhou , Xue-Kun Chen

Investigating quasi-low-dimensional bulk materials, rather than true low-dimensional materials, is of paramount importance from both fundamental and practical standpoints. Layered nitrides BaMN₂ (M = Ti, Zr, Hf) are promising thermoelectric materials because of their unique geometrical and electronic structures. In this context, we utilized first-principles calculations combined with the Boltzmann Transport Equation (BTE) to investigate the phonon dynamics and transport in BaMN₂ (M = Ti, Zr, Hf). It is found that the three-phonon limited lattice thermal conductivity (

κ_{l}

) of BaTiN₂ (8.5 W/m K) is the highest at 300 K, followed by BaZrN₂ (5.6 W/m K) and BaHfN₂ (7.6 W/m K) at 300 K, which deviates from the conventional Keyes theory’s prediction that heavier compounds generally exhibit lower

κ_{l}

. Including four-phonon scattering lowers

κ_{l}

, yet the thermal transport tendency remains. Comparative analysis of the phonon mode-resolved properties of BaZrN₂ and BaHfN₂ reveal that the transition metal (HfN₂)²⁻layer could provide additional acoustic phonon transport channel for BaHfN₂, and the higher bonding heterogeneity in BaZrN₂ induces strong phonon scattering rates for low-frequency optical modes, both resulting in the anomalous

κ_{l}

tendency. Our investigation elucidates the complex and multifaceted microscopic mechanisms governing the anomalous mass dependence of

κ_{l}

in the quasi-two-dimensional (Q-2D) BaMN₂ system.

研究准低维块状材料，而不是真正的低维材料，从基础和实际的角度来看都是至关重要的。层状氮化物BaMN2 （M = Ti, Zr, Hf）具有独特的几何结构和电子结构，是一种很有前途的热电材料。在这种情况下，我们利用第一性原理计算结合玻尔兹曼输运方程（BTE）研究了BaMN₂（M = Ti, Zr, Hf）中的声子动力学和输运。在300 K时，BaTiN₂（8.5 W/m K）的三声子极限晶格热导率（κl）最高，其次是BaZrN₂（5.6 W/m K）和BaHfN₂（7.6 W/m K），这与传统Keyes理论预测的较重的化合物通常具有较低的κl不同。包括四声子散射降低了1%，但热输运趋势仍然存在。对比分析了BaZrN₂和BaHfN₂的声子模式分辨特性，发现过渡金属（HfN₂）2−层可以为BaHfN₂提供额外的声子传输通道，而BaZrN₂中较高的键非均质性导致了低频光学模式的强声子散射率，两者都导致了异常的κl趋势。我们的研究阐明了准二维（Q-2D） BaMN₂体系中κlin异常质量依赖的复杂和多方面的微观机制。

{"title":"Bonding heterogeneity and unique transport channel leading to anomalous thermal conductivity in BaMN2 (M = Ti, Zr, Hf)","authors":"Pin-Zhen Jia , Xiao-Gen Deng , Hong-Yu Chen , Wu-Xing Zhou , Xue-Kun Chen","doi":"10.1016/j.ijheatmasstransfer.2026.128451","DOIUrl":"10.1016/j.ijheatmasstransfer.2026.128451","url":null,"abstract":"<div><div>Investigating quasi-low-dimensional bulk materials, rather than true low-dimensional materials, is of paramount importance from both fundamental and practical standpoints. Layered nitrides BaMN<sub>2</sub> (<em>M</em> = Ti, Zr, Hf) are promising thermoelectric materials because of their unique geometrical and electronic structures. In this context, we utilized first-principles calculations combined with the Boltzmann Transport Equation (BTE) to investigate the phonon dynamics and transport in BaMN₂ (<em>M</em> = Ti, Zr, Hf). It is found that the three-phonon limited lattice thermal conductivity (<span><math><msub><mi>κ</mi><mi>l</mi></msub></math></span>) of BaTiN₂ (8.5 W/m K) is the highest at 300 K, followed by BaZrN₂ (5.6 W/m K) and BaHfN₂ (7.6 W/m K) at 300 K, which deviates from the conventional Keyes theory’s prediction that heavier compounds generally exhibit lower <span><math><msub><mi>κ</mi><mi>l</mi></msub></math></span>. Including four-phonon scattering lowers<span><math><msub><mi>κ</mi><mi>l</mi></msub></math></span>, yet the thermal transport tendency remains. Comparative analysis of the phonon mode-resolved properties of BaZrN₂ and BaHfN₂ reveal that the transition metal (HfN₂)<sup>2−</sup>layer could provide additional acoustic phonon transport channel for BaHfN₂, and the higher bonding heterogeneity in BaZrN₂ induces strong phonon scattering rates for low-frequency optical modes, both resulting in the anomalous <span><math><msub><mi>κ</mi><mi>l</mi></msub></math></span> tendency. Our investigation elucidates the complex and multifaceted microscopic mechanisms governing the anomalous mass dependence of <span><math><msub><mi>κ</mi><mi>l</mi></msub></math></span>in the quasi-two-dimensional (Q-2D) BaMN₂ system.</div></div>","PeriodicalId":336,"journal":{"name":"International Journal of Heat and Mass Transfer","volume":"260 ","pages":"Article 128451"},"PeriodicalIF":5.8,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146075913","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

Comprehensive performance evaluation and advanced design optimization of triply periodic minimal surface (TPMS) heat sinks 三周期最小表面（TPMS）散热器的综合性能评价与先进设计优化

IF 5.8 2区工程技术 Q1 ENGINEERING, MECHANICAL

International Journal of Heat and Mass Transfer

Pub Date : 2026-01-30 DOI: 10.1016/j.ijheatmasstransfer.2026.128452

Jiaxuan Wang , Fengrui Zhang , Zhenyu Hou , Chenyi Qian , Junye Shi , Jiangping Chen , Binbin Yu

This study focuses on optimizing heat sink performance through innovative Triply Periodic Minimal Surface (TPMS) structures. We analyze five heat sink structures: Fischer-Koch, Gyroid, Diamond, Split-P, and Fin. Fischer-Koch shows the highest heat transfer capacity (40 W/K), while Diamond has the best overall performance. Split-P has the highest flow resistance, making it unsuitable for high-flow applications. Based on flow and heat transfer mechanisms, we identify superior regions in each structure and integrate them into a novel mixed Diamond-Fischer-Koch-Gyroid design, which has been less explored in prior literature to exploit their complementary advantages. This mixed design reduces pressure drop by 15% and achieves a maximum heat transfer capacity of 37.6 W/K under a 15 kPa flow resistance constraint. Additionally, we originally apply particle swarm optimization (PSO) to optimize the wall thickness gradient fields of Gyroid and Diamond structures. The optimized designs achieve average wall temperature reductions of 2.8 K and 3.6 K, with performance improvements of 4.52% and 4.74%, respectively. These results demonstrate the promising potential of mixed TPMS designs and PSO gradient optimization to overcome the limitations of conventional structures, providing guidance for advanced heat sink design.

本研究的重点是通过创新的三周期最小表面（TPMS）结构优化散热器性能。我们分析了五种散热器结构：Fischer-Koch， Gyroid, Diamond， Split-P和Fin. Fischer-Koch的传热能力最高（40 W/K），而Diamond的综合性能最好。Split-P具有最高的流动阻力，使其不适合高流量应用。基于流动和传热机制，我们确定了每个结构中的优势区域，并将它们整合到一个新的混合钻石-费舍尔-科赫-陀螺设计中，这在先前的文献中很少被探索，以利用它们的互补优势。这种混合设计减少了15%的压降，在15 kPa流阻约束下达到了37.6 W/K的最大换热能力。此外，我们还将粒子群优化（PSO）应用于Gyroid和Diamond结构的壁厚梯度场优化。优化后的壁温平均降低了2.8 K和3.6 K，性能分别提高了4.52%和4.74%。这些结果表明，混合TPMS设计和PSO梯度优化具有克服传统结构局限性的良好潜力，为先进的散热器设计提供了指导。

{"title":"Comprehensive performance evaluation and advanced design optimization of triply periodic minimal surface (TPMS) heat sinks","authors":"Jiaxuan Wang , Fengrui Zhang , Zhenyu Hou , Chenyi Qian , Junye Shi , Jiangping Chen , Binbin Yu","doi":"10.1016/j.ijheatmasstransfer.2026.128452","DOIUrl":"10.1016/j.ijheatmasstransfer.2026.128452","url":null,"abstract":"<div><div>This study focuses on optimizing heat sink performance through innovative Triply Periodic Minimal Surface (TPMS) structures. We analyze five heat sink structures: Fischer-Koch, Gyroid, Diamond, Split-P, and Fin. Fischer-Koch shows the highest heat transfer capacity (40 W/K), while Diamond has the best overall performance. Split-P has the highest flow resistance, making it unsuitable for high-flow applications. Based on flow and heat transfer mechanisms, we identify superior regions in each structure and integrate them into a novel mixed Diamond-Fischer-Koch-Gyroid design, which has been less explored in prior literature to exploit their complementary advantages. This mixed design reduces pressure drop by 15% and achieves a maximum heat transfer capacity of 37.6 W/K under a 15 kPa flow resistance constraint. Additionally, we originally apply particle swarm optimization (PSO) to optimize the wall thickness gradient fields of Gyroid and Diamond structures. The optimized designs achieve average wall temperature reductions of 2.8 K and 3.6 K, with performance improvements of 4.52% and 4.74%, respectively. These results demonstrate the promising potential of mixed TPMS designs and PSO gradient optimization to overcome the limitations of conventional structures, providing guidance for advanced heat sink design.</div></div>","PeriodicalId":336,"journal":{"name":"International Journal of Heat and Mass Transfer","volume":"260 ","pages":"Article 128452"},"PeriodicalIF":5.8,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146075851","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

Fundamental understanding of flow distribution control in parallel multi-channel heat transfer devices with electrohydrodynamic conduction pumping 电流体动力传导泵并联多通道传热装置流场控制的基本认识

IF 5.8 2区工程技术 Q1 ENGINEERING, MECHANICAL

International Journal of Heat and Mass Transfer

Pub Date : 2026-01-29 DOI: 10.1016/j.ijheatmasstransfer.2026.128445

Mana Masrouri, Jamal Yagoobi

Flow maldistribution in heat transfer devices such as parallel tube heat exchangers is an undesirable phenomenon that results from uneven pressure drops among the channels. Factors like non-uniform heating, geometric configuration, and system orientation can contribute to these imbalances in pressure drop, which can significantly diminish the overall energy performance of the system. In this study, Electrohydrodynamic (EHD) conduction pumping has been fundamentally studied as an active method to control the distribution of the flow in multi-channel heat exchangers. EHD conduction pumping is based on a net Coulomb force applied to a dielectric fluid. This force arises due to the formation of heterocharge layers, layers of opposite polarity, near asymmetric submerged electrodes, which are caused by enhanced dissociation of impurities in the fluid under a strong electric field. In this work, a two-dimensional domain representing a mesoscale heat exchanger using HFE-7100 as the working fluid was numerically investigated. The device comprises two parallel channels (each 7 cm long and 0.5 cm high), with electrohydrodynamic (EHD) conduction pumps embedded at the inlet. After establishing a foundational understanding of the influence of EHD conduction-driven flow distribution control on the system’s thermal performance under uneven heating loads at steady-state, an emphasis is placed on transient processes that govern the system’s response to sudden changes in operating conditions, as commonly encountered during flow distribution adjustment, going beyond the earlier work in this domain. Results show that transient charge accumulation takes 5.5 ms to reach a steady-state magnitude of (0.000826–0.000829) C/m, depending on the applied potential. This time is in the order of charge relaxation time. Additionally, the novel effect of pulsation on EHD-driven flow distribution control is investigated in detail to fundamentally understand its impact through the transient formation of the heterocharge layers. The contribution of pulsation is shown to depend strongly on the operating conditions and the applied frequency.

在平行管式换热器等换热装置中，由于通道间压降不均匀而引起的流动不均匀是一种不良现象。加热不均匀、几何结构和系统朝向等因素都可能导致压降不平衡，从而显著降低系统的整体能源性能。本研究从根本上研究了电流体动力（EHD）传导泵送作为控制多通道换热器内流动分布的一种主动方法。EHD传导泵浦是基于施加在介电流体上的净库仑力。这种力是由于在不对称浸没电极附近形成异电荷层（极性相反的层）而产生的，这是由于在强电场下流体中杂质的解离增强引起的。本文对以HFE-7100为工质的中尺度换热器的二维区域进行了数值研究。该装置包括两个平行通道（每个通道长7厘米，高0.5厘米），在入口嵌入电流体动力（EHD）传导泵。在对稳态不均匀热负荷下EHD传导驱动的流量分布控制对系统热性能的影响有了基本的了解之后，重点放在控制系统对运行条件突然变化的响应的瞬态过程上，这是在流量分布调整过程中经常遇到的，超出了该领域的早期工作。结果表明，根据外加电位的不同，瞬态电荷积累需要5.5 ms才能达到（0.000826-0.000829）C/m的稳态量级。这个时间是电荷弛豫时间的顺序。此外，详细研究了脉动对ehd驱动的流动分布控制的新影响，从根本上了解其通过异质电荷层的瞬态形成的影响。脉动的贡献很大程度上取决于工作条件和施加的频率。

{"title":"Fundamental understanding of flow distribution control in parallel multi-channel heat transfer devices with electrohydrodynamic conduction pumping","authors":"Mana Masrouri, Jamal Yagoobi","doi":"10.1016/j.ijheatmasstransfer.2026.128445","DOIUrl":"10.1016/j.ijheatmasstransfer.2026.128445","url":null,"abstract":"<div><div>Flow maldistribution in heat transfer devices such as parallel tube heat exchangers is an undesirable phenomenon that results from uneven pressure drops among the channels. Factors like non-uniform heating, geometric configuration, and system orientation can contribute to these imbalances in pressure drop, which can significantly diminish the overall energy performance of the system. In this study, Electrohydrodynamic (EHD) conduction pumping has been fundamentally studied as an active method to control the distribution of the flow in multi-channel heat exchangers. EHD conduction pumping is based on a net Coulomb force applied to a dielectric fluid. This force arises due to the formation of heterocharge layers, layers of opposite polarity, near asymmetric submerged electrodes, which are caused by enhanced dissociation of impurities in the fluid under a strong electric field. In this work, a two-dimensional domain representing a mesoscale heat exchanger using HFE-7100 as the working fluid was numerically investigated. The device comprises two parallel channels (each 7 cm long and 0.5 cm high), with electrohydrodynamic (EHD) conduction pumps embedded at the inlet. After establishing a foundational understanding of the influence of EHD conduction-driven flow distribution control on the system’s thermal performance under uneven heating loads at steady-state, an emphasis is placed on transient processes that govern the system’s response to sudden changes in operating conditions, as commonly encountered during flow distribution adjustment, going beyond the earlier work in this domain. Results show that transient charge accumulation takes 5.5 ms to reach a steady-state magnitude of (0.000826–0.000829) C/m, depending on the applied potential. This time is in the order of charge relaxation time. Additionally, the novel effect of pulsation on EHD-driven flow distribution control is investigated in detail to fundamentally understand its impact through the transient formation of the heterocharge layers. The contribution of pulsation is shown to depend strongly on the operating conditions and the applied frequency.</div></div>","PeriodicalId":336,"journal":{"name":"International Journal of Heat and Mass Transfer","volume":"260 ","pages":"Article 128445"},"PeriodicalIF":5.8,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146075863","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

Effect of sintering temperature on the microstructure, composition, and heat transfer of the GaN/diamond interface 烧结温度对氮化镓/金刚石界面微观结构、成分和传热的影响

IF 5.8 2区工程技术 Q1 ENGINEERING, MECHANICAL

International Journal of Heat and Mass Transfer

Pub Date : 2026-01-29 DOI: 10.1016/j.ijheatmasstransfer.2026.128447

Weishuai Sun , Xiaoqin Guo , Yabo Huang , Shengji Liu , Yuehan Yue , Xulei Wang , Shuai Tian , Xin Jia , Kang An , Zhengyi Zhang , Zhen Liu , Lin Gao

Gallium nitride (GaN) and a chemical vapour deposition (CVD) diamond were bonded using a semisolid nano-silver paste under low-temperature, pressure-assisted conditions. The effect of sintering temperature on the microstructure and mechanical properties of the GaN/diamond interface was investigated. The results showed that the solder morphology gradually evolved from point-like to sheet-like as the sintering temperature increased. Scanning electron microscopy observations of the GaN/diamond interface revealed that the thickness of the Ag sintering layer was approximately 40 μm. Moreover, elemental analysis of the GaN/diamond side of the sintered interface indicated a diffusion tendency of Ag. A well-defined planar interface at the nanoscale between Ag and GaN was further observed through transmission electron microscopy. The X-ray diffraction peak intensity of the Ag (111) crystal plane increased with increasing sintering temperature, while the full width at half maximum of the Ag (111) diffraction peak decreased, indicating improved nano-silver crystal quality at higher temperatures. The highest shear strength, 12.5 MPa, was obtained for samples sintered at 250°C. Interfacial solder uniformity was evaluated by monitoring the surface temperature of the heated GaN/diamond assembly, with a minimum average temperature difference of only 1.38 °C. The maximum thermal conductivity of the GaN/diamond structure reached 156.4 W/(m·K) at 250 °C. This study provides valuable insights into the heat dissipation of gallium nitride devices.

在低温、压力辅助条件下，采用半固态纳米银浆料将氮化镓（GaN）与化学气相沉积（CVD）金刚石结合。研究了烧结温度对氮化镓/金刚石界面微观结构和力学性能的影响。结果表明：随着烧结温度的升高，钎料形貌由点状逐渐演变为片状；扫描电镜观察发现，Ag烧结层厚度约为40 μm。此外，对烧结界面的氮化镓/金刚石侧进行元素分析，发现Ag有扩散趋势。通过透射电镜进一步观察到银和氮化镓之间在纳米尺度上有一个明确的平面界面。随着烧结温度的升高，Ag（111）晶面x射线衍射峰强度增大，而Ag（111）衍射峰半峰全宽减小，表明在较高温度下纳米银晶体质量得到改善。在250℃烧结时，试样的抗剪强度最高，为12.5 MPa。通过监测加热的GaN/金刚石组件的表面温度来评估界面焊料均匀性，最小平均温差仅为1.38°C。在250℃时，GaN/金刚石结构的最大导热系数达到156.4 W/（m·K）。本研究为氮化镓器件的散热提供了有价值的见解。

{"title":"Effect of sintering temperature on the microstructure, composition, and heat transfer of the GaN/diamond interface","authors":"Weishuai Sun , Xiaoqin Guo , Yabo Huang , Shengji Liu , Yuehan Yue , Xulei Wang , Shuai Tian , Xin Jia , Kang An , Zhengyi Zhang , Zhen Liu , Lin Gao","doi":"10.1016/j.ijheatmasstransfer.2026.128447","DOIUrl":"10.1016/j.ijheatmasstransfer.2026.128447","url":null,"abstract":"<div><div>Gallium nitride (GaN) and a chemical vapour deposition (CVD) diamond were bonded using a semisolid nano-silver paste under low-temperature, pressure-assisted conditions. The effect of sintering temperature on the microstructure and mechanical properties of the GaN/diamond interface was investigated. The results showed that the solder morphology gradually evolved from point-like to sheet-like as the sintering temperature increased. Scanning electron microscopy observations of the GaN/diamond interface revealed that the thickness of the Ag sintering layer was approximately 40 μm. Moreover, elemental analysis of the GaN/diamond side of the sintered interface indicated a diffusion tendency of Ag. A well-defined planar interface at the nanoscale between Ag and GaN was further observed through transmission electron microscopy. The X-ray diffraction peak intensity of the Ag (111) crystal plane increased with increasing sintering temperature, while the full width at half maximum of the Ag (111) diffraction peak decreased, indicating improved nano-silver crystal quality at higher temperatures. The highest shear strength, 12.5 MPa, was obtained for samples sintered at 250°C. Interfacial solder uniformity was evaluated by monitoring the surface temperature of the heated GaN/diamond assembly, with a minimum average temperature difference of only 1.38 °C. The maximum thermal conductivity of the GaN/diamond structure reached 156.4 W/(m·K) at 250 °C. This study provides valuable insights into the heat dissipation of gallium nitride devices.</div></div>","PeriodicalId":336,"journal":{"name":"International Journal of Heat and Mass Transfer","volume":"260 ","pages":"Article 128447"},"PeriodicalIF":5.8,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146075864","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

Numerical and experimental studies on an improved simulated annealing algorithm for inverse heat flux identification in transpiration cooling thermal protection systems 一种改进的模拟退火算法在蒸腾冷却热保护系统中反热流密度识别的数值与实验研究

IF 5.8 2区工程技术 Q1 ENGINEERING, MECHANICAL

International Journal of Heat and Mass Transfer

Pub Date : 2026-01-29 DOI: 10.1016/j.ijheatmasstransfer.2026.128404

Hang Xu , Ming-Jia Li , Xiao-Feng Yang , Yan-Xia Du , Lan-Sen Bi

Accurate identification of heat flux in transpiration cooling thermal protection systems (TPS) is critical for precise temperature control and optimization of cooling efficiency during active control processes. However, identifying surface heat flux in such systems poses more significant challenges than conventional inverse heat conduction problems (IHCP), primarily due to the complex interactions among high-Mach-number flows, compressible outflow, and microscale percolation effects in a porous medium. In this paper, the inverse heat infiltration coupling problem (IHICPP), which focuses on identifying surface heat flux in porous transpiration cooling TPS, is analyzed in detail. To transform HICPP into an optimization problem, a local thermal non-equilibrium model is employed to simulate the direct problem. The objective function is formulated using the least-squares method. An improved simulated annealing algorithm (ISAA) is proposed, which leverages gradient iteration within the simulated annealing framework to enhance identification accuracy and computational efficiency. The ISAA demonstrates merit performance across various surface heat flux waveforms and TPS thicknesses, achieving stable and reliable results. This study serves as a valuable reference for active control and optimization of TPS, offering a robust methodology for IHICPP in high-Mach-number aircraft.

在主动控制过程中，准确识别蒸腾冷却热保护系统（TPS）的热流密度对精确温度控制和优化冷却效率至关重要。然而，与传统的反热传导问题（IHCP）相比，识别此类系统中的表面热通量面临着更大的挑战，这主要是由于多孔介质中高马赫数流动、可压缩流出和微尺度渗透效应之间复杂的相互作用。本文详细分析了多孔蒸腾冷却TPS中热渗透反耦合问题（IHICPP），该问题的重点是识别表面热流密度。为了将HICPP问题转化为优化问题，采用局部热非平衡模型对直接问题进行模拟。目标函数用最小二乘法表示。提出了一种改进的模拟退火算法（ISAA），该算法利用模拟退火框架内的梯度迭代来提高识别精度和计算效率。ISAA在各种表面热通量波形和TPS厚度下都表现出良好的性能，取得了稳定可靠的结果。该研究为TPS的主动控制和优化提供了有价值的参考，为高马赫数飞机的IHICPP提供了可靠的方法。

{"title":"Numerical and experimental studies on an improved simulated annealing algorithm for inverse heat flux identification in transpiration cooling thermal protection systems","authors":"Hang Xu , Ming-Jia Li , Xiao-Feng Yang , Yan-Xia Du , Lan-Sen Bi","doi":"10.1016/j.ijheatmasstransfer.2026.128404","DOIUrl":"10.1016/j.ijheatmasstransfer.2026.128404","url":null,"abstract":"<div><div>Accurate identification of heat flux in transpiration cooling thermal protection systems (TPS) is critical for precise temperature control and optimization of cooling efficiency during active control processes. However, identifying surface heat flux in such systems poses more significant challenges than conventional inverse heat conduction problems (IHCP), primarily due to the complex interactions among high-Mach-number flows, compressible outflow, and microscale percolation effects in a porous medium. In this paper, the inverse heat infiltration coupling problem (IHICPP), which focuses on identifying surface heat flux in porous transpiration cooling TPS, is analyzed in detail. To transform HICPP into an optimization problem, a local thermal non-equilibrium model is employed to simulate the direct problem. The objective function is formulated using the least-squares method. An improved simulated annealing algorithm (ISAA) is proposed, which leverages gradient iteration within the simulated annealing framework to enhance identification accuracy and computational efficiency. The ISAA demonstrates merit performance across various surface heat flux waveforms and TPS thicknesses, achieving stable and reliable results. This study serves as a valuable reference for active control and optimization of TPS, offering a robust methodology for IHICPP in high-Mach-number aircraft.</div></div>","PeriodicalId":336,"journal":{"name":"International Journal of Heat and Mass Transfer","volume":"260 ","pages":"Article 128404"},"PeriodicalIF":5.8,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146076287","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

Parameter inversion of the concrete hydration heat source based on physics-informed neural network 基于物理信息神经网络的混凝土水化热源参数反演

IF 5.8 2区工程技术 Q1 ENGINEERING, MECHANICAL

International Journal of Heat and Mass Transfer

Pub Date : 2026-01-28 DOI: 10.1016/j.ijheatmasstransfer.2026.128409

Yahao Bu , Binghan Xue , Zhenhua Huang , Musong Yang , Zehan Zhang , Cuiying Zheng

Accurately characterizing the exothermic process of concrete hydration requires the proper determination of the hydration heat source function and its parameters. Conventional parameter identification methods require long-term experimental data and rely on data fitting or repeated calls to forward models for inversion, resulting in low efficiency and high costs. Within this study, a physics-informed neural network (PINN) framework is applied to parameter inversion of the concrete hydration heat source. In this framework, the residual of the heat conduction governing equation is computed through automatic differentiation and incorporated into the loss function as a physical constraint. By combining this physical constraint with observational data, a composite loss function is constructed, thereby enabling the simultaneous solution of forward modeling and parameter inversion within a unified framework. Using only the first three days of adiabatic temperature rise (ATR) data from different conventional concrete mix ratios, the proposed framework achieves robust inversion of the key parameters a, b, and the final ATR θ₀. The resulting temperature rise curves closely match the 28-day experimental data (R² > 0.99), demonstrating superior performance over the genetic algorithm (GA) and the Levenberg-Marquardt (LM) methods. These results demonstrate that the PINN-based framework can reduce reliance on long-term experiments while maintaining high accuracy and robustness, helping address the time-consuming and costly nature of ATR experiments, providing a practical and efficient method for rapid inversion of parameters of the concrete hydration heat source.

准确表征混凝土水化放热过程，需要正确确定水化热源函数及其参数。传统的参数辨识方法需要长期的实验数据，依赖数据拟合或反复调用正演模型进行反演，效率低，成本高。本研究将物理信息神经网络（PINN）框架应用于混凝土水化热源参数反演。在这个框架中，热传导控制方程的残差通过自动微分计算，并作为物理约束纳入损失函数。将这一物理约束与观测数据相结合，构建复合损失函数，实现了正演模拟与参数反演在统一框架内的同时求解。仅使用来自不同常规混凝土配合比的前三天绝热温升（ATR）数据，所提出的框架实现了关键参数a， b和最终ATR θ0的鲁棒反演。所得的温升曲线与28天的实验数据非常接近（R2 > 0.99），表现出优于遗传算法（GA）和Levenberg-Marquardt （LM）方法的性能。这些结果表明，基于pat的框架可以减少对长期实验的依赖，同时保持较高的准确性和鲁棒性，有助于解决ATR实验耗时和成本高的特点，为混凝土水化热源参数的快速反演提供了一种实用高效的方法。

{"title":"Parameter inversion of the concrete hydration heat source based on physics-informed neural network","authors":"Yahao Bu , Binghan Xue , Zhenhua Huang , Musong Yang , Zehan Zhang , Cuiying Zheng","doi":"10.1016/j.ijheatmasstransfer.2026.128409","DOIUrl":"10.1016/j.ijheatmasstransfer.2026.128409","url":null,"abstract":"<div><div>Accurately characterizing the exothermic process of concrete hydration requires the proper determination of the hydration heat source function and its parameters. Conventional parameter identification methods require long-term experimental data and rely on data fitting or repeated calls to forward models for inversion, resulting in low efficiency and high costs. Within this study, a physics-informed neural network (PINN) framework is applied to parameter inversion of the concrete hydration heat source. In this framework, the residual of the heat conduction governing equation is computed through automatic differentiation and incorporated into the loss function as a physical constraint. By combining this physical constraint with observational data, a composite loss function is constructed, thereby enabling the simultaneous solution of forward modeling and parameter inversion within a unified framework. Using only the first three days of adiabatic temperature rise (ATR) data from different conventional concrete mix ratios, the proposed framework achieves robust inversion of the key parameters <em>a, b</em>, and the final ATR <em>θ</em><sub>0</sub>. The resulting temperature rise curves closely match the 28-day experimental data (R<sup>2</sup> > 0.99), demonstrating superior performance over the genetic algorithm (GA) and the Levenberg-Marquardt (LM) methods. These results demonstrate that the PINN-based framework can reduce reliance on long-term experiments while maintaining high accuracy and robustness, helping address the time-consuming and costly nature of ATR experiments, providing a practical and efficient method for rapid inversion of parameters of the concrete hydration heat source.</div></div>","PeriodicalId":336,"journal":{"name":"International Journal of Heat and Mass Transfer","volume":"260 ","pages":"Article 128409"},"PeriodicalIF":5.8,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146075867","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