Relational graph-driven differential denoising and diffusion attention fusion for multimodal conversational emotion recognition

The importance of differential diffusion in solid fuel/ammonia co-combustion with NOx formation

Comparative experimental study of flame–wall interaction for hydrogen and methane

A DNS study of differential diffusion effects on the ignition of a turbulent NH <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si208.svg" display="inline" id="d1e1506"> <mml:msub> <mml:mrow/> <mml:mrow> <mml:mn>3</mml:mn> </mml:mrow> </mml:msub> </mml:math> /H <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si260.svg" display="inline" id="d1e1514"> <mml:msub> <mml:mrow/> <mml:mrow> <mml:mn>2</mml:mn> </mml:mrow> </mml:msub> </mml:math> /N <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si260.svg" display="inline" id="d1e1522"> <mml:msub> <mml:mrow/> <mml:mrow> <mml:mn>2</mml:mn> </mml:mrow> </mml:msub> </mml:math> -air mixing layer

Hydrogen-fueled internal combustion engines (H 2 -ICEs) hold strong potential as a pathway toward CO 2 -neutral propulsion. To reduce emissions, H 2 -ICEs are usually operated under fuel-lean conditions, where the flames are prone to thermo-diffusive instabilities (TDIs). These TDIs govern both local and global flame propagation, but their impact on full-scale engine combustion remains an open question. In this study, high-fidelity three-dimensional large-eddy simulations (LES) are performed at multiple mesh resolutions, with the finest grid sufficiently resolved to directly characterize flame front dynamics relevant to engine-scale combustion. The simulations reveal cellular and finger-like flame structures characteristic of TDIs throughout the entire combustion process. Analysis of the local thermo-chemical state demonstrates that differential diffusion induces pronounced mixture stratification and elevates reaction rates, resulting in super-adiabatic temperatures that strongly correlate with flame curvature. Building on these findings, the performance of the baseline artificially thickened flame (ATF) model and a recently developed thermo-diffusive (TD)-aware extension is assessed. Unlike the state-of-the-art ATF model, which suffers from grid dependence and underestimates the experimental pressure trace, the TD-aware formulation captures experimental trends more accurately and provides consistent, grid-independent integrated heat-release (IHR) traces. For the operating condition considered here, the results show that TD effects represent sub-grid-scale contributions that need to be accounted for to obtain consistent predictions of global combustion behavior under the investigated lean H 2 -ICE conditions. • LES captured cellular and finger-like thermo-diffusive flame structures. • Coarser grids suppressed fine-scale instabilities resolved at high resolution. • Local mixture stratification enhanced reactivity and caused super-adiabatic states. • ATF model showed grid bias from missing thermo-diffusive instability treatment. • Thermo-diffusive-aware ATF model reduced grid bias and improved predictive accuracy.

ABSTRACT Flame stabilization of autoignited lifted jet flames under oxygen-diluted conditions relevant to MILD combustion remains challenging. Motivated by the counterintuitive trend—namely, a decrease in liftoff height, H L , of methane/hydrogen jet flame with increasing jet velocity, U 0 , due to differential diffusion—we investigate whether this effect can be leveraged to enhance the stabilization of laminar autoignited lifted methane/hydrogen jet flames under MILD combustion conditions. Two-dimensional numerical simulations are performed using a detailed kinetic mechanism involving 57 species over a wide range of oxidizer oxygen mole fractions, X O 2 . The results show that the decreasing trend of H L with increasing U 0 becomes more pronounced as X O 2 decreases. At sufficiently high U 0 , H L even decreases with decreasing X O 2 , contrary to expectations based on conventional flame behavior in fuel/air mixtures. Complementary one- and two-dimensional simulations reveal that, as X O 2 decreases, the most reactive zone shifts toward the coflow, strengthening the differential diffusion effect. This shift increases the local hydrogen ratio, R H , upstream of the flamebase, thereby promoting earlier local ignition and a higher edge flame propagation speed, S e . The combined effects of enhanced local reactivity and increased edge flame propagation lead to a reduction in H L as X O 2 decreases. Sensitivity tests over inlet temperature and chemical-kinetic mechanisms further indicate that these trends are robust rather than confined to a narrow parameter window.

ABSTRACT This work, based on numerical simulations, discusses the effect of HBr as an inhibitor on the explosion limit of stoichiometric hydrogen‐oxygen gas mixture in a closed spherical vessel. The numerical simulation solves the complete governing equations for mass, momentum, species and energy. In addition, wall surface reactions are included to describe the destruction of chemically reactive radicals. The in‐house INSFLA code is used, which takes into account the differential diffusion and thermal diffusion (Soret effect) to describe the mass transport of species within the system. The inhibitory effect of HBr on all three explosion limits under the considered conditions is analyzed based on sensitivity analysis of the explosion limit with respect to chemical reaction rates and reaction pathway analysis. It is found that the main reason for the inhibition effect is the chain‐propagating step HBr++ which is in effect terminating. However, each of the three explosion limits is governed by different key reactions.

Cross-modality multiband differential conditional diffusion for multimodal emotion recognition in conversation

The next generations of gas turbines are expected to be ‘fuel-flexible’, i.e. to reliably operate with variable fuel mixtures ranging from pure carbon-free fuels such as ammonia and hydrogen to natural gas and blends thereof. These fuel mixtures pose challenges, however, due to variations in combustion dynamics that are primarily caused by the fuels' reactivities and their molecular transport properties. Multiple Mapping Conditioning (MMC) offers a cost-effective simulation approach that can account for realistic species transport and detailed chemical kinetics. Here, an improved mass-based particle mixing method is introduced that can capture differential diffusion by adapting the spatial displacement of the particles and effectuate mixing in different steps. This approach allows accurate modelling with a single particle cloud while maintaining a model-free closure for the chemical source term. The model is tested in a shear layer configuration with various fuel blends. The findings reveal that the new MMC mixing method improves the accuracy of species predictions when compared to standard MMC mixing approaches, and it offers an acceptable estimation of differential diffusion effects for a wide range of fuel blends.

A biodegradable stent is a mesh that is used to treat the narrow or closed part of the artery to open and restore normal blood flow, which is made of a biodegradable material.{However, since the struts in the stent are thin, a simple structural model (called the one-dimensional curved rod model) can be used to mathematically model the stent.So differential diffusion equations can describe the degradation of the stent. The movement of the blood causes microscopic trembling of the edges of the wall. Therefore, the perturbation of the corresponding differential equation by a white noise type term should correspond to the real situation. We consider some linear parabolic equations on graphs with white noise terms. To investigate the initial-boundary value problem for these equations, we reduce it to the corresponding deterministic problem.First, we prove the existence and uniqueness of the we ak solution to deterministic problem for parabolic equations on graphs.Finally, same results are obtained for linear stochastic parabolic equations on graphs.

ABSTRACT Large Eddy Simulations with flamelet-based thermochemistry are used to investigate the behavior of a premixed hydrogen-air flame stabilized by a bluff-body. Validation against experimental data is carried out first to demonstrate the model’s ability to predict both velocity field and flame structure. The capability of the model in predicting differential diffusion effects is then assessed, in particular regarding the coupling between differential diffusion, tangential strain and curvature, and their effect on mixture fraction redistribution and reaction rate variation. Results indicate that unstretched flamelet thermochemistry is capable of capturing the increase in mixture fraction caused by positive resolved strain, as well as negative variations of mixture fraction due to negative curvature. Furthermore, the model is observed to mimic the effects of negative Markstein length to a certain extent, so that positive tangential strain causes reaction rate increase. The interplay between resolved stretch and preferential diffusion is also shown to lead to a shorter flame length which is in better agreement with experimental observations as compared to simulations under unity Lewis number assumption. These findings highlight that the macroscopic effects of differential diffusion and stretch on the premixed hydrogen flame, characterized by significant strain levels, can be predicted using a flamelet-based approach and without recurring to strained flamelets database, which implies important simplifications in the combustion modeling of turbulent hydrogen-premixed flames and offers valuable insights for the design of novel combustors.

Turbulent hydrogen flames with varying operating conditions in the swirl-axial air injection AHEAD combustor were studied computationally with a multi-regime flame closure method, combustion LES / Stochastic fields. The method was validated by comparisons with measurements in isothermal and reacting flows. The velocity fields, flames, mixing fields and thermo-chemical states were analysed in detail. Further comparisons were carried out for different operating conditions to study the effect of global equivalence ratio and axial air injection ratio. On the one hand, it introduces higher axial momentum, which restricts flashback. On the other hand, increasing the global equivalence ratio or axial air injection ratio negatively affects the spatial mixing quality where the axial momentum flux plays an important role. The results also suggest that thermo-chemical states are dominantly controlled by the global equivalence ratio rather than the inlet reactant temperature or flow conditions. The effect of differential diffusion was also studied. Differential diffusion slightly increases the possibility of the upstream occurrence of the flame inner branch, which results in the inner flame branch brush becoming broader. This was found to be related to the changes of the upstream mixing field due to differential diffusion. Nevertheless, the global system is negligibly influenced by differential diffusion due to the high Reynolds number.

Differential diffusion effects on the structure of reactive flows in Marangoni-reaction-diffusion processes

This study presents a comprehensive a priori analysis of tabulated-chemistry models for both laminar and turbulent lean premixed hydrogen flames in strained counterflow configuration. Particular focus is drawn on differential and preferential diffusion effects and the synergistic interaction of thermodiffusive instabilities and turbulence that existing models struggle to capture. Through detailed assessment of various modelling approaches at unfiltered and filtered grids, we identify significant limitations in traditional unstretched flamelet manifolds, particularly their strong filter dependence and systematic reaction rate mispredictions. To address these challenges, we introduce and evaluate novel strained flamelet approaches, including: (1) a one-dimensional manifold constructed from a single strained flamelet that provides computationally efficient and reliable consumption speed predictions at coarser grids, and (2) a two-dimensional manifold combining fixed strain with varying equivalence ratio that demonstrates improved performance in predicting the local reaction rates across multiple grid resolutions. Additionally, we develop a correction methodology derived from laminar simulations that significantly improves consumption speed predictions of unstretched flamelet manifolds in turbulent settings. Unlike previous works, our solutions maintain computational efficiency without increasing manifold dimensionality, keeping memory costs unchanged. These advancements provide guidance for developing reliable LES models that properly account for differential and preferential diffusion and strain effects in practical hydrogen combustion systems.

This paper describes the results of compressible large eddy simulations, with the Stochastic Fields method, of a lean-premixed bluff-body hydrogen flame. Both the LES isothermal and the reacting fields show good agreement with the measurements, revealing the capability of the Stochastic Fields method in bluff-body hydrogen flames. For the reacting field, two diffusive transport models, the constant Schmidt/Prandtl number approximation and mixture-averaged diffusion transport are implemented independently to isolate the effects of differential diffusion on species concentrations and temperature. The mixture-averaged diffusion transport model performs better than the constant Schmidt/Prandtl in terms of mean flame shape. Mixture-averaged diffusion results in local enrichment of the flame root, higher heat release rates, temperatures, fuel consumption rates and more diffused species profiles. High-temperature burnt gas in the recirculation zone elevates the significance of molecular diffusion, related to possible re-laminarization. The results reveal the importance of the inclusion of accurate molecular diffusive transport for hydrogen flames in lean-premixed bluff-body stabilized flames. DMD analysis suggests that both diffusion methods can recover the significant dynamic behaviour of the reacting flow system. Some physical-space differences between the two approaches are also shown in terms of their DMD modes. Novelty and Significance Statement The present work describes an LES study of the effect of differential diffusion on the hydrogen lean-premixed flames. The effect is isolated by comparing LES results obtained with two molecular diffusive transport models; the constant Schmidt/Prandtl number approximation and the mixture-averaged diffusion transport model applied to a bluff-body stabilized hydrogen flame. The solutions from the two diffusion transport methods are compared and analysed regarding flow field and flame characteristics. Dynamic mode decomposition is applied for further analysis. The results highlight the importance of including the differential diffusion effects in lean bluff-body stabilized hydrogen flames. The result is likely to helpful in the simulation of a wide range of hydrogen flame applications. The present work provides an example of hydrogen flame compressible LES with relatively manageable computational costs (3.5M mesh cells) and a multi-regime-robust flame model, i.e., the Stochastic Fields method. The method avoids any assumptions regarding flame regime burning and burning mode selection or parameter tuning. The method also allows the incorporation an ‘accurate’ diffusive transport model, e.g., mixture-averaged, in a relatively straightforward manner. This may significantly help the study and simulation of hydrogen flames. The present work systematically analyses the physical and chemical processes that occurred in the lean-premixed hydrogen flame arising in the NTNU bluff-body burner. Previous computational studies within this burner is limited in number and the present work will help the development hydrogen flame simulation methods. • This work describes the effect of differential diffusion on a hydrogen premixed flame. • This work presents Stochastic Fields LES with an ‘accurate’ diffusive transport model. • The physical–chemical processes of the flame are analyzed.

Purpose This study aims to develop a new numerical framework for modelling soot formation and evolution via the soot particle size distribution (PSD) in turbulent flames. Design/methodology/approach The numerical framework couples an extended soot sectional method with a finite-rate chemistry model based on detailed chemistry, which solves primary particle and soot aggregate number densities in every section with considering turbulence–chemistry interaction. Soot aggregates and gas species are solved simultaneously with considering differential diffusion and mass exchange between soot and chemical species. A dynamic load-balancing approach with a reference mapping model is also incorporated into the numerical framework to accelerate the parallel reacting flow simulations. Findings This new numerical framework is comprehensively used to simulate soot formation and evolution in non-premixed turbulent sooting bluff body flames with different bluff body radii. With the bluff body radius increasing, the increased residence time of soot aggregates in the recirculation zone results in a significant shift of the PSD towards the larger soot aggregate side. The PSD shape always remains bimodal distribution at the centerline. Coagulation predominantly occurs at small soot aggregates, while the polycyclic aromatic hydrocarbon condensation and H-abstraction-C2H2-addition surface growth take significant effect at large soot aggregates. Originality/value Overall good quantitative and qualitative agreements of numerical results with available experimental dataset demonstrate that the new numerical framework can accurately predict the flow properties and well capture the significant soot formation and evolution processes.

Four-dimensional (4D) printing represents a paradigm shift in additive manufacturing, enabling the creation of dynamic, stimuli-responsive structures that can change shape or function over time. This study introduces a significant advancement in this field by developing 4D-printed responsive cellulosic composite (RCC) strips consisting of varying ratios of AFFINISOL (AFF, a hydroxypropyl methylcellulose derivative) and thermoplastic starch (TPS) using fused filament fabrication (FFF) technology. Further, 4D-printed RCC strips were characterized for thermal properties, which revealed glass transition temperature (Tg) between 97 and 104 °C as well as fluid uptake of approximately 70% in 4 h. Subsequent investigation, focused on evaluating the solvent-responsive shape-memory behavior of RCC strips using water as a solvent, highlighted the maximum shape recovery index (SRI), i.e., 0.98 within 105 min. Next, to demonstrate the responsive nature of the RCC strips, the shape-memory behavior was evaluated in solvents with varying polarities. The results revealed a clear interplay between solvent polarity and the rate of shape recovery of RCC strips. The strips demonstrated exceptional recovery in water; however, their recovery was severely hindered in ethanol (EtOH) with an SRI of 0.14 within 105 min and was completely absent in n-hexane. We attribute this polarity-driven response to RCC matrix-solvent interactions and the differential diffusion rate of solvents within the RCC matrix, validated using various hydroalcoholic solutions. These findings establish an avenue for innovative solvent-responsive shape-memory behavior of RCC and underscore the future potential of FFF-mediated 4D-printed RCC strips for biomedical, tissue engineering, soft robotics, and reconfigurable systems applications.

As studies continue to bring forward data on both the complexity and heterogeneity behind the tumor microenvironment, new strategies to understand and unravel the cellular interactions that regulate tumor progression and tumor cell invasion are required. Here, we present a novel and tailorable 4-well 3D culture chamber design capable of studying chemotaxis between several distinct cell types and a cancer cell population of interest. The use of a type I collagen hydrogel as the 3D substrate allowed for a differential molecule diffusion, in which rate of diffusion was associated with molecular weight. When culturing different human stromal cells (hASCs, hDMECs and hDFbs) in the outer wells while keeping VMM15 melanoma cells within the central well it was observed that hASCs and hDFbs presented directional migration throughout the collagen matrix towards the tumor cells. Further analysis revealed a higher area of migration present in the hDFbs when compared to the hASCs, supporting the potential of this system to study the recruitment of supporting cells by cancer cells and how this may impact tumor invasion.

A priori and a posteriori analyses of differential and preferential diffusion in large eddy simulations of partially premixed hydrogen–air flames

Precise and timely radar echo extrapolation is challenging in short-term meteorological nowcasting, given the inherent uncertainty and nonlinearity of weather systems. The diffusion and denoising processes in diffusion models correspond with the intrinsic mechanisms of ‘disorder enhancement’ (entropy increase) and ‘ordered structure formation’ (entropy decrease) in atmospheric motion. Nevertheless, we find that directly applying diffusion models fails to capture the key dynamics. We argue that radar echo evolution is governed by two core processes: advection-dominated bulk displacement and local growth/decay-dominated intensity changes. We thus propose a Physics-Guided Decoupled Diffusion Model (PG-DDM) which injects physical priors through a dual-path design: an Advection Path that derives the instantaneous spatial state from a single frame, and a Growth/Decay Path that captures local processes via inter-frame differences. Our method leverages these complementary physical cues to decouple the learning of motion and change, producing high-fidelity extrapolation sequences. Code is available at https://github.com/azier33/DDM.