Pub Date : 2026-03-01Epub Date: 2025-12-31DOI: 10.1016/j.jnnfm.2025.105548
Pierre Saramito
In a few lines, the Oldroyd-B, FENE-P, Giesekus and FENE-CR models are shown as satisfying the second principle of thermodynamics. In addition, entropy estimates (a priori bounds) are easily obtained, together with explicit expressions for the dissipation. For the Giesekus and FENE-CR models, these estimates are new, while for the Oldroyd-B and FENE-P, there were already established. In all cases, they are obtained her e in a clear an concise manner, instead of long derivations. This approach could also be applied to the development of new constitutive equations, and some preliminary explorations are provided. The conformation tensor is identified in a purely kinematic context, in terms of the Cauchy–Green tensor. Consequently, the formulation in terms of the logarithm of conformation tensor is reinterpreted in terms of Hencky strain and its logarithmic corotational derivative. While useful for numerical computations, this also leads to much more concise and understandable formulations, but above all, it opens up new avenues for theoretical developments. This paper presents new developments of a work initiated by the author in a recent book (Springer, 2024), which is also reviewed here in a concise manner. We briefly recall how the standard generalized materials framework extends to large-strains kinematics in Eulerian frame.
{"title":"Yet another thermodynamic environment","authors":"Pierre Saramito","doi":"10.1016/j.jnnfm.2025.105548","DOIUrl":"10.1016/j.jnnfm.2025.105548","url":null,"abstract":"<div><div>In a few lines, the Oldroyd-B, FENE-P, Giesekus and FENE-CR models are shown as satisfying the second principle of thermodynamics. In addition, entropy estimates (<em>a priori</em> bounds) are easily obtained, together with explicit expressions for the dissipation. For the Giesekus and FENE-CR models, these estimates are new, while for the Oldroyd-B and FENE-P, there were already established. In all cases, they are obtained her e in a clear an concise manner, instead of long derivations. This approach could also be applied to the development of new constitutive equations, and some preliminary explorations are provided. The conformation tensor is identified in a purely kinematic context, in terms of the Cauchy–Green tensor. Consequently, the formulation in terms of the logarithm of conformation tensor is reinterpreted in terms of Hencky strain and its logarithmic corotational derivative. While useful for numerical computations, this also leads to much more concise and understandable formulations, but above all, it opens up new avenues for theoretical developments. This paper presents new developments of a work initiated by the author in a recent book (Springer, 2024), which is also reviewed here in a concise manner. We briefly recall how the standard generalized materials framework extends to large-strains kinematics in Eulerian frame.</div></div>","PeriodicalId":54782,"journal":{"name":"Journal of Non-Newtonian Fluid Mechanics","volume":"348 ","pages":"Article 105548"},"PeriodicalIF":2.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145977791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-02DOI: 10.1016/j.jnnfm.2025.105559
Joseph V. Giliberto, Olivier Desjardins
Viscoelastic constitutive equations often model the elastic stress field through the use of an elastic dumbbell model that utilizes a conformation tensor to represent the average polymer configuration in the flow field. In a liquid–gas flow environment, the conformation tensor is a discontinuous quantity that only exists in the liquid phase. This discontinuity often presents numerical challenges that can be tackled through the use of very fine meshes at the interface to ensure the stress profile is accurately captured. In contrast, this work presents a hybrid advection scheme for the discontinuous conformation tensor field that uses a semi-Lagrangian geometric flux-based scheme in the direct vicinity of the liquid–gas interface and a MUSCL scheme in the bulk of the liquid, away from the interface. This hybrid method is found to be exactly conservative and bounded, and prevents any leakage of data across the liquid–gas interface. Verification and validation of this approach is done using the case of a gas bubble rising in a viscoelastic liquid. Results of the convergence study show that the hybrid scheme is able to converge to experimental results with 32 cells across the initial diameter of the bubble, which is one-third the resolution used in other computational studies comparing against experiments. The hybrid advection scheme is then applied to the case of a viscoelastic droplet deforming in homogeneous isotropic turbulence to investigate the influence of elastic stresses on droplet morphology. Results indicate that increasing viscoelastic stresses within the droplet significantly alters its deformation dynamics. At the moderate elastic stress levels tested, the droplet forms elongated liquid filaments delaying break-up for a longer duration. As viscoelasticity is further increased, deformation is progressively suppressed, ultimately stabilizing the droplet’s shape and preventing fragmentation.
{"title":"A sharp computational method for simulating multiphase viscoelastic flows","authors":"Joseph V. Giliberto, Olivier Desjardins","doi":"10.1016/j.jnnfm.2025.105559","DOIUrl":"10.1016/j.jnnfm.2025.105559","url":null,"abstract":"<div><div>Viscoelastic constitutive equations often model the elastic stress field through the use of an elastic dumbbell model that utilizes a conformation tensor to represent the average polymer configuration in the flow field. In a liquid–gas flow environment, the conformation tensor is a discontinuous quantity that only exists in the liquid phase. This discontinuity often presents numerical challenges that can be tackled through the use of very fine meshes at the interface to ensure the stress profile is accurately captured. In contrast, this work presents a hybrid advection scheme for the discontinuous conformation tensor field that uses a semi-Lagrangian geometric flux-based scheme in the direct vicinity of the liquid–gas interface and a MUSCL scheme in the bulk of the liquid, away from the interface. This hybrid method is found to be exactly conservative and bounded, and prevents any leakage of data across the liquid–gas interface. Verification and validation of this approach is done using the case of a gas bubble rising in a viscoelastic liquid. Results of the convergence study show that the hybrid scheme is able to converge to experimental results with 32 cells across the initial diameter of the bubble, which is one-third the resolution used in other computational studies comparing against experiments. The hybrid advection scheme is then applied to the case of a viscoelastic droplet deforming in homogeneous isotropic turbulence to investigate the influence of elastic stresses on droplet morphology. Results indicate that increasing viscoelastic stresses within the droplet significantly alters its deformation dynamics. At the moderate elastic stress levels tested, the droplet forms elongated liquid filaments delaying break-up for a longer duration. As viscoelasticity is further increased, deformation is progressively suppressed, ultimately stabilizing the droplet’s shape and preventing fragmentation.</div></div>","PeriodicalId":54782,"journal":{"name":"Journal of Non-Newtonian Fluid Mechanics","volume":"348 ","pages":"Article 105559"},"PeriodicalIF":2.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145925955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2025-12-04DOI: 10.1016/j.jnnfm.2025.105534
Rishi Kumar , K. Muralidhar , Indranil Saha Dalal
This study investigates the effects of geometric model reduction on blood flow simulations in the patient-specific descending aorta, followed by speed-accuracy trade-off analysis using 3D simulations. We demonstrate how wall shear stresses (WSS) can be reliably estimated for such realistic arteries using significantly faster simulations of highly idealized equivalent geometries, for any blood rheology model. CFD simulations (3D) are performed at two levels of geometry reduction employing realistic pulsatile inflow and pressure outlet boundary conditions and utilizing both Newtonian and non-Newtonian blood rheology models, including the one developed recently by Apostolidis and Beris. The first level of reduction does not retain effects due to local asymmetry but can approximate various flow parameters and patterns, while showing a significant computational speedup. However, further simplification to an idealized smooth geometry loses all information about the vortex structures and flow circulation. The non-Newtonian models retain more accuracy than the Newtonian models in geometry reductions, as quantified by correlations defined in this study. The idealized smooth geometry, combined with area correction, yields WSS estimates that closely approximate those of the actual artery. This study is expected to be applicable in geometric reductions (and speed enhancements) for more complex patient-specific 3D simulations while maintaining accuracy.
{"title":"Effects of geometric modeling and blood rheology in patient-specific arterial blood flow simulations with speed-accuracy trade-off analysis","authors":"Rishi Kumar , K. Muralidhar , Indranil Saha Dalal","doi":"10.1016/j.jnnfm.2025.105534","DOIUrl":"10.1016/j.jnnfm.2025.105534","url":null,"abstract":"<div><div>This study investigates the effects of geometric model reduction on blood flow simulations in the patient-specific descending aorta, followed by speed-accuracy trade-off analysis using 3D simulations. We demonstrate how wall shear stresses (WSS) can be reliably estimated for such realistic arteries using significantly faster simulations of highly idealized equivalent geometries, for any blood rheology model. CFD simulations (3D) are performed at two levels of geometry reduction employing realistic pulsatile inflow and pressure outlet boundary conditions and utilizing both Newtonian and non-Newtonian blood rheology models, including the one developed recently by Apostolidis and Beris. The first level of reduction does not retain effects due to local asymmetry but can approximate various flow parameters and patterns, while showing a significant computational speedup. However, further simplification to an idealized smooth geometry loses all information about the vortex structures and flow circulation. The non-Newtonian models retain more accuracy than the Newtonian models in geometry reductions, as quantified by correlations defined in this study. The idealized smooth geometry, combined with area correction, yields WSS estimates that closely approximate those of the actual artery. This study is expected to be applicable in geometric reductions (and speed enhancements) for more complex patient-specific 3D simulations while maintaining accuracy.</div></div>","PeriodicalId":54782,"journal":{"name":"Journal of Non-Newtonian Fluid Mechanics","volume":"348 ","pages":"Article 105534"},"PeriodicalIF":2.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145665694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-02-14DOI: 10.1016/j.jnnfm.2026.105567
Hoang Minh Khoa Nguyen, Dong-Wook Oh
This study presents a numerical analysis of predicting fiber alignment in fused filament fabrication (FFF) under varying nozzle geometries and deposition conditions. Computational fluid dynamics simulations were performed using a coupled level-set and volume-of-fluid approach to resolve the melt flow containing carbon fibers inside the nozzle and after extrusion. Fiber alignment was calculated through the orientation tensors with a quadratic closure approximation. Two nozzle designs, a straight channel nozzle (SCN) and an orifice-embedded nozzle (OEN), were compared across different deposition gap cases. The results reveal that nozzle geometry and deposition conditions strongly affect the local flow kinematics and, consequently, the distribution of fiber orientations after extrusion. In the SCN, regardless of the gap between the nozzle tip and the deposition bed, squeeze-induced shear resulted in most fibers aligning parallel to the flow and extrusion direction. In contrast, the OEN produced more substantial perpendicular alignment across the filament owing to orifice-induced extensional flow. Furthermore, the model numerically reproduced key phenomena observed in OEN flow visualization experiments, including the formation of a characteristic 'M'-shaped perpendicular alignment profile and the subsequent disappearance of its lower peak as the deposition gap decreases. These findings offer guidelines for tailoring anisotropic properties in fiber-reinforced printed parts.
{"title":"Numerical analysis on the effect of the orifice embedded nozzle and the squeeze flow on fiber alignment in the fused filament fabrication","authors":"Hoang Minh Khoa Nguyen, Dong-Wook Oh","doi":"10.1016/j.jnnfm.2026.105567","DOIUrl":"10.1016/j.jnnfm.2026.105567","url":null,"abstract":"<div><div>This study presents a numerical analysis of predicting fiber alignment in fused filament fabrication (FFF) under varying nozzle geometries and deposition conditions. Computational fluid dynamics simulations were performed using a coupled level-set and volume-of-fluid approach to resolve the melt flow containing carbon fibers inside the nozzle and after extrusion. Fiber alignment was calculated through the orientation tensors with a quadratic closure approximation. Two nozzle designs, a straight channel nozzle (SCN) and an orifice-embedded nozzle (OEN), were compared across different deposition gap cases. The results reveal that nozzle geometry and deposition conditions strongly affect the local flow kinematics and, consequently, the distribution of fiber orientations after extrusion. In the SCN, regardless of the gap between the nozzle tip and the deposition bed, squeeze-induced shear resulted in most fibers aligning parallel to the flow and extrusion direction. In contrast, the OEN produced more substantial perpendicular alignment across the filament owing to orifice-induced extensional flow. Furthermore, the model numerically reproduced key phenomena observed in OEN flow visualization experiments, including the formation of a characteristic 'M'-shaped perpendicular alignment profile and the subsequent disappearance of its lower peak as the deposition gap decreases. These findings offer guidelines for tailoring anisotropic properties in fiber-reinforced printed parts.</div></div>","PeriodicalId":54782,"journal":{"name":"Journal of Non-Newtonian Fluid Mechanics","volume":"348 ","pages":"Article 105567"},"PeriodicalIF":2.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147395606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-29DOI: 10.1016/j.jnnfm.2026.105562
Taha Rezaee
Localized wall slip can reorganize viscoelastic flows far beyond the immediate vicinity of the boundary, yet the physical mechanisms governing this spatial response remain unclear. In this study, we investigate steady annular flow of a viscoelastic fluid described by the exponential Phan–Thien–Tanner (EPTT) model, focusing on how a spatially localized modification of the slip length influences the bulk flow field. The resulting changes in polymeric shear stress, axial velocity, and first normal stress difference are quantified using root-mean-square (RMS) response measures that characterize the intensity, spatial extent, and centroid of the slip-induced disturbance.
Flow-type maps reveal a distinctive kinematic signature associated with the slip-induced response. A persistent extensional band develops in the mid-gap region downstream of the slip perturbation and is consistently flanked by rotation-dominated patches. Within the extensional band, polymers undergo enhanced stretching, whereas the adjacent rotational regions locally suppress stress production. This spatial arrangement gives rise to a downstream-curved plateau in the first normal stress difference, which sets both the shape and the reach of the disturbance in the bulk flow. The repeated appearance of this extension–rotation pattern across all cases indicates that it constitutes the kinematic origin of the observed long-range stress response.
{"title":"Slip-induced kinematic structuring and stress propagation in viscoelastic annular flow","authors":"Taha Rezaee","doi":"10.1016/j.jnnfm.2026.105562","DOIUrl":"10.1016/j.jnnfm.2026.105562","url":null,"abstract":"<div><div>Localized wall slip can reorganize viscoelastic flows far beyond the immediate vicinity of the boundary, yet the physical mechanisms governing this spatial response remain unclear. In this study, we investigate steady annular flow of a viscoelastic fluid described by the exponential Phan–Thien–Tanner (EPTT) model, focusing on how a spatially localized modification of the slip length influences the bulk flow field. The resulting changes in polymeric shear stress, axial velocity, and first normal stress difference are quantified using root-mean-square (RMS) response measures that characterize the intensity, spatial extent, and centroid of the slip-induced disturbance.</div><div>Flow-type maps reveal a distinctive kinematic signature associated with the slip-induced response. A persistent extensional band develops in the mid-gap region downstream of the slip perturbation and is consistently flanked by rotation-dominated patches. Within the extensional band, polymers undergo enhanced stretching, whereas the adjacent rotational regions locally suppress stress production. This spatial arrangement gives rise to a downstream-curved plateau in the first normal stress difference, which sets both the shape and the reach of the disturbance in the bulk flow. The repeated appearance of this extension–rotation pattern across all cases indicates that it constitutes the kinematic origin of the observed long-range stress response.</div></div>","PeriodicalId":54782,"journal":{"name":"Journal of Non-Newtonian Fluid Mechanics","volume":"348 ","pages":"Article 105562"},"PeriodicalIF":2.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147395720","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}
In recent years, ion-selective membranes and membrane-based separation technologies have garnered significant attention due to their increasing integration in various industries, including energy storage and electrolyzer applications, which enable chemical extraction and/or separation via relevant phenomena such as electrodialysis, desalination, flow electrodes, capacitive deionization, and redox-flow battery systems. The interaction between the membrane surface and the electro-rheological (ER) properties of fluid modulates the inherent ion transport dynamics. The induced electric current subsequently alters the flow field, thereby either enhancing or inhibiting the overall separation efficiency, depending on the applied electric field strength. Additionally, given the non-uniform ionic concentration distribution near the membrane surface, electro-convective currents ultimately lead to a net over-limiting current, followed by a relative suppression of advective ion transport. Such irregular loading and unloading cycles may lead to excessive ion accumulation on electrode surfaces, accelerating dendrite formation, which in turn degrades electrode performance and compromises membrane integrity. Therefore, the present study investigates the role of shear-thinning electrolytes in mitigating electroconvection near ion-selective membranes. A computational model is employed to solve the coupled Poisson-Nernst–Planck equation and the momentum equations, which leads to the evolution of ion distribution profiles and electrokinetic flow instabilities. The extensive numerical simulations yielded the flow attributes in terms of instantaneous velocity, concentration contours, streamlines, ionic current density, and average kinetic energy. In contrast, prolonged chaotic convection facilitates a more uniform distribution of ions within the electrolyte. The enhanced shear thinning effect sharpens both velocity and ionic concentration gradients adjacent to the membrane surface, thereby increasing ionic flux. In general, shear-thinning electrolytes present a promising strategy for mitigating dendrite formation, ultimately improving the operational stability and longevity of electrochemical devices.
{"title":"Influence of shear-thinning rheology on electroconvection around ion-selective membrane","authors":"Saurabh Maurya , Mohit Trivedi , Neelkanth Nirmalkar","doi":"10.1016/j.jnnfm.2025.105545","DOIUrl":"10.1016/j.jnnfm.2025.105545","url":null,"abstract":"<div><div>In recent years, ion-selective membranes and membrane-based separation technologies have garnered significant attention due to their increasing integration in various industries, including energy storage and electrolyzer applications, which enable chemical extraction and/or separation via relevant phenomena such as electrodialysis, desalination, flow electrodes, capacitive deionization, and redox-flow battery systems. The interaction between the membrane surface and the electro-rheological (ER) properties of fluid modulates the inherent ion transport dynamics. The induced electric current subsequently alters the flow field, thereby either enhancing or inhibiting the overall separation efficiency, depending on the applied electric field strength. Additionally, given the non-uniform ionic concentration distribution near the membrane surface, electro-convective currents ultimately lead to a net over-limiting current, followed by a relative suppression of advective ion transport. Such irregular loading and unloading cycles may lead to excessive ion accumulation on electrode surfaces, accelerating dendrite formation, which in turn degrades electrode performance and compromises membrane integrity. Therefore, the present study investigates the role of shear-thinning electrolytes in mitigating electroconvection near ion-selective membranes. A computational model is employed to solve the coupled Poisson-Nernst–Planck equation and the momentum equations, which leads to the evolution of ion distribution profiles and electrokinetic flow instabilities. The extensive numerical simulations yielded the flow attributes in terms of instantaneous velocity, concentration contours, streamlines, ionic current density, and average kinetic energy. In contrast, prolonged chaotic convection facilitates a more uniform distribution of ions within the electrolyte. The enhanced shear thinning effect sharpens both velocity and ionic concentration gradients adjacent to the membrane surface, thereby increasing ionic flux. In general, shear-thinning electrolytes present a promising strategy for mitigating dendrite formation, ultimately improving the operational stability and longevity of electrochemical devices.</div></div>","PeriodicalId":54782,"journal":{"name":"Journal of Non-Newtonian Fluid Mechanics","volume":"348 ","pages":"Article 105545"},"PeriodicalIF":2.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-03-04DOI: 10.1016/j.jnnfm.2026.105573
Hongfei Xie , Ying Meng , Kang Luo , Hongliang Yi
Viscoelastic thermal convection turbulence is an important branch of viscoelastic Rayleigh–Bérnard convection (VRBC). It is well established that the aspect ratio (AR) has a significant influence on turbulent RBC; however, its impact on turbulent VRBC remains poorly understood, particularly at large AR. In this study, large-eddy simulation (LES) based on the finite volume method (FVM) is employed to investigate turbulent VRBC over a range of aspect ratios. Three eddy-viscosity models are implemented and systematically evaluated against DNS results to assess their performance. Using the validated LES framework, turbulent VRBC at different aspect ratios is further investigated. The present results reveal a strong dependence on the Weissenberg number () at large aspect ratios. Specifically, Heat transfer enhancement(HTE) is observed at low elasticity ( 1.0), whereas a transition to heat transfer reduction (HTR) occurs at higher elasticity ( 1.0). HTE is closely associated with flow structure modulation: at low elasticity, the number of large-scale structures in the flow increases, which enhances the overall heat transfer efficiency. In contrast, HTR arises from elasticity-induced energy transfer from the mean flow to turbulent fluctuations, which weakens the mean circulation and consequently reduces the heat transfer efficiency.
{"title":"Application of Large Eddy Simulation models to Viscoelastic thermal convection turbulence","authors":"Hongfei Xie , Ying Meng , Kang Luo , Hongliang Yi","doi":"10.1016/j.jnnfm.2026.105573","DOIUrl":"10.1016/j.jnnfm.2026.105573","url":null,"abstract":"<div><div>Viscoelastic thermal convection turbulence is an important branch of viscoelastic Rayleigh–Bérnard convection (VRBC). It is well established that the aspect ratio (AR) has a significant influence on turbulent RBC; however, its impact on turbulent VRBC remains poorly understood, particularly at large AR. In this study, large-eddy simulation (LES) based on the finite volume method (FVM) is employed to investigate turbulent VRBC over a range of aspect ratios. Three eddy-viscosity models are implemented and systematically evaluated against DNS results to assess their performance. Using the validated LES framework, turbulent VRBC at different aspect ratios is further investigated. The present results reveal a strong dependence on the Weissenberg number (<span><math><mrow><mi>W</mi><mi>i</mi></mrow></math></span>) at large aspect ratios. Specifically, Heat transfer enhancement(HTE) is observed at low elasticity (<span><math><mrow><mi>W</mi><mi>i</mi><mo><</mo></mrow></math></span> 1.0), whereas a transition to heat transfer reduction (HTR) occurs at higher elasticity (<span><math><mrow><mi>W</mi><mi>i</mi><mo>></mo></mrow></math></span> 1.0). HTE is closely associated with flow structure modulation: at low elasticity, the number of large-scale structures in the flow increases, which enhances the overall heat transfer efficiency. In contrast, HTR arises from elasticity-induced energy transfer from the mean flow to turbulent fluctuations, which weakens the mean circulation and consequently reduces the heat transfer efficiency.</div></div>","PeriodicalId":54782,"journal":{"name":"Journal of Non-Newtonian Fluid Mechanics","volume":"348 ","pages":"Article 105573"},"PeriodicalIF":2.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147395608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-02-23DOI: 10.1016/j.jnnfm.2026.105568
Bernardo P. Brener , Matheus S.S. Macedo , Anselmo Pereira , Roney L. Thompson
Viscoelastic turbulence (e.g. the inertially-induced turbulence affected by the presence of polymers), is a central problem in Fluid Mechanics, combining the nonlinear complexity of turbulence with the rheology of polymer solutions. While Direct Numerical Simulations (DNS) have revealed important physical mechanisms, their high computational cost and the lack of robust Reynolds-averaged closures remain significant challenges. In this work, we introduce a novel machine learning framework that augments low-cost, low-fidelity Reynolds-averaged Navier–Stokes simulations (RANS) of Newtonian turbulent channel flows to predict high-fidelity results, as provided by Direct Numerical Simulations (DNS) of viscoelastic turbulent channel flows. A three-step data-driven strategy is conducted: (i) augmenting Newtonian DNS into viscoelastic DNS closures, (ii) enhancing low-cost RANS into Newtonian DNS accuracy, and (iii) predicting viscoelastic DNS-like outcomes directly from Newtonian RANS through deep neural networks. To bypass the ill-conditioning of the RANS equations and non-negligible errors in the Reynolds stress tensor provided by DNS, we propose to provide the closure with the sum of the polymeric and turbulent stresses, allowing it to be computed indirectly using first-order statistics only. Results show that the proposed framework achieves accurate predictions across all three tasks, with the final step offering DNS-like accuracy at the cost of RANS simulations. These findings demonstrate the feasibility of machine learning–assisted viscoelastic turbulence modeling in accurately predicting drag-reducing flows.
{"title":"Neural-network closure modeling for viscoelastic drag-reducing channel flows","authors":"Bernardo P. Brener , Matheus S.S. Macedo , Anselmo Pereira , Roney L. Thompson","doi":"10.1016/j.jnnfm.2026.105568","DOIUrl":"10.1016/j.jnnfm.2026.105568","url":null,"abstract":"<div><div>Viscoelastic turbulence (e.g. the inertially-induced turbulence affected by the presence of polymers), is a central problem in Fluid Mechanics, combining the nonlinear complexity of turbulence with the rheology of polymer solutions. While Direct Numerical Simulations (DNS) have revealed important physical mechanisms, their high computational cost and the lack of robust Reynolds-averaged closures remain significant challenges. In this work, we introduce a novel machine learning framework that augments low-cost, low-fidelity Reynolds-averaged Navier–Stokes simulations (RANS) of Newtonian turbulent channel flows to predict high-fidelity results, as provided by Direct Numerical Simulations (DNS) of viscoelastic turbulent channel flows. A three-step data-driven strategy is conducted: (i) augmenting Newtonian DNS into viscoelastic DNS closures, (ii) enhancing low-cost RANS into Newtonian DNS accuracy, and (iii) predicting viscoelastic DNS-like outcomes directly from Newtonian RANS through deep neural networks. To bypass the ill-conditioning of the RANS equations and non-negligible errors in the Reynolds stress tensor provided by DNS, we propose to provide the closure with the sum of the polymeric and turbulent stresses, allowing it to be computed indirectly using first-order statistics only. Results show that the proposed framework achieves accurate predictions across all three tasks, with the final step offering DNS-like accuracy at the cost of RANS simulations. These findings demonstrate the feasibility of machine learning–assisted viscoelastic turbulence modeling in accurately predicting drag-reducing flows.</div></div>","PeriodicalId":54782,"journal":{"name":"Journal of Non-Newtonian Fluid Mechanics","volume":"348 ","pages":"Article 105568"},"PeriodicalIF":2.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147395610","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-02-03DOI: 10.1016/j.jnnfm.2026.105566
Alejandra M. Mil-Martínez , René O. Vargas , Aldo Gómez-López , Juan P. Escandón , Lorenzo Martínez-Suástegui , Timothy N. Phillips
This study presents multiscale simulations of complex fluids confined between parallel plates under simple shear flow, employing a finitely extensible nonlinear elastic (FENE) transient network model. The model integrates microscopic kinetic equations for microstructural breaking and recombination processes with macroscopic flow equations, enabling the prediction of nonlinear velocity and stress fluctuations. A hybrid micro–macro numerical framework is developed to capture the coupling between microstructural dynamics and macroscopic rheology. Numerical experiments explore the influence of kinetic rate constants, viscosity ratio, elasticity, extension length, and inertia on flow instabilities. The results reveal that considering viscosity as a function of microstructural kinetics induces fluctuations in the velocity field. These fluctuations occur when the rate of interaction between microstructures reaches a certain value. The fluctuations decrease when the system is dilute or elasticity is increased, and increase for short microstructural chain extensions and increasing inertia. These findings establish a direct connection between molecular-scale restructuring and macroscopic flow, thereby contributing to the fundamental understanding of flow instabilities and providing guidance for modelling complex fluids.
{"title":"Rheo-kinetic predictions using a finitely extensible nonlinear elastic transient network model, Part I: Simple shear flow","authors":"Alejandra M. Mil-Martínez , René O. Vargas , Aldo Gómez-López , Juan P. Escandón , Lorenzo Martínez-Suástegui , Timothy N. Phillips","doi":"10.1016/j.jnnfm.2026.105566","DOIUrl":"10.1016/j.jnnfm.2026.105566","url":null,"abstract":"<div><div>This study presents multiscale simulations of complex fluids confined between parallel plates under simple shear flow, employing a finitely extensible nonlinear elastic (FENE) transient network model. The model integrates microscopic kinetic equations for microstructural breaking and recombination processes with macroscopic flow equations, enabling the prediction of nonlinear velocity and stress fluctuations. A hybrid micro–macro numerical framework is developed to capture the coupling between microstructural dynamics and macroscopic rheology. Numerical experiments explore the influence of kinetic rate constants, viscosity ratio, elasticity, extension length, and inertia on flow instabilities. The results reveal that considering viscosity as a function of microstructural kinetics induces fluctuations in the velocity field. These fluctuations occur when the rate of interaction between microstructures reaches a certain value. The fluctuations decrease when the system is dilute or elasticity is increased, and increase for short microstructural chain extensions and increasing inertia. These findings establish a direct connection between molecular-scale restructuring and macroscopic flow, thereby contributing to the fundamental understanding of flow instabilities and providing guidance for modelling complex fluids.</div></div>","PeriodicalId":54782,"journal":{"name":"Journal of Non-Newtonian Fluid Mechanics","volume":"348 ","pages":"Article 105566"},"PeriodicalIF":2.8,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147395721","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号