arXiv subjects

H. S. Udaykumar

Publications and source records attributed to H. S. Udaykumar.

At least 19 recordsLinked to original sources

arXiv2026

Attention Is All You Need (to Avoid Spurious Oscillations)

Can attention move a shock across several cells in one update without breaking it? We develop a conservative, fixed grid finite-volume scheme in which a CFL-conditioned attention flux selects upstream information according to the transport required by the current time step. One-dimensional inviscid Burgers transport is used as the central mechanism test: the same learned flux remains reliable in the conventional small-step regime and, with a time step four times larger, preserves sharp shocks while using one stage per update. A standard fifth-order WENO scheme with third-order strong-stability-preserving Runge-Kutta time integration (WENO-5+SSP-RK3) is included alongside controlled Forward Euler comparisons to separate flux selection from time integration. The learned attention shifts upstream with the local transport reach and becomes more selective near shocks; inference-time interventions and retrained ablations show that transport-scale information and state-dependent selection contribute directly to performance. Directional two-dimensional scalar Burgers transport and the one-dimensional shallow-water system then test whether the conservation-scale-selection principle transfers beyond the original scalar setting. The results support attention as a learnable information stencil for conservative large-step shock transport, while identifying finite candidate reach and problem-dependent robustness as the present limits.

cs.LG↗

arXiv2026

G-PARC: Graph-Physics Aware Recurrent Convolutional Neural Networks for Spatiotemporal Dynamics on Unstructured Meshes

Physics-aware recurrent convolutional networks (PARC) have demonstrated strong performance in predicting nonlinear spatiotemporal dynamics by embedding differential operators directly into the computational graph of a neural network. However, pixel-based convolutions are restricted to static, uniform Cartesian grids, making them ill-suited to following evolving localized structures in an efficient manner. Graph neural networks (GNNs) naturally handle irregular spatial discretizations, but existing graph-based physics-aware deep learning (PADL) methods have difficulty handling extreme nonlinear regimes. To address these limitations, we propose Graph PARC (G-PARC), which uses moving least squares (MLS) kernels to approximate spatial derivatives on unstructured graphs, and embeds the derivatives of governing partial differential equations into the network's computational graph. G-PARC achieves better accuracy with 2-3x fewer parameters than MeshGraphNet, MeshGraphKAN, and GraphSAGE, replacing the traditional encoder-processor-decoder framework with analytically computed differential operators. We demonstrate that G-PARC (1) generalizes across nonuniform spatial and temporal discretizations; (2) handles moving meshes required for structural deformation; and (3) outperforms existing graph-based PADL methods on nonlinear benchmarks including fluvial hydrology, planar shock waves, and elastoplastic dynamics. By embedding explicit physical operators within the flexibility of GNNs, G-PARC enables accurate modeling of extreme nonlinear phenomena on complex computational domains, moving PADLbeyond idealized Cartesian grids.

cs.LG↗

arXiv2026

High-fidelity simulations of shock initiation of an energetic crystal-binder system due to flyer impact

Meso-scale simulations of energy localization at hotspots provide closure models for multiscale frameworks of shock-to-detonation transition (SDT). Validation of such meso-scale calculations is challenging as direct comparison with experiments is constrained both by limitations of data acquisition in the experiments (e.g., of temperature fields) and modeling over-simplifications in the simulations. To address the latter problem and bring modeling closer to experiments, we advance a high-fidelity meso-scale computational framework for interface-resolved reactive calculations of shock initiation in plastic-bonded explosives (PBXs). Accurate resolution of shock and interfacial dynamics is achieved through higher-order (5th-order WENO) schemes, and sharp interface treatments are implemented for physically accurate material-material interactions. Recently obtained atomistics-consistent material models are used for HMX, with the grid resolution taken down to atomistic scale (O(nm)). The crystal geometries are obtained directly from experiments via nano-CT imaging. The impacting flyer plate, energetic crystal and binder are tracked as distinct phases, and flyer-binder impact and separation are simulated, capturing the flyer deformation and the effects of relief waves from the flyer surface. By combining these high-fidelity modeling components, we evaluate how closely simulations can approach experimental data. Overall, this work provides as assessment of which aspects of numerical treatment and material modeling have the greatest impact on meso-scale simulations of flyer-induced initiation of PBXs, and points to where further improvements are necessary.

physics.comp-ph↗

arXiv2026

Size is Not the Solution: Deformable Convolutions for Effective Physics Aware Deep Learning

Physics-aware deep learning (PADL) enables rapid prediction of complex physical systems, yet current convolutional neural network (CNN) architectures struggle with highly nonlinear flows. While scaling model size addresses complexity in broader AI, this approach yields diminishing returns for physics modeling. Drawing inspiration from Hybrid Lagrangian-Eulerian (HLE) numerical methods, we introduce deformable physics-aware recurrent convolutions (D-PARC) to overcome the rigidity of CNNs. Across Burgers' equation, Navier-Stokes, and reactive flows, D-PARC achieves superior fidelity compared to substantially larger architectures. Analysis reveals that kernels display anti-clustering behavior, evolving into a learned "active filtration" strategy distinct from traditional h- or p-adaptivity. Effective receptive field analysis confirms that D-PARC autonomously concentrates resources in high-strain regions while coarsening focus elsewhere, mirroring adaptive refinement in computational mechanics. This demonstrates that physically intuitive architectural design can outperform parameter scaling, establishing that strategic learning in lean networks offers a more effective path forward for PADL than indiscriminate network expansion.

cs.LG↗

arXiv2025

Reduced Order Modeling of Energetic Materials Using Physics-Aware Recurrent Convolutional Neural Networks in a Latent Space (LatentPARC)

Physics-aware deep learning (PADL) has gained popularity for use in complex spatiotemporal dynamics (field evolution) simulations, such as those that arise frequently in computational modeling of energetic materials (EM). Here, we show that the challenge PADL methods face while learning complex field evolution problems can be simplified and accelerated by decoupling it into two tasks: learning complex geometric features in evolving fields and modeling dynamics over these features in a lower dimensional feature space. To accomplish this, we build upon our previous work on physics-aware recurrent convolutions (PARC). PARC embeds knowledge of underlying physics into its neural network architecture for more robust and accurate prediction of evolving physical fields. PARC was shown to effectively learn complex nonlinear features such as the formation of hotspots and coupled shock fronts in various initiation scenarios of EMs, as a function of microstructures, serving effectively as a microstructure-aware burn model. In this work, we further accelerate PARC and reduce its computational cost by projecting the original dynamics onto a lower-dimensional invariant manifold, or 'latent space.' The projected latent representation encodes the complex geometry of evolving fields (e.g. temperature and pressure) in a set of data-driven features. The reduced dimension of this latent space allows us to learn the dynamics during the initiation of EM with a lighter and more efficient model. We observe a significant decrease in training and inference time while maintaining results comparable to PARC at inference. This work takes steps towards enabling rapid prediction of EM thermomechanics at larger scales and characterization of EM structure-property-performance linkages at a full application scale.

cond-mat.mtrl-sci↗

arXiv2025

Multi-resolution Physics-Aware Recurrent Convolutional Neural Network for Complex Flows

We present MRPARCv2, Multi-resolution Physics-Aware Recurrent Convolutional Neural Network, designed to model complex flows by embedding the structure of advection-diffusion-reaction equations and leveraging a multi-resolution architecture. MRPARCv2 introduces hierarchical discretization and cross-resolution feature communication to improve the accuracy and efficiency of flow simulations. We evaluate the model on a challenging 2D turbulent radiative layer dataset from The Well multi-physics benchmark repository and demonstrate significant improvements when compared to the single resolution baseline model, in both Variance Scaled Root Mean Squared Error and physics-driven metrics, including turbulent kinetic energy spectra and mass-temperature distributions. Despite having 30% fewer trainable parameters, MRPARCv2 outperforms its predecessor by up to 50% in roll-out prediction error and 86% in spectral error. A preliminary study on uncertainty quantification was performed, and we also analyzed the model's performance under different levels of abstractions of the flow, specifically on sampling subsets of field variables. We find that the absence of physical constraints on the equation of state (EOS) in the network architecture leads to degraded accuracy. A variable substitution experiment confirms that this issue persists regardless of which physical quantity is predicted directly. Our findings highlight the advantages of multi-resolution inductive bias for capturing multi-scale flow dynamics and suggest the need for future PIML models to embed EOS knowledge to enhance physical fidelity.

physics.flu-dyn↗

arXiv2025

A physics-aware deep learning model for shear band formation around collapsing pores in shocked reactive materials

Modeling shock-to-detonation phenomena in energetic materials (EMs) requires capturing complex physical processes such as strong shocks, rapid changes in microstructural morphology, and nonlinear dynamics of chemical reaction fronts. These processes participate in energy localization at hotspots, which initiate chemical energy release leading to detonation. This study addresses the formation of hotspots in crystalline EMs subjected to weak-to-moderate shock loading, which, despite its critical relevance to the safe storage and handling of EMs, remains underexplored compared to the well-studied strong shock conditions. To overcome the computational challenges associated with direct numerical simulations, we advance the Physics-Aware Recurrent Convolutional Neural Network (PARCv2), which has been shown to be capable of predicting strong shock responses in EMs. We improved the architecture of PARCv2 to rapidly predict shear localizations and plastic heating, which play important roles in the weak-to-moderate shock regime. PARCv2 is benchmarked against two widely used physics-informed models, namely, Fourier neural operator and neural ordinary differential equation; we demonstrate its superior performance in capturing the spatiotemporal dynamics of shear band formation. While all models exhibit certain failure modes, our findings underscore the importance of domain-specific considerations in developing robust AI-accelerated simulation tools for reactive materials.

cs.LG↗

arXiv2025

Latent Representation Learning of Multi-scale Thermophysics: Application to Dynamics in Shocked Porous Energetic Material

Coupling of physics across length and time scales plays an important role in the response of microstructured materials to external loads. In a multi-scale framework, unresolved (subgrid) meso-scale dynamics is upscaled to the homogenized (macro-scale) representation of the heterogeneous material through closure models. Deep learning models trained using meso-scale simulation data are now a popular route to assimilate such closure laws. However, meso-scale simulations are computationally taxing, posing practical challenges in training deep learning-based surrogate models from scratch. In this work, we investigate an alternative meta-learning approach motivated by the idea of tokenization in natural language processing. We show that one can learn a reduced representation of the micro-scale physics to accelerate the meso-scale learning process by tokenizing the meso-scale evolution of the physical fields involved in an archetypal, albeit complex, reactive dynamics problem, \textit{viz.}, shock-induced energy localization in a porous energetic material. A probabilistic latent representation of \textit{micro}-scale dynamics is learned as building blocks for \textit{meso}-scale dynamics. The \textit{meso-}scale latent dynamics model learns the correlation between neighboring building blocks by training over a small dataset of meso-scale simulations. We compare the performance of our model with a physics-aware recurrent convolutional neural network (PARC) trained only on the full meso-scale dataset. We demonstrate that our model can outperform PARC with scarce meso-scale data. The proposed approach accelerates the development of closure models by leveraging inexpensive micro-scale simulations and fast training over a small meso-scale dataset, and can be applied to a range of multi-scale modeling problems.

physics.comp-ph↗

arXiv2024

On resolving meso-scale calculations of pore-collapse-generated hotspots in energetic crystals for consistency with atomistic models

Meso-scale calculations of pore collapse and hotspot formation in energetic crystals provide closure models to macro-scale hydrocodes for predicting the shock sensitivity of energetic materials. To this end, previous works obtained atomistics-consistent material models for two common energetic crystals, HMX and RDX, such that pore collapse calculations adhered closely to molecular dynamics (MD) results on key features of energy localization, particularly the profiles of the collapsing pores, appearance of shear bands, and the transition from viscoplastic to hydrodynamic collapse. However, some important aspects such as the temperature distributions in the hotspot were not as well captured. One potential issue was noted but not resolved adequately in those works, namely the grid resolution that should be employed in the meso-scale calculations for various pore sizes and shock strengths. Conventional computational mechanics guidelines for selecting meshes as fine as possible, balancing computational effort, accuracy and grid independence, were shown not to produce physically consistent features associated with shear localization. Here, we examine the physics of pore collapse, shear band evolution and structure, and hotspot formation, for both HMX and RDX; we then evaluate under what conditions atomistics-consistent models yield physically correct (considering MD as ground truth) hotspots for a range of pore diameters, from nm to microns, and for a wide range of shock strengths. The study provides insights into the effects of pore size and shock strength on pore collapse and hotspots, identifying aspects such as size-independent behaviors, and proportion of energy contained in shear as opposed to jet impact-heated regions of the hotspot. Areas for further improvement of atomistics-consistent material models are also indicated.

cond-mat.mes-hall↗

arXiv2024

PARCv2: Physics-aware Recurrent Convolutional Neural Networks for Spatiotemporal Dynamics Modeling

Modeling unsteady, fast transient, and advection-dominated physics problems is a pressing challenge for physics-aware deep learning (PADL). The physics of complex systems is governed by large systems of partial differential equations (PDEs) and ancillary constitutive models with nonlinear structures, as well as evolving state fields exhibiting sharp gradients and rapidly deforming material interfaces. Here, we investigate an inductive bias approach that is versatile and generalizable to model generic nonlinear field evolution problems. Our study focuses on the recent physics-aware recurrent convolutions (PARC), which incorporates a differentiator-integrator architecture that inductively models the spatiotemporal dynamics of generic physical systems. We extend the capabilities of PARC to simulate unsteady, transient, and advection-dominant systems. The extended model, referred to as PARCv2, is equipped with differential operators to model advection-reaction-diffusion equations, as well as a hybrid integral solver for stable, long-time predictions. PARCv2 is tested on both standard benchmark problems in fluid dynamics, namely Burgers and Navier-Stokes equations, and then applied to more complex shock-induced reaction problems in energetic materials. We evaluate the behavior of PARCv2 in comparison to other physics-informed and learning bias models and demonstrate its potential to model unsteady and advection-dominant dynamics regimes.

cs.LG↗

arXiv2023

Artificial intelligence approaches for materials-by-design of energetic materials: state-of-the-art, challenges, and future directions

Artificial intelligence (AI) is rapidly emerging as an enabling tool for solving various complex materials design problems. This paper aims to review recent advances in AI-driven materials-by-design and their applications to energetic materials (EM). Trained with data from numerical simulations and/or physical experiments, AI models can assimilate trends and patterns within the design parameter space, identify optimal material designs (micro-morphologies, combinations of materials in composites, etc.), and point to designs with superior/targeted property and performance metrics. We review approaches focusing on such capabilities with respect to the three main stages of materials-by-design, namely representation learning of microstructure morphology (i.e., shape descriptors), structure-property-performance (S-P-P) linkage estimation, and optimization/design exploration. We provide a perspective view of these methods in terms of their potential, practicality, and efficacy towards the realization of materials-by-design. Specifically, methods in the literature are evaluated in terms of their capacity to learn from a small/limited number of data, computational complexity, generalizability/scalability to other material species and operating conditions, interpretability of the model predictions, and the burden of supervision/data annotation. Finally, we suggest a few promising future research directions for EM materials-by-design, such as meta-learning, active learning, Bayesian learning, and semi-/weakly-supervised learning, to bridge the gap between machine learning research and EM research.

cond-mat.mtrl-sci↗

arXiv2023

PARC: Physics-Aware Recurrent Convolutional Neural Networks to Assimilate Meso-scale Reactive Mechanics of Energetic Materials

The thermo-mechanical response of shock-initiated energetic materials (EM) is highly influenced by their microstructures, presenting an opportunity to engineer EM microstructure in a "materials-by-design" framework. However, the current design practice is limited, as a large ensemble of simulations is required to construct the complex EM structure-property-performance linkages. We present the Physics-Aware Recurrent Convolutional (PARC) Neural Network, a deep-learning algorithm capable of learning the mesoscale thermo-mechanics of EM from a modest number of high-resolution direct numerical simulations (DNS). Validation results demonstrated that PARC could predict the themo-mechanical response of shocked EM with a comparable accuracy to DNS but with notably less computation time. The physics awareness of PARC enhances its modeling capabilities and generalizability, especially when challenged in unseen prediction scenarios. We also demonstrate that visualizing the artificial neurons at PARC can shed light on important aspects of EM thermos-mechanics and provide an additional lens for conceptualizing EM.

cond-mat.mtrl-sci↗

arXiv2023

Challenges and opportunities for machine learning in multiscale computational modeling

Many mechanical engineering applications call for multiscale computational modeling and simulation. However, solving for complex multiscale systems remains computationally onerous due to the high dimensionality of the solution space. Recently, machine learning (ML) has emerged as a promising solution that can either serve as a surrogate for, accelerate or augment traditional numerical methods. Pioneering work has demonstrated that ML provides solutions to governing systems of equations with comparable accuracy to those obtained using direct numerical methods, but with significantly faster computational speed. These high-speed, high-fidelity estimations can facilitate the solving of complex multiscale systems by providing a better initial solution to traditional solvers. This paper provides a perspective on the opportunities and challenges of using ML for complex multiscale modeling and simulation. We first outline the current state-of-the-art ML approaches for simulating multiscale systems and highlight some of the landmark developments. Next, we discuss current challenges for ML in multiscale computational modeling, such as the data and discretization dependence, interpretability, and data sharing and collaborative platform development. Finally, we suggest several potential research directions for the future.

cs.LG↗

arXiv2023

A physics-aware deep learning model for energy localization in multiscale shock-to-detonation simulations of heterogeneous energetic materials

Predictive simulations of the shock-to-detonation transition (SDT) in heterogeneous energetic materials (EM) are vital to the design and control of their energy release and sensitivity. Due to the complexity of the thermo-mechanics of EM during the SDT, both macro-scale response and sub-grid mesoscale energy localization must be captured accurately. This work proposes an efficient and accurate multiscale framework for SDT simulations of EM. We introduce a new approach for SDT simulation by using deep learning to model the mesoscale energy localization of shock-initiated EM microstructures. The proposed multiscale modeling framework is divided into two stages. First, a physics-aware recurrent convolutional neural network (PARC) is used to model the mesoscale energy localization of shock-initiated heterogeneous EM microstructures. PARC is trained using direct numerical simulations (DNS) of hotspot ignition and growth within microstructures of pressed HMX material subjected to different input shock strengths. After training, PARC is employed to supply hotspot ignition and growth rates for macroscale SDT simulations. We show that PARC can play the role of a surrogate model in a multiscale simulation framework, while drastically reducing the computation cost and providing improved representations of the sub-grid physics. The proposed multiscale modeling approach will provide a new tool for material scientists in designing high-performance and safer energetic materials.

cond-mat.mtrl-sci↗

arXiv2022

Mechanisms of shock-induced initiation at micro-scale defects in energetic crystal-binder systems

Crystals of energetic materials, such as HMX, embedded in plastic binders are the building blocks of plastic-bonded explosives. Such heterogeneous energetic materials contain microstructural features such as sharp corners, interfaces between crystal and binder, intra- and extra-granular voids, and other defects. Energy localization or hotspots arise during shock interaction with the microstructural heterogeneities, leading to the initiation of PBXs. In this paper, high-resolution numerical simulations are performed to elucidate the mechanistic details of shock-induced initiation in a PBX; we examine four different mechanisms: Shock-focusing at sharp corners or edges and its dependency on the shape of the crystal, and the strength of the applied shock; debonding between crystal and binder interfaces; collapse of voids in the binder located near an HMX crystal; and the collapse of voids within HMX crystals. Insights are obtained into the relative contributions of these mechanisms to the ignition and growth of hotspots. Understanding these mechanisms of energy localization and their relative importance for hotspot formation and initiation sensitivity of PBXs will aid in the design of energetic material-driven systems with controlled sensitivity, to prevent accidental initiation and ensure reliable performance.

cond-mat.mtrl-sci↗

arXiv2020

A simulation-derived surrogate model for the vaporization rate of aluminum droplets heated by a passing shock wave

The vaporization rate of aluminum droplets in shocked flows plays a crucial role in determining the energy release rate during the combustion of the aluminized energetic materials. In this paper, the physics of the vaporization of aluminum droplets in shocked flows is numerically investigated. Surrogate models for the temporally averaged Sherwood number and Nusselt number, cast as functions of shock Mach number and Reynolds number, are developed from the simulation-based data. The results show that the Sherwood number and the Nusselt number of the droplet increase monotonically with the Reynolds number. On the other hand, the Sherwood number and the Nusselt number exhibit non-monotonic behavior with increasing shock Mach number due to the transition of the post-shock flow from subsonic to the supersonic speeds as the shock Mach number is increased from 1.1 to 3.5. In contrast with available models in the literature that are commonly used in process scale computations of aluminum droplet vaporization, the current models for the Sherwood number and the Nusselt number are applicable over a wide range of the Reynolds number and the Mach number and will be useful in the macro-scale multi-phase simulations of the combustion of aluminumized energetic materials in high-speed flows.

physics.flu-dyn↗

arXiv2020

Deep learning for synthetic microstructure generation in a materials-by-design framework for heterogeneous energetic materials

The sensitivity of heterogeneous energetic (HE) materials (propellants, explosives, and pyrotechnics) is critically dependent on their microstructure. Initiation of chemical reactions occurs at hot spots due to energy localization at sites of porosities and other defects. Emerging multi-scale predictive models of HE response to loads account for the physics at the meso-scale, i.e. at the scale of statistically representative clusters of particles and other features in the microstructure. Meso-scale physics is infused in machine-learned closure models informed by resolved meso-scale simulations. Since microstructures are stochastic, ensembles of meso-scale simulations are required to quantify hot spot ignition and growth and to develop models for microstructure-dependent energy deposition rates. We propose utilizing generative adversarial networks (GAN) to spawn ensembles of synthetic heterogeneous energetic material microstructures. The method generates qualitatively and quantitatively realistic microstructures by learning from images of HE microstructures. We show that the proposed GAN method also permits the generation of new morphologies, where the porosity distribution can be controlled and spatially manipulated. Such control paves the way for the design of novel microstructures to engineer HE materials for targeted performance in a materials-by-design framework.

cond-mat.mtrl-sci↗

arXiv2019

Structure-Property Linkage in Shocked Multi-Material Flows Using A Level-Set Based Eulerian Image-To-Computation Framework

Morphology and dynamics at the meso-scale play crucial roles in the overall macro- or system-scale flow of heterogeneous materials. In a multi-scale framework, closure models upscale unresolved sub-grid (meso-scale) physics and therefore encapsulate structure-property (S-P) linkages to predict performance at the macro-scale. This work establishes a route to structure-property linkage, proceeding all the way from imaged micro-structures to flow computations in one unified level set-based framework. Level sets are used to: 1) Define embedded geometries via image segmentation; 2) Simulate the interaction of sharp immersed boundaries with the flow field, and 3) Calculate morphological metrics to quantify structure. Meso-scale dynamics are computed to calculate sub-grid properties, i.e. closure models for momentum and energy equations. The structure-property linkage is demonstrated for two types of multi-material flows: interaction of shocks with a cloud of particles and reactive meso-mechanics of pressed energetic materials. We also present an approach to connect local morphological characteristics in a microstructure containing topologically complex features with the shock response of imaged samples of such materials. This paves the way for using geometric machine learning techniques to associate imaged morphologies with their properties.

physics.comp-ph↗