Search NASA⌕ Search

SEARCH · Search NASA

Results for “diffusion surrogate”

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
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Effectiveness of denoising diffusion probabilistic models for fast and high-fidelity whole-event simulation in high-energy heavy-ion experiments

Artificial intelligence (AI) generative models, such as generative adversarial networks (GANs), variational autoencoders, and normalizing flows, have been widely used and studied as efficient alternatives for traditional scientific simulations. However, they have several drawbacks, including training instability and inability to cover the entire data distribution, especially for regions where data are rare. This is particularly challenging for whole-event, full-detector simulations in high-energy heavy-ion experiments, such as sPHENIX at the Relativistic Heavy Ion Collider and Large Hadron Collider experiments, where thousands of particles are produced per event and interact with the detector. This work investigates the effectiveness of denoising diffusion probabilistic models (DDPMs) as an AI-based generative surrogate model for the sPHENIX experiment that includes the heavy-ion event generation and response of the entire calorimeter stack. DDPM performance in sPHENIX simulation data is compared with a popular rival, GANs. Results show that both DDPMs and GANs can reproduce the data distribution where the examples are abundant (low-to-medium calorimeter energies). Nonetheless, DDPMs significantly outperform GANs, especially in high-energy regions where data are rare. Additionally, DDPMs exhibit superior stability compared to GANs. The results are consistent between both central and peripheral centrality heavy-ion collision events. Moreover, DDPMs offer a substantial speedup of approximately a factor of 100 compared to the traditional Geant4 simulation method.

42 ENGINEERING↗
DOE OSTI2024

Bayesian reduced-order deep learning surrogate model for dynamic systems described by partial differential equations

We propose a reduced-order deep-learning surrogate model for dynamic systems described by time-dependent partial differential equations. This method employs space–time Karhunen–Loève expansions (KLEs) of the state variables and space-dependent KLEs of space-varying parameters to identify the reduced (latent) dimensions. Subsequently, a deep neural network (DNN) is used to map the parameter latent space to the state variable latent space. An approximate Bayesian method is developed for uncertainty quantification (UQ) in the proposed KL-DNN surrogate model. The KL-DNN method is tested for the linear advection–diffusion and nonlinear diffusion equations, and the Bayesian approach for UQ is compared with the deep ensembling (DE) approach, commonly used for quantifying uncertainty in DNN models. It was found that the approximate Bayesian method provides a more informative distribution of the PDE solutions in terms of the coverage of the reference PDE solutions (the percentage of nodes where the reference solution is within the confidence interval predicted by the UQ methods) and log predictive probability. The DE method is found to underestimate uncertainty and introduce bias. For the nonlinear diffusion equation, we compare the KL-DNN method with the Fourier Neural Operator (FNO) method and find that KL-DNN is 10% more accurate and needs less training time than the FNO method.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Surrogate Hessian accelerated structural optimization for stochastic electronic structure theories

In this work, we present an efficient energy-based method for structural optimization with stochastic electronic structure theories, such as diffusion quantum Monte Carlo (DMC). This method is based on robust line-search energy minimization in reduced parameter space, exploiting approximate but accurate Hessian information from a surrogate theory, such as density functional theory. The surrogate theory is also used to characterize the potential energy surface, allowing for simple but reliable ways to maximize statistical efficiency while retaining controllable accuracy. We demonstrate the method by finding the minimum DMC energy structures of the selected flake-like aromatic molecules, such as benzene, coronene, and ovalene, represented by 2, 6, and 19 structural parameters, respectively. In each case, the energy minimum is found within two parallel line-search iterations. The method is near-optimal for a line-search technique and suitable for a broad range of applications. It is easily generalized to any electronic structure method where forces and stresses are still under active development and implementation, such as diffusion Monte Carlo, auxiliary-field Monte Carlo, and stochastic configuration interaction, as well as deterministic approaches such as the random-phase approximation. Accurate and efficient means of geometry optimization could shed light on a broad class of materials and molecules, showing high sensitivity of induced properties to structural variables.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Learning effective stochastic differential equations from microscopic simulations: Linking stochastic numerics to deep learning

We identify effective stochastic differential equations (SDEs) for coarse observables of fine-grained particle- or agent-based simulations; these SDEs then provide useful coarse surrogate models of the fine scale dynamics. We approximate the drift and diffusivity functions in these effective SDEs through neural networks, which can be thought of as effective stochastic ResNets. The loss function is inspired by, and embodies, the structure of established stochastic numerical integrators (here, Euler–Maruyama and Milstein); our approximations can thus benefit from backward error analysis of these underlying numerical schemes. They also lend themselves naturally to “physics-informed” gray-box identification when approximate coarse models, such as mean field equations, are available. Existing numerical integration schemes for Langevin-type equations and for stochastic partial differential equations can also be used for training; we demonstrate this on a stochastically forced oscillator and the stochastic wave equation. Our approach does not require long trajectories, works on scattered snapshot data, and is designed to naturally handle different time steps per snapshot. We consider both the case where the coarse collective observables are known in advance, as well as the case where they must be found in a data-driven manner.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Property Measurements of NaCl-UCl 3 and LiF-NaF-KF Molten Salts Doped with Surrogate Fission Products

Knowing thermophysical and thermochemical property values of salts over the expected range of operating temperatures is essential to modeling and simulation efforts that support the commercialization of molten salt reactor (MSR) technologies. Salt properties being measured at Argonne include thermal transitions, phase behavior, heat capacity, density, surface tension, volumetric thermal expansion, thermal diffusivity, thermal conductivity, and viscosity. The properties of several salts of interest including NaCl-UCl 3 and LiF-NaF-KF (FLiNaK) have been measured at Argonne and reported previously. These property measurements are suitable for use in evaluating reactor performance during startup and the early operating life of the reactor. The physical and chemical behavior of the fuel salt is altered by the buildup of fission products and the resulting changes in heat transfer and other properties must be understood by reactor developers. This report summarizes melting behavior, heat capacity, and thermal diffusivity values that were measured for mixtures of FLiNaK and NaCl-UCl 3 salts that have been doped with surrogate fission products. Additional heat capacity measurements were also performed on a separate mixture of FLiNaK without dopants. Measured properties are suitable for incorporation into the Molten Salt Thermal Properties Database–Thermochemical (MSTDB–TC) and the Molten Salt Thermal Properties Database–Thermophysical (MSTDB–TP) to support the development of MSRs. Salts were prepared by doping eutectic mixtures of NaCl-UCl 3 and FLiNaK with surrogate fission products. These salts were prepared for use in salt spill experiments being performed at Argonne and were subsequently used for thermal property measurements. Thermal transitions and heat capacities were measured by using differential scanning calorimetry (DSC) and thermal diffusivity was measured by using laser flash analysis (LFA) at temperatures spanning the typical MSR operating range. The measured property values of the doped salts were compared to measured property values of salts without fission products to quantify the effects of the doped fission products.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2020

A buoyancy–shear–drag-based turbulence model for Rayleigh–Taylor, reshocked Richtmyer–Meshkov, and Kelvin–Helmholtz mixing

A phenomenological turbulence model for Rayleigh–Taylor, reshocked Richtmyer–Meshkov, and Kelvin–Helmholtz instability-induced mixing is developed using a general buoyancy–shear–drag model. Analytic solutions to the simplified model equations corresponding to each instability are derived separately, which are then used to calibrate the model coefficients to predict specific values of the mixing layer growth parameters and exponents. The buoyancy-shear–drag equations for the bubble and spike mixing layer widths are then solved numerically, and a turbulent diffusivity (or viscosity) is constructed dimensionally as the product of the mixing layer width, h, and its time-derivative, dh/dt. Surrogate turbulent fields are then constructed by multiplying a presumed, approximate self-similar spatial profile by appropriate functions of h and dh/dt. Using several simplifying approximations, the turbulent diffusion equations satisfied by the mean mass fraction and mean shear velocity are solved analytically in a reference frame moving with the mean advection velocity, and it is shown that the mean fields evolve in space and time as expected. The explicit modeling and solution of turbulent transport equations (e.g., for the turbulent kinetic energy and its dissipation rate or a lengthscale) are not required. By using separate equations for the bubble and spike evolution, the model naturally captures the asymmetry induced by an increasing density contrast between the heavy and light fluids. The model also includes molecular dissipation and diffusion, and therefore, can describe transition to fully-developed turbulence if the initial turbulent diffusivity/viscosity is much smaller than the molecular diffusivity/viscosity. This model is applied to constant-acceleration Rayleigh–Taylor, impulsively reshocked Richtmyer–Meshkov, and variable-density Kelvin–Helmholtz mixing layers to demonstrate its utility. It is shown that the numerical solutions of the model calibrated using specific values of the instability growth parameters and exponents: (1) produces mixing layer widths in agreement with the expected self-similar growth power laws; (2) gives spatiotemporal profiles of turbulent fields that are expected and consistent with previous results; (3) predicts the expected power-law growths and decays of the spatially-integrated turbulent fields, and; (4) gives spatiotemporal profiles of the mean fields that are expected and consistent with previous results. This relatively simple model can be implemented numerically, and is expected to be useful for fundamental as well as astrophysical and inertial confinement fusion applications.

42 ENGINEERING↗
DOE OSTI2023

Towards fast, accurate predictions of RF simulations via data-driven modeling: Forward and lateral models

Three machine learning techniques (multilayer perceptron, random forest, and Gaussian process) provide fast surrogate models for lower hybrid current drive (LHCD) simulations. A single GENRAY/CQL3D simulation without radial diffusion of fast electrons requires several minutes of wall-clock time to complete, which is acceptable for many purposes, but too slow for integrated modeling and real-time control applications. More accurate simulations with fast electron diffusion are even slower, requiring multiple hours of run time with parallel processing. The machine learning models use a database of 16,000+ GEN-RAY/CQL3D simulations for training, validation, and testing. Latin hypercube sampling methods implemented in πScope ensure that the database covers the range of 9 input parameters (n e0 , T e0 , I p , B t , R 0 , n ∥︀ , Z e f f , V loop , P LHCD ) with sufficient density in all regions of parameter space. The surrogate models reduce the computation time from minutes-hours to ms with high accuracy across the input parameter space. Data-driven surrogate models also allow for solving inverse and “lateral” problems. A surrogate model for the inverse problem maps from a desired current drive or power deposition profile to a set of input parameters that would result in such a profile, while a surrogate model for the lateral problem maps from a measured experimental quantity such as hard x-ray emission to a current drive or power deposition profile. In conclusion, the πScope database creation workflow is flexible and applicable to other RF simulation codes such as TORIC.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2022

Towards Fast and Accurate Predictions of Radio Frequency Power Deposition and Current Profile via Data-driven Modeling

Three machine learning techniques (multilayer perceptron, random forest, and Gaussian process) provide fast surrogate models for lower hybrid current drive (LHCD) simulations. A single GENRAY/CQL3D simulation without radial diffusion of fast electrons requires several minutes of wall-clock time to complete, which is acceptable for many purposes, but too slow for integrated modeling and real-time control applications. The machine learning models use a database of 16,000+ GENRAY/CQL3D simulations for training, validation, and testing. Latin hypercube sampling methods ensure that the database covers the range of 9 input parameters ($n_{e0}$, $T_{e0}$, $I_p$, $B_t$, $R_0$, $n_{||}$, $Z_{eff}$, $V_{loop}$, $P_{LHCD}$) with sufficient density in all regions of parameter space. The surrogate models reduce the inference time from minutes to ~ms with high accuracy across the input parameter space.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2022

Towards fast and accurate predictions of radio frequency power deposition and current profile via data-driven modelling: applications to lower hybrid current drive

Three machine learning techniques (multilayer perceptron, random forest and Gaussian process) provide fast surrogate models for lower hybrid current drive (LHCD) simulations. A single GENRAY/CQL3D simulation without radial diffusion of fast electrons requires several minutes of wall-clock time to complete, which is acceptable for many purposes, but too slow for integrated modelling and real-time control applications. The machine learning models use a database of more than 16 000 GENRAY/CQL3D simulations for training, validation and testing. Latin hypercube sampling methods ensure that the database covers the range of nine input parameters ( $n_{e0}$ , $T_{e0}$ , $I_p$ , $B_t$ , $R_0$ , $n_{\|}$ , $Z_{{\rm eff}}$ , $V_{{\rm loop}}$ and $P_{{\rm LHCD}}$ ) with sufficient density in all regions of parameter space. The surrogate models reduce the inference time from minutes to $\sim$ ms with high accuracy across the input parameter space.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

CaloTrilogy: Toward a Breakthrough in One-Step, End-to-End, Physics-Guided Shower Generation for Modern Calorimeters

High-precision calorimeter simulation at current and future colliders imposes rapidly growing computational demands, motivating the development of machine-learning surrogates for traditional Monte Carlo tools such as Geant4. Flow matching and diffusion-based generative models have become leading approaches for high-dimensional fast simulation because of their sample quality, but typically require ${\cal O}(100)$ function evaluations at inference and often rely on auxiliary networks to constrain global observables, compromising streamlined end-to-end generation. We introduce a unified framework that improves the balance between speed, shower quality, and physics fidelity. The method combines: (i) an average velocity field integrator that enables sampling in one or a few evaluations; (ii) a learned generative prior in shower space, constructed from data rather than random noise; and (iii) physics-guided loss terms that impose inductive biases on key observables during training. These elements are training time regularizers, preserving end-to-end inference with no additional cost. With only one or a few evaluation steps, the model achieves shower quality competitive with state-of-the-art flow and diffusion approaches, tested on several public high granularity calorimeter datasets. The results demonstrate inter-layer shower structure consistent with the underlying physics, providing a strong candidate for future fast simulation workflows.

Jiang, Cheng [Edinburgh U.]↗
DOE OSTI2026

D$e^+e^-$ffusion: Capturing the Beam-Beam Physics of $e^+e^-$ Collisions with Diffusion Models

Beam-induced backgrounds at high-luminosity $e^+e^-$ colliders, such as the FCC-ee, are dominated by incoherent pair creation (IPC), and require computationally expensive simulations with dedicated Monte Carlo (MC) event generators. Reliable detector and machine-detector interface studies necessitate event samples that are several orders of magnitude larger than what is practically attainable with existing MC. To alleviate this bottleneck, we present D$e^+e^-$ffusion, a denoising diffusion probabilistic model that operates as a permutation-equivariant, set-valued surrogate for fast IPC simulation. Trained on a small GuineaPig++ sample, D$e^+e^-$ffusion faithfully reproduces the marginal and joint kinematic, angular, and positional distributions of all three IPC production processes. In addition, we assess the fidelity at the detector level by propagating both Geant4 and D$e^+e^-$ffusion events through a Geant4 simulation of the CLD vertex detector and by training a transformer-based two-sample classifier; the classifier achieves an area under the ROC curve of $0.553 \pm 0.016$. The trained model generates events nearly four orders of magnitude faster than Geant4, paving the way for a fast-simulation surrogate for FCC-ee design studies.

Chahine, Antonio [Imperial Coll., London]↗
DOE OSTI2025

Surrogate modeling of Cellular-Potts agent-based models as a segmentation task using the U-Net neural network architecture

The Cellular-Potts model is a powerful and ubiquitous framework for developing computational models for simulating complex multicellular biological systems. Cellular-Potts models (CPMs) are often computationally expensive due to the explicit modeling of interactions among large numbers of individual model agents and diffusive fields described by partial differential equations (PDEs). In this work, we develop a convolutional neural network (CNN) surrogate model using a U-Net architecture that accounts for periodic boundary conditions. We use this model to accelerate the evaluation of a mechanistic CPM previously used to investigate in vitro vasculogenesis. The surrogate model was trained to predict 100 computational steps ahead (Monte-Carlo steps, MCS), accelerating simulation evaluations by a factor of 562 times compared to single-core CPM code execution on CPU. Over short timescales of up to 3 recursive evaluations, or 300 MCS, our model captures the emergent behaviors demonstrated by the original Cellular-Potts model such as vessel sprouting, extension and anastomosis, and contraction of vascular lacunae. This approach demonstrates the potential for deep learning to serve as a step toward efficient surrogate models for CPM simulations, enabling faster evaluation of computationally expensive CPM simulations of biological processes.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Combustion characteristics and detailed simulations of surrogates for a Tier II gasoline certification fuel

An experimental and numerical study of combustion of a gasoline certification fuel (‘indolene’), and four (S4) and five (S5) component surrogates for it, is reported for the configurations of an isolated droplet burning with near spherical symmetry in the standard atmosphere, and a single cylinder engine designed for advanced compression ignition of pre-vaporized fuel. The intent was to compare performance of the surrogate for these different combustion configurations and to assess the broader applicability of the kinetic mechanism and property database for the simulations. A kinetic mechanism comprised of 297 species and 16797 reactions was used in the simulations that included soot chemistry and accounted for unsteady transport, liquid diffusion inside the droplet, radiative heat transfer, and variable properties. The droplet data showed a clear preference for the S5 surrogate in terms of burning rate. The simulations showed generally very good agreement with measured droplet, flame, and soot shell diameters. Measurements of combustion timing, in-cylinder pressure, and mass-averaged gas temperature were also well predicted with a slight preference for the S5 surrogate. Preferential vaporization was not evidenced from the evolution of droplet diameter but was clearly revealed in simulations of the evolution of mixture fractions inside the droplets. The influence of initial droplet diameter (Do) on droplet burning was strong, with S5 burning rates decreasing with increasing Do due to increasing radiation losses from the flame. Flame extinction was predicted for Do =3.0 mm as a radiative loss mechanism but not predicted for smaller Do for the conditions of the simulations.

Guo, Songtao↗
DOE OSTI2022

Combustion characteristics and detailed simulations of surrogates for a Tier II gasoline certification fuel

An experimental and numerical study of combustion of a gasoline certification fuel (‘indolene’), and four (S4) and five (S5) component surrogates for it, is reported for the configurations of an isolated droplet burning with near spherical symmetry in the standard atmosphere, and a single cylinder engine designed for advanced compression ignition of pre-vaporized fuel. The intent was to compare performance of the surrogate for these different combustion configurations and to assess the broader applicability of the kinetic mechanism and property database for the simulations. A kinetic mechanism comprised of 297 species and 16,797 reactions was used in the simulations that included soot formation and evolution, and accounted for unsteady transport, liquid diffusion inside the droplet, radiative heat transfer, and variable properties. The droplet data showed a clear preference for the S5 surrogate in terms of burning rate. The simulations showed generally very good agreement with measured droplet, flame, and soot shell diameters. Measurements of combustion timing, in-cylinder pressure, and mass-averaged gas temperature were also well predicted with a slight preference for the S5 surrogate. Preferential vaporization was not evidenced from the evolution of droplet diameter but was clearly revealed in simulations of the evolution of mixture fractions inside the droplets. Here, the influence of initial droplet diameter (D o ) on droplet burning was strong, with S5 burning rates decreasing with increasing D o due to increasing radiation losses from the flame. Flame extinction was predicted for D o =3.0 mm as a radiative loss mechanism but not predicted for smaller D o for the conditions of the simulations.

42 ENGINEERING↗
DOE OSTI2022

Fusion RF Modeling Machine Learning (FusionML_RF) v1.0

FusionML_RF consists of multiple codes and trained machine learning (ML) models that perform low-cost output modeling from the Genray-CQL3D. Three machine learning techniques (multilayer perceptron, random forest, and Gaussian process) provide fast surrogate models for lower hybrid current drive (LHCD) simulations. For example, completing a single GENRAY/CQL3D simulation without radial diffusion of fast electrons requires several minutes of wall-clock time. On the other hand, these ML models achieve ~ms of inference time with high accuracy across the input parameter space. This software collection consists of multiple components. (1) codes that use ML methods and precomputed Genray-CQL3D simulation output to build regression models that enable approximate computations of Genray-CLQ3D outputs from arbitrary but physically meaningful input parameters (surrogate modeling); (2) three trained models created by the team, using a database of 16,000+ GENRAY/CQL3D simulations, to study the performance of ML models for surrogate modeling; (3) codes that load the trained models and simulation data, and then compute mean squared error between the models' predictions and the ground truth of simulation output data. This collection is being made available in conjunction with a scientific publication about the work to promote reusability and provide an artifact of the scientific work.

Bai, Zhe↗
DOE OSTI2025

Defect Diffusion Graph Neural Networks for Materials Discovery in High-Temperature Energy Applications

Here, the migration of crystallographic defects dictates material properties and performance for a plethora of technological applications. Density functional theory (DFT)-based nudged elastic band (NEB) calculations are a powerful computational technique for predicting defect migration activation energy barriers, yet they become prohibitively expensive for high-throughput screening of defect diffusivities. Without introducing hand-crafted (i.e., chemistry- or structure-specific) descriptors, we propose a generalized deep learning approach to train surrogate models for NEB energies of vacancy migration by hybridizing graph neural networks with transformer encoders and simply using pristine host structures as input. With sufficient training data, computationally efficient and simultaneous inference of vacancy defect thermodynamics and migration activation energies can be obtained to compute temperature-dependent vacancy diffusivities and to down-select candidates for more thorough DFT analysis or experiments. Thus, as we specifically demonstrate for potential water-splitting materials, candidates with desired defect thermodynamics, kinetics, and host stability properties can be more rapidly targeted from open-source databases of experimentally validated or hypothetical materials.

14 SOLAR ENERGY↗
DOE OSTI2022

Manifold Learning-Based Polynomial Chaos Expansions for High-Dimensional Surrogate Models

In this work we introduce a manifold learning-based method for uncertainty quantification (UQ) in systems describing complex spatiotemporal processes. Our first objective is to identify the embedding of a set of high-dimensional data representing quantities of interest of the computational or analytical model. For this purpose, we employ Grassmannian diffusion maps, a two-step nonlinear dimension reduction technique which allows us to reduce the dimensionality of the data and identify meaningful geometric descriptions in a parsimonious and inexpensive manner. Polynomial chaos expansion is then used to construct a mapping between the stochastic input parameters and the diffusion coordinates of the reduced space. An adaptive clustering technique is proposed to identify an optimal number of clusters of points in the latent space. The similarity of points allows us to construct a number of geometric harmonic emulators which are finally utilized as a set of inexpensive pretrained models to perform an inverse map of realizations of latent features to the ambient space and thus perform accurate out-of-sample predictions. Thus, the proposed method acts as an encoder-decoder system which is able to automatically handle very high-dimensional data while simultaneously operating successfully in the small-data regime. The method is demonstrated on two benchmark problems and on a system of advection-diffusion-reaction equations which model a first-order chemical reaction between two species. In all test cases, the proposed method is able to achieve highly accurate approximations which ultimately lead to the significant acceleration of UQ tasks.

42 ENGINEERING↗
DOE OSTI2024

Property Measurements of LiF-NaF-KF Molten Salts Doped with Surrogate Fission Products

Measurements of molten salt properties including transition temperatures, phase behavior, heat capacity, density, volumetric thermal expansion, surface tension, viscosity thermal diffusivity, thermal conductivity, and vapor pressure are performed at Argonne. The properties of several fuel and coolant salts including eutectic LiF-NaF-KF (FLiNaK) have been measured. These properties are suitable for use in evaluating reactor performance during startup and early operating conditions.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗