Engineering PapersSearch

SEARCH · Engineering Papers

Results for “Reduced-order”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2

Toward real-time optimization through model reduction and model discrepancy sensitivities

Optimization problems arise in a range of scenarios, from optimal control to model parameter estimation. In many applications, such as the development of digital twins, it is essential to solve these optimization problems within wall-clock-time limitations. However, this is often unattainable for complex systems, such as those modeled by nonlinear partial differential equations. One strategy for mitigating this issue is to construct a reduced-order model (ROM) that enables more rapid optimization. In particular, the use of nonintrusive ROMs—those that do not require access to the full-order model at evaluation time—is popular because they facilitate the computation of optimization solutions within the wall-clock time requirements. However, the optimization solution will be unreliable if the iterates move outside the ROM training data. This article proposes the use of hyper-differential sensitivity analysis with respect to model discrepancy (HDSA-MD) as a computationally efficient tool to augment ROM-constrained optimization and improve its reliability. The proposed approach consists of two phases: (i) an offline phase where several full-order model evaluations are computed to train the ROM, and (ii) an online phase where a ROM-constrained optimization problem is solved, a limited number of full-order model evaluations are computed, and HDSA-MD is used to enhance the optimization solution. Numerical results are demonstrated for two examples, atmospheric contaminant control and wildfire ignition location estimation, in which a ROM is trained offline using inaccurate atmospheric data. In conclusion, the HDSA-MD update yields a significant improvement in the ROM-constrained optimization solution using only one full-order model evaluation online with corrected atmospheric data.

PDE-constrained optimization

Deposition Height Prediction in Directed Energy Deposition

Using 316L stainless steel as a model material, reduced-order models are developed to predict capture efficiency, deposition height, and site-specific hardness in directed energy deposition. Capture efficiency is predicted over a 15 to 55 pct range using a dimensionless number derived from processing conditions and thermophysical properties. Deposition height is predicted over a 0.3 to 1.3 mm range without in situ sensing or prior training data, using two models based on the same mass and energy-balance principles. Predictions are compared with machine learning approaches. A quantitative relationship links deposition height, primary dendrite arm spacing (PDAS), and hardness: heights of 0.3 to 1.1 mm correspond to PDAS values of 2.7 to 5.1 µm and Vickers hardness (HV) of 160 to 219. Thinner layers cool more rapidly, producing finer microstructures and higher hardness. Samples fabricated with in situ variations in deposition height exhibited up to 55 HV differences between thick and thin regions, demonstrating that local control of deposition height enables predictive, site-specific hardness within a single build. These results establish deposition height prediction as a pathway for a priori process design and property control in directed energy deposition for 316L stainless steel.

Kunkel, William [Univ. of Wisconsin, Madison, WI (

Ensemble Kalman filter for data assimilation coupled with low-resolution computations techniques applied in fluid dynamics

This paper presents an innovative Reduced-order model (ROM) for merging experimental and simulation data using data assimilation (DA) to estimate the "True" state of a fluid dynamics system, leading to more accurate predictions. Our methodology introduces a novel approach by implementing the ensemble Kalman filter (EnKF) within a reduced-dimensional framework, grounded in a robust theoretical foundation and applied to fluid dynamics. To address the substantial computational demands of DA, the proposed ROM employs low-resolution (LR) techniques to drastically reduce computational costs. This innovative approach involves downsampling datasets for DA computations, followed by an advanced reconstruction technique based on low-cost singular value decomposition (lcSVD). The lcSVD method, a key innovation in this paper, has never been applied to DA before and offers a highly efficient way to enhance resolution with minimal computational resources. Our results demonstrate significant reductions in both computation time and RAM usage through these LR techniques without compromising the accuracy of the estimations. For instance, in a turbulent test case, for a data compression rate of 15.9, the LR approach can achieve a speed-up of 13.7 and a RAM compression of 90.9% while maintaining a low relative root mean square error (RRMSE) of 2.6%, compared to 0.8% in the high-resolution (HR) reference. Furthermore, we highlight the effectiveness of the EnKF in estimating and predicting the state of fluid flow systems based on limited observations and given low-fidelity numerical data. This paper highlights the potential of the proposed DA method in fluid dynamics applications, particularly for improving computational efficiency in CFD and related fields. Its ability to balance accuracy with low computational and memory costs makes it especially suitable for large-scale and real-time applications, such as environmental monitoring or engineering design. This method will be incorporated into ModelFLOWs-app.

Data Assimilation

Higher-order LaSDI: Reduced order modeling with multiple time derivatives

Solving complex partial differential equations (PDEs) is essential across scientific disciplines but often requires numerical models that can be prohibitively expensive in time-sensitive applications. Reduced-order models (ROMs) address this challenge by exploiting low-dimensional structure to create fast approximations. The Latent Space Dynamics Identification (LaSDI) framework has demonstrated success in learning ROMs for parameterized PDE families, but remains limited to first-order systems. Here, in this paper, we propose Higher-Order LaSDI (HLaSDI), which extends the LaSDI framework to PDEs with arbitrary order of time derivatives. This generalization significantly expands the applicability of LaSDI-based methods to systems previously outside their scope, including hyperbolic PDEs. We demonstrate HLaSDI’s accuracy and efficiency on several linear and nonlinear benchmark problems.

97 MATHEMATICS AND COMPUTING

Regional surrogates for predictive control of digital twins

Digital twins of complex systems must involve a model that is fast, generalizable, and usable for real-time control. For example, high-fidelity nonlinear multiphysics simulations can capture laser-material interactions, but are too slow for optimization or model predictive control (MPC). Reduced-order models, used to accelerate such computation, frequently fail to generalize to unseen inputs or control states. We show theoretically that this failure is intrinsic, i.e., that a learned model is non-unique outside the sampled subspace when its low-rank structure arises from limited excitation and clustered eigenvalues, rather than from a user-imposed truncation alone. Motivated by this result, we propose a control-ready regional surrogate-construction framework for both autonomous and nonautonomous dynamics; it employs Koopman lifting to represent nonlinearities, while preserving spatial locality. We illustrate our approach by constructing a control-ready surrogate for the digital twin of a thermal component of additive-manufacturing process. Our surrogate, localized in space through a von Neumann stencil, is learned from noisy high-fidelity simulations that emulate thermal-camera images collected during the manufacturing. It is linear in thermo-physically augmented states so that MPC reduces to a convex quadratic program. The surrogate requires no online correction, generalizes to unseen scan paths and power profiles of the laser, and is more than three orders of magnitude faster than a finite-difference solver. Furthermore, when the MPC sequence computed on the digital twin is applied to this solver, closed-loop temperature regulation is recovered, showing that the surrogate preserves control-relevant input-output behavior.

Data-driven model

Latent space dynamics identification for interface tracking with application to shock-induced pore collapse

Capturing sharp, evolving interfaces remains a central challenge in reduced-order modeling, especially when data is limited and the system exhibits localized nonlinearities or discontinuities. Here, we propose LaSDI-IT (Latent Space Dynamics Identification for Interface Tracking), a data-driven framework that combines low-dimensional latent dynamics learning with explicit interface-aware encoding to enable accurate and efficient modeling of physical systems involving moving material boundaries. At the core of LaSDI-IT is a revised autoencoder architecture that jointly reconstructs the physical field and an indicator function representing material regions or phases, allowing the model to track complex interface evolution without requiring detailed physical models or mesh adaptation. The latent dynamics are learned through linear regression in the encoded space and generalized across parameter regimes using Gaussian process interpolation with greedy sampling. We demonstrate LaSDI-IT on the problem of shock-induced pore collapse in high explosives, a process characterized by sharp temperature gradients and dynamically deforming pore geometries. The method achieves relative prediction errors below 9% across the parameter space, accurately recovers key quantities of interest such as pore area and hot spot formation, and matches the performance of dense training with only half the data. This latent dynamics prediction was 10 6 times faster than the conventional high-fidelity simulation, proving its utility for multi-query applications. These results highlight LaSDI-IT as a general, data-efficient framework for modeling discontinuity-rich systems in computational physics, with potential applications in multiphase flows, fracture mechanics, and phase change problems.

Gaussian process

Learning Latent Representations to Bridge Coarse-Grained and Atomistic Resolutions in Polymer Simulations

We present a machine-learning-based framework for learning reduced-order representations of polymer chain conformations across coarse-grained (CG) and united-atom (UA) fidelities. By employing linear singular value decomposition and nonlinear autoencoders, we compress high-dimensional polymer configurations into latent spaces with minimal loss of structural accuracy. Crucially, we demonstrate a near-perfect linear mapping between CG and UA latent spaces, enabling an efficient super-resolution back-mapping procedure that reconstructs high-fidelity UA configurations from CG simulations. While minor structural inaccuracies occur, they are effectively corrected through a brief molecular dynamics relaxation, forming a practical hybrid machine learning−physics scheme. This approach establishes the key structural prerequisites for accelerated polymer dynamics simulations: a compact and accurate latent encoding of polymer chain conformations and a validated multi-fidelity mapping that permits reconstruction of UA structures from CG configurations. The extension of this framework to explicit time evolution within the latent space, enabling dynamics to be propagated at CG fidelity and decoded to UA resolution only when required, represents a natural and well-motivated direction for future work.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH

Customizable wave tailoring nonlinear materials enabled by bilevel inverse design

Abstract Passive wave transformation via nonlinearity is ubiquitous in settings from acoustics to optics and electromagnetics. It is well known that different nonlinearities yield different effects on propagating signals, which raises the question of “what precise nonlinearity is the best for a given wave tailoring application?” In this work, considering a one-dimensional spring-mass chain connected by polynomial springs (a variant of the Fermi-Pasta-Ulam-Tsingou system), we introduce a bilevel inverse design method which couples the shape optimization of structures for tailored constitutive responses with reduced-order nonlinear dynamical inverse design. We apply it to two qualitatively distinct problems—minimization of peak transmitted kinetic energy from impact, and pulse shape transformation—demonstrating our method’s breadth of applicability. For the impact problem, we obtain two fundamental insights. First, small differences in nonlinearity can drastically change the dynamic response of the system, from severely under- to outperforming a comparative linear system. Second, the oft-used strategy of impact mitigation via “energy locking” bistability can be significantly outperformed by our optimal nonlinearity. We validate this case with impact experiments and find excellent agreement. This study establishes a framework for broader passive nonlinear mechanical wave tailoring material design, with applications to computing, signal processing, shock mitigation, and autonomous materials.

Science & Technology - Other Topics

Numerical Simulation of a Natural Convection–Driven Air-Cooled Reactor Cavity Cooling System Experiment

Ensuring the efficient removal of decay heat from the reactor vessel is essential for the safety of advanced reactor technologies. Several Generation-IV concepts incorporate variations in the reactor vessel cooling systems to achieve this objective. High-temperature gas-cooled reactors utilize a reactor cavity cooling system (RCCS), a passive ex-vessel system designed to operate without active components or external power during accident conditions. The RCCS removes decay heat primarily through radiative and convective heat transfer mechanisms. Here, this study presents a comprehensive validation of a computational fluid dynamics Reynolds-averaged Navier-Stokes model for the University of Wisconsin-Madison air-cooled RCCS facility. Validation was conducted for both high- and low-power natural convection cases under a uniform heating profile. Near-wall resolution was found to be critical for accurately modeling natural convection in the RCCS; employing an all-𝑦 + wall treatment resulted in wall temperature discrepancies exceeding 50 °⁢𝐶 compared to a wall-resolved mesh. Thermal-hydraulic behaviors under natural and forced convection conditions were compared within the heated cavity and RCCS. A turbulence model sensitivity analysis indicated that low-Reynolds number k-ɛ, k-ω shear stress transport (SST), and Reynolds stress transport models produce similar wall temperature predictions. A buoyancy modeling sensitivity study revealed that the Boussinesq approximation significantly underpredicted thermal-hydraulic behavior in the RCCS. Based on these findings, modeling recommendations are provided. The validated data set along with identified sensitivities refine the modeling of natural convection in the RCCS. The information produced by this study supports RCCS design, optimization, and safety evaluations, enabling the calibration and verification of reduced-order thermal-hydraulic models.

22 - GENERAL STUDIES OF NUCLEAR REACTORS

Active learning emulators for nuclear two-body scattering in momentum space

In this work we extend the active learning emulators for two-body scattering in coordinate space with error estimation, recently developed by Maldonado et al. [Phys. Rev. C 112, 024002], to coupled-channel scattering in momentum space. Our full-order model (FOM) solver is based on the Lippmann-Schwinger integral equation for the scattering t-matrix as opposed to the radial Schrödinger equation. We use (Petrov-)Galerkin projections and high-fidelity calculations at a few snapshots across the parameter space of the interaction to construct efficient reduced-order models (ROMs), trained by a greedy algorithm for locally optimal snapshot selection. Both the FOM solver and the corresponding ROMs are implemented efficiently in Python using Google's JAX library. We present results for emulating scattering phase shifts in coupled and uncoupled channels and cross sections, and assess the accuracy of the developed ROMs and their computational speedup factors. We also develop emulator error estimation for both the t-matrix and the total cross section. The software framework for reproducing and extending our results is publicly available. Together with our recent advances in developing active-learning emulators for three-body scattering, these emulator frameworks set the stage for full Bayesian calibrations of chiral nuclear interactions and optical models against scattering data with quantified emulator errors.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS

High-Fidelity Building Emulator

This dataset provides high-fidelity time series data for an emulated commercial office building sited in the Chicago, IL area during a Typical Meteorological Year (TMY). This dataset consists of air-side HVAC measurements and control inputs, and it includes normal operations as well as various implemented faults (with associated ground truth measurements) implemented on selected days. This data could be used to quantify and compare the impacts of different faults, and it could also be used as training or validation data for machine learning algorithms (e.g., reduced-order modelling, fault detection and diagnosis).

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI

Development of a Techno-Economic Analysis Framework for a Solar Thermochemical Fuel Production Process

Synthetic liquid fuels can provide a drop-in substitute for fossil-based fuels in sectors such as aviation and maritime, where electrification is not a viable option due to the need for high specific energy density. However, for these alternative fuels to be adopted at a commercial scale, their price must be competitive compared to their fossil-based counterparts. The reverse water-gas shift (RWGS) reaction offers a promising pathway, using hydrogen (sourced from electrolysis) and carbon dioxide as the feed and reacting to produce syngas - a mixture of H2 and CO at a specific ratio. Syngas is a useful precursor that can be converted into fuels and chemicals via known downstream processes, such as liquid transportation fuels via Fischer-Tropsch (FT) synthesis. The RWGS reaction is currently not applied in commercial scale, unlike the rest of the components in the process chain (electrolyzers and syngas-to-fuel synthesis units). The RWGS reaction poses several challenges due to its restrictive thermodynamics. Being an equimolar reaction, high temperatures and a large excess of H2 are needed to achieve reasonable CO2 conversion at equilibrium. This has detrimental effects on practical process implementation and the quality of syngas that can be produced, with direct effect on the energy and capital requirements, as well as the need for expensive downstream separation. In this work, we are proposing to develop a new concentrating solar thermal (CST) compatible RWGS reactor, performing the reaction in a 2-step chemical looping process using metal oxide at a temperature range of 600-800 degrees Celsius. By decoupling the reactor from the solar receiver, the Generation 3 (Gen3) CST technology could be utilized, together with its proposed thermal energy storage (TES) technology, benefitting from a good match to the required temperatures. CST technology is a viable option for supplying the heat that could be rapidly deployed in scale, thus being a good match to the gas-to-liquid (GTL) process which requires a large minimal scale to be commercially viable. The integration of TES with CST also allows operating the plant at large annual capacity factors and avoids multiple shutdown/startup cycles, thus fitting into the steady-state operation mode that most GTL processes require. The main innovation in the proposed design hinges on a countercurrent reaction design using a packed bed reactor. In 2019 Metcalfe et al. showed the benefits of countercurrent species exchange could be realized in a redox chemical-looping processes, by storing the favorable countercurrent chemical potential profiles in a packed bed of non-stoichiometric oxide. Metcalfe et al. applied this breakthrough concept to the WGS reaction, which is conventionally a co-feed catalytic process, showing a dramatic improvement. Bulfin et al. (2023) performed a similar proof-of-concept demonstration for the RWGS reaction using CeO2, achieving cumulative and peak CO2 conversions of 88% and 95%, respectively, compared to a thermodynamic limit of 58% for the co-feed catalytic process at the same conditions. In our new REGENLOOP project, we are developing a reactor prototype from the heat-exchange packed bed reactor-type, a commonly used reactor in the chemical industry. The endothermic heat of reduction will be supplied to the reactor using CST, while the same heat transfer fluid (HTF) mechanism will be used to extract the exothermic heat of oxidation. An array of multiple reactors is used to supply constant high-purity CO stream, that is then mixed with H2 from electrolysis to produce a high-purity syngas at the required composition. By removing the CO-CO2 separation after the RWGS process, significant energy and cost reduction can be achieved. A physics-based TEA framework is currently being developed, covering all the major plant processes, from the solar collection through storage, chemical looping RWGS, GTL, and auxiliary unit operations, up to the liquid hydrocarbon product. This modeling framework will utilize reduced-order models for the chemical looping RWGS and TES, CST modeling using SolarPILOT, and Aspen Plus for the GTL. By using this combined physics-based approach, the effects of design/operating parameters on the performance and cost can be elucidated. In our presentation, the modeling framework will be presented in detail, including preliminary cost predictions of using this plant configuration under a few selected relevant case studies. This study will be used to identify the major cost drivers, informing further system design and optimization needed to chart the way for a commercially viable pathway.

14 SOLAR ENERGY

Control Parameter Sensitivity Study for Inverter-Based-Resource Dominated Grids: A Small Signal Stability Approach and Framework

The growing adoption of renewable energy is driving the prevalence of inverter-based resources (IBRs) within power grids. Future power grids will integrate both grid-following IBRs (GFM-IBRs) and grid-forming IBRs (GFL-IBRs) alongside synchronous generators. Therefore, it is crucial to perform stability studies that account for all components and especially control interactions related to IBRs. Extensive research has performed to study the IBR-related stability, however, the sensitivity study of IBRs' control parameters on system stability has not been adequately studied yet, especially from a systematic way. Therefore, this paper conducts a small signal stability analysis for a generic grid with multiple types of resources and develops an analytical framework for assessing the sensitivity of control parameters affecting stability margins. To achieve that, the non-autonomous reduced-order non-linear dynamic model is developed for a generic power system with multiple synchronous generator-based resources (SGBRs), GFM-IBRs, and GFL-IBRs. Based on the analytic model, a systematic framework for parametric sensitivity on systems' asymptotic stability is developed. A parameter sensitivity analysis based on eigenvalue methods is proposed. The impact of the droop controllers of GFM-IBRs, PQ-dispatch and the PLL controller of GFL-IBR on the system asymptotic stability is discussed. This sensitivity study is aiming to provide deep insights on control parameters' impact on system stability, and gives direction for parameter tuning in case of instability.

24 POWER TRANSMISSION AND DISTRIBUTION

Overview of IMPACT Data Acquisition System and Data Reduction Process

This report documents the development of the data acquisition system (DAS) and data reduction methodologies for the Irradiated Material Property Accelerated Characterization Test (IMPACT) experiment at the Advanced Test Reactor (ATR). The IMPACT experiment is designed to enable in-pile measurement of thermal conductivity in metallic nuclear fuels, specifically U-10Zr, using an instrumented thermal conductivity probe. The DAS supports both passive temperature monitoring and active thermal interrogation of the probe through controlled AC and DC excitation. Significant modifications to laboratory-scale systems were required to accommodate the higher resistance paths associated with the in-pile application. Custom electronics and relay-controlled measurement sequencing were developed to enable the measurement and sufficient power delivery to the sensing region. A reduced-order, axisymmetric thermal model based on the thermal quadrupoles method is presented to support data interpretation. This model enables efficient evaluation of transient heat transfer behavior and facilitates solution of the inverse problem required to extract thermal properties from measured signals. Multiple boundary condition formulations are discussed to address varying experimental time scales and geometries. Additionally, machine learning techniques are introduced to support data reduction and improve confidence in inverse solutions. Convolutional neural networks are applied to identify the presence of gas gaps and other evolving geometric features that significantly impact thermal response during irradiation. These efforts contribute to the broader integration of digital twin frameworks and real-time modeling capabilities within the Advanced Fuels Campaign.

46 - INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AN

A Physics-Based Digital Twin for Wave Elevation and Seabed Moment Estimation of Offshore Monopiles: Preprint

In this work, we present a proof of concept of a physics-based digital twin for a monopile structure (with overhead inertia) subjected to wave loading. The digital twin is formulated using reduced-order models derived from first principles and combined with a Kalman filter for state estimation. The proposed framework estimates the monopile top motion, the wave elevation, and the section forces and moments along the pile using primarily acceleration measurements at the monopile top. Key innovations include the use of a hydrodynamic shape function to represent distributed wave loading in a compact and computationally efficient manner, and the introduction of a shaping filter to augment the state-space with wave kinematics. Synthetic measurement data are generated using OpenFAST and used as a reference to assess the performance of the digital twin. Results demonstrate that the wave elevation can be accurately reconstructed without direct sea-state measurements as long as the wave regime is inertia-dominated. Under the ideal tested conditions, the total hydrodynamic force and sea-bed bending moment are estimated with relative errors on the order of 1% and correlation coefficients exceeding 96%. Future work will evaluate the estimator's performance under operational uncertainties and more complex loading conditions.

17 WIND ENERGY

Surrogates for Valve-Controlled Pipe Flow: Accelerating Nuclear Reactor Design

Neural surrogate models are developed to replace expensive steady-state RANS CFD simulations for valve-controlled pipe flow in nuclear reactor design. Using parametric CFD data generated with MOOSE Pronghorn across a range of valve geometry and flow conditions, three approaches are compared: a POD-based reduced-order model, a structured UNet on a cylindrical grid, and unstructured models (DeepONet and BiStride MeshGraphNet) on nondimensionalized point clouds. POD achieves the highest accuracy (99%) with fast inference but requires storing all solution snapshots, while the DeepONet and BSMS-GNN both achieve ~89% accuracy at sub-second inference, with the BSMS-GNN offering superior geometric generalizability. These surrogates enable rapid ranking of candidate valve designs and can warm-start CFD solvers to accelerate convergence, supporting agentic design iteration on the Prometheus platform.

42 - ENGINEERING

A Scalable Reduced‐Order Model for the Steady Navier–Stokes Equations

Scaling up new scientific technologies from laboratory to industry often involves demonstrating performance on a larger scale. Computer simulations can accelerate design and predictions in the deployment process, though traditional numerical methods are computationally intractable even for intermediate pilot plant scales. Recently, the component reduced order modeling method has been developed to tackle this challenge by combining projection reduced order modeling and discontinuous Galerkin domain decomposition. However, while many scientific or engineering applications involve nonlinear physics, this method has only been demonstrated for various linear systems. In this work, the component reduced order modeling method is extended to steady Navier–Stokes flow, with application to general nonlinear physics in view. The large‐scale, global domain is decomposed into a combination of small‐scale unit component. Linear subspaces for flow velocity and pressure are identified via proper orthogonal decomposition over sample snapshots collected from each small‐scale unit component. Velocity bases are augmented with a pressure supremizer to satisfy the inf–sup condition for stable pressure prediction. Two different nonlinear reduced order modeling methods are employed and compared for efficient evaluation of nonlinear advection: A third‐order tensor projection operator and the empirical quadrature procedure. The proposed method is demonstrated on the flow over arrays of five different unit objects, achieving a 23‐fold speedup with less than 4% relative error in domains up to 256 times larger than the unit components. Furthermore, a numerical experiment with the pressure supremizer strongly indicates the need for a supremizer for stable pressure prediction. A comparison between the tensorial approach and the empirical quadrature procedure revealed a slight advantage of the empirical quadrature procedure. The framework is compared with an alternating Schwarz‐based reduced‐order approach, demonstrating improved efficiency and robustness for the DG‐based global solver while retaining flexibility for sub‐scale iterative solvers. The method is further extended to a coupled advection–diffusion and Navier–Stokes system, illustrating its applicability to multi‐physics problems and its potential for more general, inter‐coupled nonlinear systems.

42 ENGINEERING

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 spatiotemporal dynamics simulations, such as those in computational modeling of energetic materials (EM). 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. We build upon our previous work on physics-aware recurrent convolutional neural networks (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. Here, 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.

Mathematics and Computing