Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “surrogates”

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 1,261 records · Page 70
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

A high-order Shifted Interface Method for Lagrangian shock hydrodynamics

Here, we present a new method for two-material Lagrangian hydrodynamics, which combines the Shifted Interface Method (SIM) with a high-order Finite Element Method. Our approach relies on an exact (or sharp) material interface representation, that is, it uses the precise location of the material interface. The interface is represented by the zero level-set of a continuous high-order finite element function that moves with the material velocity. This strategy allows to evolve curved material interfaces inside curved elements. By reformulating the original interface problem over a surrogate (approximate) interface, located in proximity of the true interface, the SIM avoids cut cells and the associated problematic issues regarding implementation, numerical stability, and matrix conditioning. Accuracy is maintained by modifying the original interface conditions using Taylor expansions. We demonstrate the performance of the proposed algorithms on established numerical benchmarks in one, two and three dimensions.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Active operator learning with predictive uncertainty quantification for partial differential equations

With the increased prevalence of neural operators being used to provide rapid solutions to partial differential equations (PDEs), understanding the accuracy of model predictions and the associated error levels is necessary for deploying reliable surrogate models in scientific applications. Existing uncertainty quantification (UQ) frameworks employ ensembles or Bayesian methods, which can incur substantial computational costs during both training and inference. Here, we propose a lightweight predictive UQ method tailored for Deep operator networks (DeepONets) that also generalizes to other operator networks. Numerical experiments on linear and nonlinear PDEs demonstrate that the framework’s uncertainty estimates are unbiased and provide accurate out-of-distribution uncertainty predictions with a sufficiently large training dataset. Our framework provides fast inference and uncertainty estimates that can efficiently drive outer-loop analyses that would be prohibitively expensive with conventional solvers. We demonstrate how predictive uncertainties can be used in the context of Bayesian optimization and active learning problems to yield improvements in accuracy and data-efficiency for outer-loop optimization procedures. In the active learning setup, we extend the framework to Fourier Neural Operators (FNO) and describe a generalized method for other operator networks. To enable real-time deployment, we introduce an inference strategy based on precomputed trunk outputs and a sparse placement matrix, reducing evaluation time by more than a factor of five. Our method provides a practical route to uncertainty-aware operator learning in time-sensitive settings.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Efficient data-driven regression for reduced-order modeling of spatial pattern formation

We present an efficient data-driven regression approach for constructing reduced-order models (ROMs) of reaction-diffusion systems exhibiting pattern formation. The ROMs are learned non-intrusively from available training data of physically accurate numerical simulations. The method can be applied to general nonlinear systems through the use of polynomial model form, while not requiring knowledge of the underlying physical model, governing equations, or numerical solvers. The process of learning ROMs is posed as a low-cost least-squares problem in a reduced-order subspace identified via Proper Orthogonal Decomposition (POD). Numerical experiments on classical pattern-forming systems–including the Schnakenberg and Mimura–Tsujikawa models–demonstrate that higher-order surrogate models significantly improve prediction accuracy while maintaining low computational cost. The proposed method provides a flexible, non-intrusive model reduction framework, well suited for the analysis of complex spatio-temporal pattern formation phenomena.

Data-driven modeling↗
DOE OSTI2020

Anion-exchanged and quaternary ammonium functionalized MIL-101-Cr metal-organic framework (MOF) for ReO 4 – /TcO 4 – sequestration from groundwater

There are few effective technologies for the sequestration of highly water-soluble pertechnetate (TcO 4 – ) from contaminated water despite the urgency of environmental and public health concerns. Here, anion exchanged and cetyltrimethylammonium bromide (CTAB) functionalized MIL-101-Cr-NO 3 were investigated for perrhenate (ReO 4 – ), a surrogate of TcO 4 – , sequestration from artificial groundwater. Cl – , I – , and CF 3 SO 3 – exchanged MIL-101-Cr proved more effective at ReO 4 – removal than the parent MIL-101-Cr-F. Compared to the parent framework, CTAB functionalized MIL-101-Cr-NO 3 increased ReO 4 – removal capacity from 39 to 139 mg/g, improved the reaction kinetics from ~30 to <10 min to reach full adsorption capacity and the selectivity for ReO 4 – over competing NO 3 – , CO 3 2– , SO 4 2– , and Cl – . Spectroscopic data indicated that the chemical speciation of Re in the exchanged MIL-101-Cr remained ReO 4 – , indicating synergistic sequestration through both anion exchange and non-ion exchange binding with the positively charged ligand of CTAB. These studies foreshadow potential applications of MOFs for the remediation of 99 TcO 4 – from contaminated environments.

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗
DOE OSTI2025

Influence of temperature, oxygen partial pressure, and microstructure on the high-temperature oxidation behavior of the SiC Layer of TRISO particles

Tristructural isotropic (TRISO)-coated fuel particles are designed for use in high-temperature gas-cooled nuclear reactors, featuring a structural SiC layer that may be exposed to oxygen-rich environments over 1000 °C. Surrogate TRISO particles were tested in 0.2–20 kPa O 2 atmospheres to observe the differences in oxidation behavior. Oxide growth mechanisms remained consistent from 1200–1600 °C for each P O$_2$ , with activation energies of 228 ± 7 kJ/mol for 20 kPa O 2 and 188 ± 8 kJ/mol for 0.2 kPa O 2 . At 1600 °C, kinetic analysis revealed a change in oxide growth mechanisms between 0.2 and 6 kPa O2. In 0.2 kPa O 2 , oxidation produced raised oxide nodules on pockets with nanocrystalline SiC. Oxidation mechanisms were determined using Atom probe tomography. Active SiC oxidation occurred in C-rich grain boundaries with low P O$_2$ , leading to SiO 2 buildup in porous nodules. Here, this phenomenon was not observed at any temperature in 20 kPa O 2 environments.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

A deep learning-based workflow for fast prediction of 3D state variables in geological carbon storage: A dimension reduction approach

Deep learning (DL) models are extensively used as surrogate models for high-fidelity simulations of multiphase fluid flow in porous media at large scales, enabling fast forecasts of the spatial–temporal evolution of three-dimensional (3D) state variables in geological carbon storage (GCS). However, training these models in high-dimensional space remains computationally demanding and prone to overfitting because of limited training data. This paper presents a novel workflow to address these challenges by integrating dimension reduction (DR) methods. Here, the proposed workflow employed pre-trained DR models to extract the latent variables of geological models and state variables and utilized the multi-layer perceptron (MLP) for constructing mapping functions between the input and output variables in latent spaces. Subsequently, the pre-trained reconstruction models converted the MLP-predicted latent state variables to their original high-dimensional form. Furthermore, we proposed a novel strategy for the DR and reconstruction of 3D saturation fields to account for the unique data characteristics of sparsity, nonuniformity, and discontinuity. The proposed strategy applied PCA and inverse PCA for 2D average saturation fields and developed a DL-based 3D reconstruction model, leveraging three 2D average saturation fields as input to produce a 3D saturation field as output. The pre-training of DR and reconstruction models and training of MLP models were conducted on 84 Gulf of Mexico (GoM) simulations and evaluated on 12 testing simulations. Each simulation contained 720 monthly time steps, with the first 360 months as the injection period and the rest as the post-injection period. The proposed workflow, incorporating DR and DL models, accurately predicts the normalized 3D pressure fields, achieving mean square error (MSE) of 2.92 × 10 -7 compared to the ground truth obtained from a full-physics simulator. Furthermore, the proposed strategy outperformed PCA and convolutional autoencoder (CAE) models on 3D saturation fields, resulting in minor workflow prediction errors with an MSE of 2.93 × 10 -5 . The results suggest the proposed workflow provides sufficient predictive fidelity across temporal and spatial scales, and enables a speedup of 160 times compared to the full-physics simulator, facilitating improved decision-making and risk assessment for large-scale GCS management in real-time scenarios.

3D reconstruction model↗
DOE OSTI2022

A process optimization framework for laser direct energy deposition: Densification, microstructure, and mechanical properties of an Fe-Cr alloy

Laser Direct Energy Deposition (DED) is a metal additive manufacturing technique with the ability to fabricate large and complex parts through deposition of metal powders. However, achieving high-density parts and targeted build heights using DED can be challenging due to the large number of highly sensitive process variables. This work proposes a robust fabrication parameter optimization framework to generate process maps for primary parameters in DED, including laser power, scan speed, mass flow rate, hatch spacing, and layer height. Simple single-track experiments were utilized to map out the parameter space, and a combination of geometric criteria for hatch spacing and layer height were proposed to determine parameter sets that achieve both targeted build heights and mitigate porosity formation. Using this framework, specimens with >99 % density and consistent mechanical properties were successfully fabricated over a wide range of process parameters for an Fe-9wt.%Cr (Fe9Cr) alloy, a surrogate for radiation damage-resistant reduced activation ferritic/martensitic (RAFM) steels. Processing these materials using DED is of particular interest in the development of plasma facing components for nuclear fusion applications. The microstructure and mechanical properties of as-printed Fe9Cr were characterized using optical and electron microscopy, X-ray diffraction, and uniaxial tensile tests. As-printed Fe9Cr displayed ~25 % elongation and ultimate tensile strengths of up to 475 MPa which is comparable to similar wrought alloys. Finally, the proposed framework will allow for accelerated DED parameter optimization for novel alloy systems, as well as open the possibility for local microstructure control while simultaneously mitigating defect formation.

42 ENGINEERING↗
DOE OSTI2023

Learning macroscopic internal variables and history dependence from microscopic models

This paper concerns the study of history dependent phenomena in heterogeneous materials in a two-scale setting where the material is specified at a fine microscopic scale of heterogeneities that is much smaller than the coarse macroscopic scale of application. Here, we specifically study a polycrystalline medium where each grain is governed by crystal plasticity while the solid is subjected to macroscopic dynamic loads. The theory of homogenization allows us to solve the macroscale problem directly with a constitutive relation that is defined implicitly by the solution of the microscale problem. However, the homogenization leads to a highly complex history dependence at the macroscale, one that can be quite different from that at the microscale. In this paper, we examine the use of machine-learning, and especially deep neural networks, to harness data generated by repeatedly solving the finer scale model to: (i) gain insights into the history dependence and the macroscopic internal variables that govern the overall response; and (ii) to create a computationally efficient surrogate of its solution operator, that can directly be used at the coarser scale with no further modeling. We do so by introducing a recurrent neural operator (RNO), and show that: (i) the architecture and the learned internal variables can provide insight into the physics of the macroscopic problem; and (ii) that the RNO can provide multiscale, specifically FE 2 , accuracy at a cost comparable to a conventional empirical constitutive relation.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Oxide evolution on the SiC layer of TRISO particles during extended air oxidation

Tristructural isotropic (TRISO) fuel particles have been primarily developed for high-temperature gas-cooled nuclear reactors and can be subjected to oxidizing environments for extended periods in an off-normal accident scenario. Surrogate TRISO fuel particles were oxidized in air at 1,000 or 1,100 °C for up to 120 h. Here, the oxide scale morphology and thickness were studied via scanning electron microscopy, focused ion beam, and atomic force microscopy. TRISO particles oxidized at 1,100 °C exhibited a highly crystalline oxide scale, which led to significant cracking and irregularly shaped closed porosity, whereas those oxidized at 1,000 °C possessed a primarily amorphous oxide scale, which contained small, rounded internal pores and no larger defects. The observed phenomena deviated from the expected behavior based on models for oxide growth on flat-plate and fiber SiC. The oxidation kinetics of TRISO fuel particles in high-temperature air were investigated without mechanically deforming the surface and were analyzed with respect to oxide morphology.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Effect of sucrose on the oxidation-reduction conditions and retention of rhenium during vitrification of low-activity waste

Technetium speciation and solubility during nuclear waste vitrification are affected by oxygen partial pressure in the melt. Thus, reducing agents such as sucrose, which are routinely added to nuclear waste feeds to repress foaming, can significantly affect the volatilization of Tc and increase its retention in glass. Using Re as a surrogate for Tc, we measured the effect of sucrose addition to two low-activity waste melter feeds on the Re retention, oxygen partial pressure, and iron redox as functions of temperature. As expected, Re retention in glass increased as the sucrose content increased. Furthermore, this is attributed to sucrose reaction with nitrates and nitrites in the feed. Destruction of nitrates and nitrites reduces the salt phase fraction and thus likely promotes Re diffusion into the glass-forming melt. In agreement with available literature, we discuss that a modest increase in the sucrose content (compared to current practice) can significantly improve the Tc retention without reducing Fe 2+ to metal.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2023

Microstructural analysis of tristructural isotropic particles in high-temperature steam mixed gas atmospheres

High-temperature gas-cooled reactors (HTGRs) use tristructural isotropic (TRISO) particles embedded in a graphitic matrix material to form the integral fuel element. Potential off-normal reactor conditions for HTGRs include steam ingress with temperatures above 1,000 °C. Fuel element exposure to steam can cause the graphitic matrix material to evolve, forming an atmosphere composed of oxidants and oxidation products and potentially exposing the TRISO particles to these conditions. Investigating the oxidation response of TRISO particles exposed to a mixed gas atmosphere will provide insight into the stability under off-normal conditions. In this study, surrogate TRISO particles were exposed to high temperatures (T = 1,200 °C) in flowing steam (3% < pH 2 O < 21%) and CO (pCO < 1%) to determine the oxidation behavior of the SiC layer when exposed to various mixed gas atmospheres. Scanning electron microscopy, x-ray diffraction, and focused ion beam milling was used to determine the impact of CO and steam on the oxidation behavior of the SiC layer. Therefore, the data presented demonstrates how the SiC layer showed strong oxidation resistance due to limited SiO 2 growth and maintained its structural integrity under these off-normal conditions.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

Phonon modal analysis of thermal transport in ThO 2 with point defects using equilibrium molecular dynamics

Defects can significantly degrade the thermal conductivity of ThO 2 , an advanced nuclear fuel material as well as a surrogate for other fluorite-structured materials. Here, we investigate how point defects in ThO 2 impact phonon mode-resolved thermal transport. By incorporating phonon modes from lattice dynamics, we decompose the trajectory and heat flux to phonon normal mode space and extract key phonon properties, including phonon relaxation times and their contributions to thermal conductivity. We implement two methods. The first method is based on the Green Kubo formalism to resolve the contribution of each phonon mode to thermal conductivity. The second resolves the lifetime of individual phonon modes and the thermal conductivity is calculated using the Boltzmann transport equation within relaxation time approximation. Notably, a lower contribution of acoustic modes is revealed compared to perturbative approaches considering only three-phonon scattering processes. The effects of four types of point defects are evaluated. The strongest impact on a reduction in thermal conductivity is from Th interstitials, followed by Th vacancies. O interstitials/vacancies have a similar impact, albeit smaller than defects on the thorium sublattice. These observations are consistent with previous studies.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Characterization of the microstructure of yttrium hydride under proton irradiation

High moderation per unit volume solid moderator materials like yttrium hydride (YH x ) are necessary for compact nuclear microreactors. However, the phase stability and hydrogen transport processes of YH x under high-temperature irradiation are largely unknown. Proton irradiation was conducted on YH x at 300 °C and 580 °C to 0.2 dpa using 1 MeV or 2 MeV protons in a high-vacuum environment. The hydrogen concentration was determined before and after irradiation using elastic recoil detection analysis, and microstructural evolution was examined via post-irradiation scanning transmission electron microscopy and Raman spectroscopy. Dislocation loops and cavities were observed in all conditions; their distribution was correlated with the bombarding proton energy and ion irradiation temperature. This work revealed that hydrogen retention is proportional to the formation of traps for hydrogen gas atoms and identified pathways for hydrogen release. The relative contributions of bulk or fast diffusion paths, such as grain boundaries, delamination boundaries, and stacking faults are discussed; the primary mechanisms of hydrogen loss are likely based on diffusion, ruling out artefacts of the experimental design. In conclusion, the study suggests proton irradiation may be a strong surrogate to study hydrogen transport in hydride moderator materials under irradiation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

Stoichiometric effects on grain growth in zirconium carbide coatings for high-temperature nuclear fuel

Interest in coated particle fuel for space nuclear propulsion (SNP) has expanded in recent years due to successful demonstrations of the resiliency of the coatings to extreme environments. For SNP applications, the coating layer for the particle design needs to be able to withstand exposure to high temperature hydrogen during operating conditions. ZrC has been proposed as a protective layer, however, it is important to understand the high temperature behavior to ensure adequate protection to this fuel. In this study, surrogate ZrC coated particles were heat treated at 1900 °C up to 300 min, to examine how the microstructure evolves when exposed to high temperature. Scanning electron microscopy and electron backscatter diffraction (EBSD) were conducted to determine grain size and grain boundary character and orientation to determine the degree of change in the ZrC layer post heat treatment. Raman spectroscopy provided insight to understand how the as-fabricated stoichiometry of each sample contributed to the differences in grain growth behavior. Despite the as-fabricated samples showing a similar initial grain size and grain boundary character, the samples with a higher amount of excess carbon exhibited smaller grain areas and slower growth rates when exposed to 1900 °C. In conclusion, this investigation details the as-fabricated microstructure of the ZrC layer, specifically grain size, evolved under high temperature as this can impact the performance of the fuel under operating conditions.

EBSD↗
DOE OSTI2026

FLiBe compatibility of fusion-relevant materials

There is current interest in a 2LiF-BeF2 (FLiBe) molten salt blanket for fusion applications, however, very little compatibility data are available for fusion relevant materials. Initial static capsule testing results for 500 h at 550°-750 °C are presented for V-4Cr-4Ti, Fe-8Cr-2W (alloy F82H) compared to type 316H stainless steel and monolithic SiC as a surrogate for SiC/SiC composites. In addition, sintered Be12Ti specimens with varying additions of Be (0–20%) and a Be2C specimen were exposed with F82H coupons at 650 °C in individual Mo capsules. The V-4Cr-4Ti and SiC specimens exposed at 650 °C had small mass changes. Steel specimens showed increasing mass losses with increasing exposure temperature. However, when Be12Ti-Be specimens were included in a capsule with an F82H coupon, increasing F82H mass gains were observed with increasing Be content due to the formation of a Fe-Be surface layer. No layer and no mass gain were observed for the F82H specimen tested with Be2C.

Pint, Bruce [Oak Ridge National Laboratory (ORNL) ↗
DOE OSTI2025

MOOSE ProbML: Parallelized probabilistic machine learning and uncertainty quantification for computational energy applications

Here, this paper presents the development and demonstration of massively parallel probabilistic machine learning (ML) and uncertainty quantification (UQ) capabilities within the Multiphysics Object-Oriented Simulation Environment (MOOSE), an open-source computational platform for parallel finite element and finite volume analyses. In addressing the computational expense and uncertainties inherent in complex multiphysics simulations, this paper integrates Gaussian process (GP) variants, active learning, Bayesian inverse UQ, adaptive forward UQ, Bayesian optimization, evolutionary optimization, and Markov chain Monte Carlo (MCMC) within MOOSE. It also elaborates on the interaction among key MOOSE systems — Sampler, MultiApp, Reporter, and Surrogate — in enabling these capabilities. The modularity offered by these systems enables development of a multitude of probabilistic ML and UQ algorithms in MOOSE. Example code demonstrations include parallel active learning and parallel Bayesian inference via active learning. The impact of these developments is illustrated through five applications relevant to computational energy applications: UQ of nuclear fuel fission product release, using parallel active learning Bayesian inference; very rare events analysis in nuclear microreactors using active learning; advanced manufacturing process modeling using multi-output GPs (MOGPs) and dimensionality reduction; fluid flow using deep GPs (DGPs); and tritium transport model parameter optimization for fusion energy, using batch Bayesian optimization. These capabilities are part of the MOOSE framework.

97 - MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Computation of conventional and alternative jet fuel sensitivity to lean blowout

Large Eddy Simulations (LES) are performed to compute the sensitivity of a conventional (A-2) and an alternate bio-jet (C-1) fuel to Lean Blowout (LBO). A realistic aviation gas turbine engine combustor configuration is considered. Reliable experimental LBO data and OH* chemiluminescence data for the conventional and alternate jet fuel in the combustor configuration have recently become available. The present work utilizes a highly automated, on-the-fly meshing strategy, along with adaptive mesh refinement, to demonstrate the feasibility of capturing the realistic combustion processes. A Lagrangian framework, with initial conditions specified using measurements of spray statistics, is used to model the fuel spray. Newly developed compact reaction mechanisms based on fuel surrogates are validated for the A-2 and the C-1 fuels. The compact reaction mechanisms are implemented using a detailed finite rate chemistry solver. Spray statistics computed by the present LES simulations compare well with available measurements at stable flame conditions near the lean blowout limit. The computed shape of the stable flame as represented by line integrated OH concentrations compares well with the experimental OH* chemiluminescence data. Lean blowout is reached by gradually decreasing the fuel flow rate in the computations, similar to that in the experiments. The results of the LES simulations effectively capture the fuel composition effects and estimate the sensitivity of the LBO limits to the fuel type. The computed trends in LBO limits agree within engineering accuracy with the experimental results for conventional and alternative aviation fuels. The methodology for predicting the fuel composition effects on the lean blowout limits in a fully resolved realistic, complex combustor is established for the first time.

Adaptive mesh refinement↗