Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Computer model”

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 1,603 records · Page 89
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

SGD-Net: Efficient Model-Based Deep Learning with Theoretical Guarantees

Deep unfolding networks have recently gained popularity for solving imaging inverse problems. However, the computational and memory complexity of data-consistency layers within traditional deep unfolding networks scales with the number of measurements, limiting their applicability to large-scale imaging inverse problems. We propose SGD-Net as a new methodology for improving the efficiency of deep unfolding through stochastic approximations of the data-consistency layers. Our theoretical analysis shows that SGD-Net can be trained to approximate batch deep unfolding networks to an arbitrary precision. Our simulations on intensity diffraction tomography and sparse-view computed tomography show that SGD-Net can match the performance of the traditional batch network at a fraction of training and testing complexity.Deep unfolding networks have recently gained popularity for solving imaging inverse problems. However, the computational and memory complexity of data-consistency layers within traditional deep unfolding networks scales with the number of measurements, limiting their applicability to large-scale imaging inverse problems. We propose SGD-Net as a new methodology for improving the efficiency of deep unfolding through stochastic approximations of the data-consistency layers. Our theoretical analysis shows that SGD-Net can be trained to approximate batch deep unfolding networks to an arbitrary precision. Our simulations on intensity diffraction tomography and sparse-view computed tomography show that SGD-Net can match the performance of the traditional batch network at a fraction of training and testing complexity.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Accelerating Scientific Computing in the Post-Moore’s Era

Novel uses of graphical processing units for accelerated computation revolutionized the field of high-performance scientific computing by providing specialized workflows tailored to algorithmic requirements. As the era of Moore’s law draws to a close, many new non–von Neumann processors are emerging as potential computational accelerators, including those based on the principles of neuromorphic computing, tensor algebra, and quantum information. While development of these new processors is continuing to mature, the potential impact on accelerated computing is anticipated to be profound. We discuss how different processing models can advance computing in key scientific paradigms: machine learning and constraint satisfaction. Significantly, each of these new processor types utilizes a fundamentally different model of computation, and this raises questions about how to best use such processors in the design and implementation of applications. While many processors are being developed with a specific domain target, the ubiquity of spin-glass models and neural networks provides an avenue for multi-functional applications. Furthermore, this also hints at the infrastructure needed to integrate next-generation processing units into future high-performance computing systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Thermal and dynamic partition of dumbbell interstitials in complex concentrated alloys

Complex concentrated alloys (CCAs) are promising candidates for applications in extreme conditions, such as irradiation where interstitial mediated diffusion is important. In CCAs with N principal elements, $\frac{N(N+1)}{2}$ types of dumbbell interstitials exist. Currently, there is no way to predict the thermal partition (fractional concentration at equilibrium) and the dynamic partition (fractional time an interstitial spends during diffusion) of each type of dumbbell interstitial. To mitigate this issue, this work proposes a theoretical model for computing the equilibrium concentrations and thermal partition of dumbbell interstitials in CCAs and validates the model using grand canonical Monte Carlo simulations. Lattice kinetic Monte Carlo simulations show that the thermal partition is equivalent to the dynamic partition, and both are governed by composition and formation energies of dumbbells. As a result, the model proposed provides a foundation for understanding radiation enhanced diffusion and induced segregation in CCAs under irradiation.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Modern Monte Carlo methods for efficient uncertainty quantification and propagation: A survey

Uncertainty quantification (UQ) includes the characterization, integration, and propagation of uncertainties that result from stochastic variations and a lack of knowledge or data in the natural world. Monte Carlo (MC) method is a sampling-based approach that has widely used for quantification and propagation of uncertainties. However, the standard MC method is often time-consuming if the simulation-based model is computationally intensive. This article gives an overview of modern MC methods to address the existing challenges of the standard MC in the context of UQ. Specifically, multilevel Monte Carlo (MLMC) extending the concept of control variates achieves a significant reduction of the computational cost by performing most evaluations with low accuracy and corresponding low cost, and relatively few evaluations at high accuracy and corresponding high cost. Multifidelity Monte Carlo (MFMC) accelerates the convergence of standard Monte Carlo by generalizing the control variates with different models having varying fidelities and varying computational costs. Multimodel Monte Carlo method (MMMC), having a different setting of MLMC and MFMC, aims to address the issue of UQ and propagation when data for characterizing probability distributions are limited. Multimodel inference combined with importance sampling is proposed for quantifying and efficiently propagating the uncertainties resulting from small data sets. All of these three modern MC methods achieve a significant improvement of computational efficiency for probabilistic UQ, particularly uncertainty propagation. Finally, an algorithm summary and the corresponding code implementation are provided for each of the modern MC methods. The extension and application of these methods are discussed in detail.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Organic Photovoltaic Efficiency Predictor: Data-Driven Models for Non-Fullerene Acceptor Organic Solar Cells

In the design of organic solar cells, there has been a need for materials with high power conversion efficiencies. Scharber’s model is commonly used to predict efficiency; however, it exhibits poor performance with new non-fullerene acceptor (NFA) devices, since it was designed for fullerene-based devices. In this work, an empirical model is proposed that can be a more accurate alternative for NFA organic solar cells. Additionally, many screening studies use computationally expensive methods. A model based on using semiempirical simplified time-dependent density functional theory (sTD-DFT) as an alternative method can accelerate the calculations and yield a similar accuracy. The models presented in this paper, termed organic photovoltaic efficiency predictor (OPEP) models, have shown significantly lower errors than previous models, with OPEP/B3LYP yielding errors of 1.53% and OPEP/sTD- DFT of 1.55%. As a result, the proposed computational models can be used for the fast and accurate screening of new high-efficiency NFAs/donor pairs.

14 SOLAR ENERGY↗
DOE OSTI2024

Leveraging regional mesh refinement to simulate future climate projections for California using the Simplified Convection-Permitting E3SM Atmosphere Model Version 0

Abstract. The spatial heterogeneity related to complex topography in California demands high-resolution (< 5 km) modeling, but global convection-permitting climate models are computationally too expensive to run multi-decadal simulations. We developed a 3.25 km California climate modeling framework by leveraging regional mesh refinement (CARRM) using the U.S. Department of Energy (DOE)'s global Simple Cloud-Resolving E3SM Atmosphere Model (SCREAM) version 0. Four 5-year time periods (2015–2020, 2029–2034, 2044–2049, and 2094–2099) were simulated by nudging CARRM outside California to 1° coupled simulation of E3SMv1 under the Shared Socioeconomic Pathways (SSP)5-8.5 future scenario. The 3.25 km grid spacing adds considerable value to the prediction of the California climate changes, including more realistic high temperatures in the Central Valley and much improved spatial distributions of precipitation and snowpack in the Sierra Nevada and coastal stratocumulus. Under the SSP5-8.5 scenario, CARRM simulation predicts widespread warming of 6–10 °C over most of California, a 38 % increase in statewide average 30 d winter–spring precipitation, a near-complete loss of the alpine snowpack, and a sharp reduction in shortwave cloud radiative forcing associated with marine stratocumulus by the end of the 21st century. We note a climatological wet precipitation bias for the CARRM and discuss possible reasons. We conclude that SCREAM RRM is a technically feasible and scientifically valid tool for climate simulations in regions of interest, providing an excellent bridge to global convection-permitting simulations.

58 GEOSCIENCES↗
DOE OSTI2019

Porting a 3D seismic modeling code (SW4) to CORAL machines

Seismic waves fourth order (SW4) solves the seismic wave equations on Cartesian and curvilinear grids using large compute clusters with O (100,000) cores. In this article we discuss the porting of SW4 to run on the CORAL architecture using the RAJA performance portability abstraction layer. The performances of key kernels using RAJA and CUDA are compared to estimate the performance penalty of using the portability abstraction layer. Code changes required for efficiency on GPUs and minimizing time spent in Message Passing Interface (MPI) are discussed. This article describes a path for efficiently porting large code bases to GPU-based machines while avoiding the pitfalls of a new architecture in the early stages of its deployment. Current bottlenecks in the code are discussed along with possible architectural or software mitigations. SW4 runs 28× faster on one 4-GPU CORAL node than on a CTS-1 node (Dual Intel Xeon E5-2695 v4). SW4 is now in routine use on problems of unprecedented resolution (203 billion grid points) and scale on 1,200 nodes of Summit.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

ChemoGraph: Interactive Visual Exploration of the Chemical Space

Exploratory analysis of the chemical space is an important task in the field of cheminformatics. For example, in drug discovery research, chemists investigate sets of thousands of chemical compounds in order to identify novel yet structurally similar synthetic compounds to replace natural products. Manually exploring the chemical space inhabited by all possible molecules and chemical compounds is impractical, and therefore presents a challenge. To fill this gap, we present ChemoGraph, a novel visual analytics technique for interactively exploring related chemicals. In ChemoGraph, we formalize a chemical space as a hypergraph and apply novel machine learning models to compute related chemical compounds. It uses a database to find related compounds from a known space and a machine learning model to generate new ones, which helps enlarge the known space. Moreover, ChemoGraph highlights interactive features that support users in viewing, comparing, and organizing computationally identified related chemicals. With a drug discovery usage scenario and initial expert feedback from a case study, we demonstrate the usefulness of ChemoGraph.

chemical space exploration↗
DOE OSTI2024

Uncertainty Propagation from Experiment Measurements to Modeling Approaches: A Case for SMR Steam Entrainment Testing

To license new and advanced reactor designs, regulators must be convinced that their unique safety cases—relative to existing large scale reactors—have been adequately addressed by the designed reactor protection systems. In water cooled small modular reactors (SMRs), droplet entrainment in steam flow has significant implications on the progression of accident scenarios due to its compact design features, which requires representative test data applicable to SMR designs. Computer code, modeling and simulation (M&S) tools and models require adequate verification, assessment, and qualification. This includes M&S results validation against scaled empirical data within allowable uncertainty bands to gain regulatory approvals during the various stages of reactor system design, demonstration, and commercialization. However, measurement uncertainty within the empirical datasets and test data applicability ranges requires careful consideration of M&S inputs (i.e., boundary conditions, and initial conditions), and verification and validation efforts. This study focuses on uncertainty quantification in designing scaled test facilities for SMR applications with appropriate measurements and a standard data-reduction method to estimate thermal hydraulics characteristics parameters that incorporate physics phenomena of interest. In addition, this study supports the evaluation model development and assessment process using M&S that interfaces with advanced computing tools and digital twin capabilities. This will allow synchronization between experiment and modeling approaches for droplet entrainment testing and analysis, improving diagnostics, prognostics, and decision-making to accelerate regulatory approval.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2021

Suitability of Altair FEKO's Computational Electromagnetic Techniques for Tree Model Scattering and Penetration at Ku-band

Foliage penetration (FOPEN) radar at lower frequencies (VHF, UHF) is a well-studied area with many contributions. However, there is growing interest in using higher Ku-band frequencies (12-18 GHz) for FOPEN. Specifically, the reduced wavelength sizes provide some key saliencies for developing more optimized detection solutions. The disadvantage is that exploiting Ku-band for FOPEN is complicated because higher frequencies have pronounced scattering effects due to their smaller wavelengths. A methodology h as been developed to model and simulate FOPEN problems that characterize the phenomenology of Ku-band electromagnetic ( EM ) wave transmissions through moderate foliage. The details of this research (i.e. the realistic tree models, simulation setup and results) are documented in multiple reports. The main focus of this report is to describe the preliminary validation and verification of Altair FEKO, the computational EM (CEM) software used for this research, as well as present a simplified symmetrical tree model and an introductory CAD tree model.

42 ENGINEERING↗
DOE OSTI2024

Prediction of O and OH Adsorption on Transition Metal Oxide Surfaces from Bulk Descriptors

In the search for stable and active catalysts, density functional theory and machine learning (ML) based models can accelerate the screening of materials. While stability is conveniently addressed on the bulk level of computation, the modelling of catalytic activity requires expensive surface simulations. Here, in this work, we develop models for the surface adsorption energy of O and OH intermediates across a consistent and extensive dataset of pure transition metal oxide surfaces. We show that adsorption energies across metal oxidation states of +2 to +6 are well captured from the metal-oxygen bond strength extracted from the bulk level calculation. Specifically, we calculate the integrated crystal orbital Hamiltonian population (ICOHP) of the metal-oxygen bond in the bulk oxide and employ a simple normalization scheme to obtain a strong correlation with adsorption energetics. By combining our ICOHP descriptor with non DFT features in a Gaussian Process regression (GPR) model, we achieve high model accuracy with mean absolute errors of 0.166 and 0.219 eV for OH and O adsorption, respectively. By targeting the O-OH adsorption energy difference with our GPR model, we predict the the oxygen evolution reaction (OER) activity from bulk descriptors only. Furthermore, we utilize the strong correlation between the COHP and metal oxygen bond lengths to rapidly predict adsorption energetics and catalytic activity from the optimized bulk geometry. Our approach can enable an efficient search for active catalysts by eliminating the need for surface calculations in the initial screening phase.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Empirically-calibrated H100 node power models for accurate AI training energy estimation

Accurately quantifying the energy use of artificial intelligence (AI) training is critical for infrastructure planning, carbon accounting, and sustainable data center operation, but few studies have directly measured the power consumption of production workloads on contemporary hardware. By combining empirical measurements from Brookhaven National Laboratory during AI training on 8-graphics-processing-unit H100 systems with open-source benchmarking data, we develop statistical models relating computational intensity to node-level power consumption. We measure the gap between manufacturer-rated thermal design power (TDP) and actual power demand during AI training. Our analysis reveals that even computationally intensive workloads operate at only 76% of the 10.2 kW TDP rating. Our architecture-specific model, calibrated to floating-point operations, predicts energy consumption with 11.4% mean absolute percentage error, significantly outperforming TDP-based approaches (27%–37% error). We identified distinct power signatures between transformer and convolutional neural network architectures, with transformers showing characteristic fluctuations that may impact grid stability. These results provide a measurement-grounded basis for improving AI training energy estimates, enabling more reliable infrastructure sizing, cost projections, and environmental impact assessments.

Newkirk, Alex C↗
DOE OSTI2020

Transient Weighted Moving-Average Model of Photovoltaic Module Back-Surface Temperature

Accurate modeling of photovoltaic (PV) performance requires the precise calculation of module temperature. Currently, most temperature models rely on steady-state assumptions that do not account for the transient climatic conditions and thermal mass of the module. On the other hand, complex physics-based transient models are computationally expensive and difficult to parameterize. In order to address this, a new approach to transient thermal modeling was developed, in which the steady-state predictions from previous timesteps are weighted and averaged to accurately predict the module temperature at finer time scales. This model is informed by 3-D finite-element analyses, which are used to calculate the effect of wind speed and module unit mass on module temperature. The model, in application, serves as an added filter over existing steady-state models that smooths out erroneous values that are a result of intermittency in solar resource. Validation of this moving-average model has shown that it can improve the overall PV energy performance model accuracy by as much as 0.58% over steady-state models based on mean absolute error improvements and can significantly reduce the variability between the model predictions and measured temperature times series data.

14 SOLAR ENERGY↗
DOE OSTI2025

Modeling MTS pyrolysis and SiC deposition kinetics using principal component analysis and neural networks

Accurate chemical kinetics modeling is crucial for improving the efficiency of chemical processing and synthesis of ceramic matrix composites. Detailed kinetic models are computationally expensive due to the large number of transported chemical species, while the simplified physics-based models, such as single-step global mechanisms, are efficient but often overlook key chemical intermediates and pathways. Recent deep learning approaches promise accurate and cost-effective models. Yet, they require additional closures for the transported nonlinear latent variables, complicating integration with existing solvers. In this work, we develop a hybrid linear—nonlinear reduced model for silicon carbide deposition from methyltrichlorosilane precursor by combining principal component analysis (PCA) and autoencoder (AE) neural network (NN) approaches. PCA is used to identify a smaller set of linear transport variables, enabling direct reuse of conventional transport solvers. NNs then reconstruct the full chemical state from these reduced variables. We demonstrate the method on a chemical vapor deposition reactor—comprising a gas-phase pyrolysis plug flow reactor and a heterogeneous surface reactor—over a wide range of temperatures, pressures, and residence times. Our PCA–AE model achieves high accuracy with only five transported scalars, achieving an eightfold cost reduction compared to detailed mechanisms, in both a priori (using data from the test set only) and a posteriori (coupled with a differential equation solver). In conclusion, notable errors arise primarily near training domain boundaries and for long residence times, indicating the need for domain shift indicators and better long-horizon predictions in future reduced chemistry model development.

autoencoder neural networks↗
DOE OSTI2022

Iterative Workflow for Quantification and Minimization of Reduced Chemistry-Induced Uncertainties in Reacting Flow Simulation

Reduced chemistry models mitigate computational cost but introduce two sources of uncertainties in reacting flow simulation, including chemical information loss due to model reduction, and approximation errors due to non-optimal projection. We present an iterative workflow for quantification and minimization of reduced chemistry-induced uncertainties in reacting flow simulations.

chemistry↗
DOE OSTI2024

Decomposing a renewable energy design and dispatch model

We address a mixed-integer linear programming model which selects a cost-minimizing set of available technologies with which to design a renewable energy system and prescribe their associated dispatch decisions. Realistically sized instances of such models pose computational challenges. To this end, we develop a Lagrangian heuristic based on a decomposition methodology which partitions the model into blocks and optimizes these more manageable, smaller subproblems. It also provides a lower bound to assess solution quality. In conclusion, we apply this methodology to the National Renewable Energy Laboratory's Renewable Energy Integration and Optimization (REopt TM ) model to generate near-optimal solutions to realistic instances containing, on average, approximately 300,000 variables and at least as many constraints, with a mean 30% optimality gap improvement using a five-minute solution time limit, compared to directly solving the original monolith.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

A Boundary Element Model for Assessing Large‐Scale Pressurization in Faulted Geological Storage Systems

Assessing large-scale pressurization at the regional scale—a possible outcome of large subsurface storage applications such as wastewater injection and geological carbon sequestration—presents significant computational challenges. These challenges are particularly pronounced when accounting for complex geologic structures with multiple reservoir and caprock layers, fault zones, and wells. This study introduces a computationally efficient model that integrates single-phase semi-analytical solutions with a boundary element (BE) approach. The model simulates pressure propagation in multilayered 3D systems, including vertical faults, caprock, basement, and confining units. We apply this new model to a representative scenario involving CO 2 injection near a partially sealing fault with verification against an independent two-phase flow model. Results demonstrate that our model accurately captures far-field pressure responses and that, outside the CO 2 plume zone, pressure predictions from single-phase and two-phase models are nearly identical. This supports the use of single-phase models like ours for efficient estimation of far-field pressure changes. Additionally, we demonstrate its effectiveness at a large scale, incorporating multiple wells and faults. With its ability to represent multiple wells, fault zones, and geological heterogeneity, our model is well suited for assessments of basin-scale pressurization. Its computational efficiency also makes it a promising tool for integration with optimization frameworks aimed at designing and managing injection strategies in faulted storage systems.

Cihan, A. [Lawrence Berkeley National Laboratory (↗