Search NASA⌕ Search

SEARCH · Search NASA

Results for “computational frameworks”

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 991 records · Page 55
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Confidence Assessment for Automatic Target Recognition

Confidence assessment is critical for effective automatic target recognition (ATR). Productive use and interpretation of ATR results by analysts or downstream algorithms requires not only algorithmic declarations of target presence and identity, but also algorithmic assessment of the certainty of those declarations in comparison to the certainties of alternative target-identity possibilities. Unfortunately, despite its importance, confidence assessment is an understudied, underdeveloped, and often-neglected function of ATR systems. This lack of regard stems not only from the difficulty of accurate algorithmic determination of target-identity certainty, but also from a general lack of understanding and careful consideration about what confidence should actually represent. We present a framework for confidence assessment that establishes a clear definition of confidence and provides a straightforward theoretical basis for its calculation. This framework is grounded in a hypothesis-theoretic consideration of ATR and it springs from from a handful of axiomatic principles concerning the nature and meaning of confidence in this context. This framework establishes a rigorous mathematical definition of confidence and it provides equations relating confidence to other information that is almost always provided by ATRs. We present an approach for computing confidence within this framework, using an advance process of ATR characterization followed by a simple computation at the time of ATR execution. We discuss practical difficulties with our approach, and we suggest methods for effective mitigation of these difficulties in implemented systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

A perspective on Bayesian methods applied to materials discovery and design

For more than two decades, there has been increasing interest in developing frameworks for the accelerated discovery and design of novel materials that could enable promising and transformative technologies. The Integrated Computational Materials Engineering (ICME) program called for integrating computational tools to establish linkages along process-structure-property-performance (PSPP) chains. The Materials Genome Initiative called for integrating experiments and computations within data science frameworks as a strategy to accelerate the materials development cycle. While these frameworks and paradigms have been quite influential, traditional ICME or data science-based approaches tend to have some limitations, mainly when querying the materials space is costly and very little information is available. Bayesian methods are more suitable in this context due to their efficiency gains. To this end, the materials discovery problem is framed as a Bayesian Optimization (BO). Different examples in which BO has been applied to solve materials discovery problems are presented. The methods/examples discussed include BO under model uncertainty, multi-information source BO, multi-objective and multi-constraint BO, and batch BO. Bayesian Materials Discovery is a promising area of research that is likely to become more influential as more attention is put on autonomous materials discovery platforms. Therefore, a discussion is provided on the potential development of such methods to increase the ability of existing platforms in materials discovery. Here, the ultimate goal is to pave the way to autonomous materials discovery.

36 MATERIALS SCIENCE↗
DOE OSTI2015

Predicting Weak Lensing Statistics from Halo Mass Reconstructions (Final Paper)

As dark matter does not absorb or emit light, its distribution in the universe must be inferred through indirect effects such as the gravitational lensing of distant galaxies. While most sources are only weakly lensed, the systematic alignment of background galaxies around a foreground lens can constrain the mass of the lens which is largely in the form of dark matter. In this paper, I have implemented a framework to reconstruct all of the mass along lines of sight using a best-case dark matter halo model in which the halo mass is known. This framework is then used to make predictions of the weak lensing of 3,240 generated source galaxies through a 324 arcmin 2 field of the Millennium Simulation. The lensed source ellipticities are characterized by the ellipticity-ellipticity and galaxy-mass correlation functions and compared to the same statistic for the intrinsic and ray-traced ellipticities. In the ellipticity-ellipticity correlation function, I and that the framework systematically under predicts the shear power by an average factor of 2.2 and fails to capture correlation from dark matter structure at scales larger than 1 arcminute. The model predicted galaxy-mass correlation function is in agreement with the ray-traced statistic from scales 0.2 to 0.7 arcminutes, but systematically underpredicts shear power at scales larger than 0.7 arcminutes by an average factor of 1.2. Optimization of the framework code has reduced the mean CPU time per lensing prediction by 70% to 24 ± 5 ms. Physical and computational shortcomings of the framework are discussed, as well as potential improvements for upcoming work.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

Sketching the pion's valence-quark generalised parton distribution

In order to learn effectively from measurements of generalised parton distributions (GPDs), it is desirable to compute them using a framework that can potentially connect empirical information with basic features of the Standard Model. We sketch an approach to such computations, based upon a rainbow-ladder (RL) truncation of QCD's Dyson-Schwinger equations and exemplified via the pion's valence dressed-quark GPD, H-pi(V)(chi, xi, t). Our analysis focuses primarily on xi = 0, although we also capitalise on the symmetry-preserving nature of the RL truncation by connecting H-pi(V)(chi, xi = +/- 1, t) with the pion's valence-quark parton distribution amplitude. We explain that the impulse-approximation used hitherto to define the pion's valence dressed-quark GPD is generally invalid owing to omission of contributions from the gluons which bind dressed-quarks into the pion. A simple correction enables us to identify a practicable improvement to the approximation for H(pi)(V)p(chi, 0, t), expressed as the Radon transform of a single amplitude. Therewith we obtain results for H pi V(chi, 0, t) and the associated impact-parameter dependent distribution, q(pi)(V)(chi, vertical bar(b) over right arrow (perpendicular to)vertical bar), which provide a qualitatively sound picture of the pion's dressed-quark structure at a hadronic scale. We evolve the distributions to a scale zeta = 2 GeV, so as to facilitate comparisons in future with results from experiment or other nonperturbative methods. (C) 2014 Published by Elsevier B. V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/3.0/).

Mezrag, C.↗
DOE OSTI2025

Data Readiness for Scientific AI at Scale

This paper examines how Data Readiness for AI (DRAI) principles apply to leadership-scale scientific datasets used to train foundation models. We analyze archetypal workflows across four representative domains—climate, nuclear fusion, bio/health, and materials—to identify common preprocessing patterns and domain-specific constraints. We introduce a two-dimensional readiness framework that combines canonical preprocessing patterns with a five-level operational readiness scale, both tailored to high-performance computing (HPC) environments. This framework helps outline key challenges in transforming large-scale scientific data into formats suitable for scalable AI training. Together, these dimensions form a conceptual maturity matrix that characterizes scientific data readiness and guides infrastructure development toward standardized, cross-domain support for scalable and reproducible AI for science.

Brewer, Wes [ORNL] (ORCID:0000000236393956)↗
DOE OSTI2021

Breaking Down the Computational Barriers to Real-Time Urban Flood Forecasting

Flooding impacts are on the rise globally, and concentrated in urban areas. Currently, there are no operational systems to forecast flooding at spatial resolutions that can facilitate emergency preparedness and response actions mitigating flood impacts. We present a framework for real-time flood modeling and uncertainty quantification that combines the physics of fluid motion with advances in probabilistic methods. The framework overcomes the prohibitive computational demands of high-fidelity modeling in real-time by using a probabilistic learning method relying on surrogate models that are trained prior to a flood event. This shifts the overwhelming burden of computation to the trivial problem of data storage, and enables forecasting of both flood hazard and its uncertainty at scales that are vital for time-critical decision-making before and during extreme events. The framework has the potential to improve flood prediction and analysis and can be extended to other hazard assessments requiring intense high-fidelity computations in real-time.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2011

Information causality and noisy computations

We reformulate the information causality in a more general framework by adopting the results of signal propagation and computation in a noisy circuit. In our framework, the information causality leads to a broad class of Tsirelson inequalities. This fact allows us to subject information causality to experimental scrutiny. A no-go theorem for reliable nonlocal computation is also derived. Information causality prevents any physical circuit from performing reliable computations.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2016

Efficacy of the SU(3) scheme for ab initio large-scale calculations beyond the lightest nuclei

We report on the computational characteristics of ab initio nuclear structure calculations in a symmetry-adapted no-core shell model (SA-NCSM) framework. We examine the computational complexity of the current implementation of the SA-NCSM approach, dubbed LSU3shell, by analyzing ab initio results for 6 Li and 12 C in large harmonic oscillator model spaces and SU3-selected subspaces. We demonstrate LSU3shell’s strong-scaling properties achieved with highly-parallel methods for computing the many-body matrix elements. Results compare favorably with complete model space calculations and significant memory savings are achieved in physically important applications. In particular, a well-chosen symmetry-adapted basis affords memory savings in calculations of states with a fixed total angular momentum in large model spaces while exactly preserving translational invariance.

74 ATOMIC AND MOLECULAR PHYSICS↗
DOE OSTI2014

A flexible method to calculate the distributions of discretization errors in operator-split codes with stochastic noise in problem data

This report investigates a technique to calculate the distributions of discretization errors for a model of advection-diffusion-reaction with stochastic noise in problem data. The focus is on operator-split discretization methods. The error is decomposed into components due to the splitting and due to the discretization within each component. We present a method to estimate the distributions of the total discretization error for a quantity of interest as well as the aforementioned error components. These computations require information about the propagation of uncertainty within and between code modules, for which purpose we use a flexible code framework for uncertainty quantification. Computational examples illustrate that in certain cases the distribution of numerical error can have a significant impact on the code output. Using information about the components of the numerical error, decisions can be made regarding discretization parameters to efficiently reduce the impact of numerical error on the distribution of model outputs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

Efficacy of the SU(3) scheme for ab initio large-scale calculations beyond the lightest nuclei

We report on the computational characteristics of ab initio nuclear structure calculations in a symmetry-adapted no-core shell model (SA-NCSM) framework. We examine the computational complexity of the current implementation of the SA-NCSM approach, dubbed LSU3shell, by analyzing ab initio results for 6 Li and 12 C in large harmonic oscillator model spaces and SU(3)-selected subspaces. We demonstrate LSU3shell's strong-scaling properties achieved with highly-parallel methods for computing the many-body matrix elements. Results compare favorably with complete model space calculations and signi cant memory savings are achieved in physically important applications. In particular, a well-chosen symmetry-adapted basis a ords memory savings in calculations of states with a fixed total angular momentum in large model spaces while exactly preserving translational invariance.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2021

Scalable Approaches to Selecting Key Entities in Large Networked Infrastructure Systems

This work aims at bringing advances in discrete optimization algorithms to solving practical engineering problems at scale. Often times, in many engineering design problems, there is a need to select a small set of influential or representative elements from a large ground set of entities in an optimal fashion. Submodular optimization provides for a formal way to solve such problems. Common examples with infrastructure systems involve sensor placement and identification of key entities with certain objectives. However, scaling these approaches to large infrastructure systems can be challenging because of the high computational complexity of the overall framework that include the optimization algorithms as well as high-complexity compute-oracles that provide the necessary objective function values. In this work, we explore a well-studied and widely-applicable paradigm, namely leader-selection in a multi-agent networked setting in the context of scalable methodologies. We demonstrate novel frameworks that utilize variations of accelerated submodular optimization algorithms along with linear-algebraic methods that can help accelerate the oracle computations. We further explore this combination in conjunction with graph partitioning paradigms to take advantage of the accelerated algorithms in a distributed setting. Finally we demonstrate the key findings on a practical problem in an operational setting. For this, we leverage an example road network with approximately 18k nodes and 27k edges in a traffic control application, where we seek a limited number of k=200 key intersections. This problem can be solved in a serial setting in just under 5 hours providing more than 2 orders of magnitude speed-up over methods that do not consider acceleration techniques.

Visweswara Sathanur, Arun↗
DOE OSTI2024

OLCF’s Advanced Computing Ecosystem (ACE): FY24 Efforts for the DOE Integrated Research Infrastructure (IRI) Program

This report highlights significant strides made by Oak Ridge National Laboratory’s Oak Ridge Leadership Computing Facility (OLCF) in advancing computational research and infrastructure. Through the Advanced Computing Ecosystem (ACE) strategic initiative, OLCF has been successfully integrated with Doe’s Integrated Research Infrastructure (IRI) program, establishing itself as a critical framework for enhancing scientific computing capabilities across various domains. This report outlines the activities, accomplishments, and future directions of ACE, emphasizing its role in developing cutting-edge technologies, supporting science pilots, and fostering collaborations that drive scientific innovation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

NeuroSymbolic Approaches as a Vector for Assured Artificial Intelligence

The deployment of artificial intelligence systems in critical applications requires higher levels of assurance for safety, security, and interpretability. While neurosymbolic (NESY) approaches combining neural networks with symbolic reasoning offer potential advantages for assured AI, existing differentiable neurosymbolic frameworks face significant limitations including computational overhead and performance constraints. This report investigates the ISED (InferSampleEstimateDescend) framework as an alternative approach that enables neurosymbolic learning without requiring endtoend differentiability. We evaluate ISED’s utility for geointelligence applications by comparing neurosymbolic models against standard neural networks on aircraft classification tasks using the RarePlanes and MTARSI imagery datasets. Our results demonstrate that while ISEDbased models achieve slightly lower accuracy (89.7% vs 92.1% on RarePlanes; 91.1% vs 92.5% on MTARSI), they provide critical explainability capabilities that enable tracing incorrect predictions back to specific attribute misclassifications. We also present an automated pipeline that generates both attributeclass mappings and neurosymbolic model architectures from natural language descriptions, significantly reducing the manual effort required for NESY model deployment. These findings suggest that ISED offers a promising direction for developing assured AI systems where interpretability and reasoning transparency are prioritized alongside performance.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Multidisciplinary design optimization in Architecture, Engineering, and Construction: a detailed review and call for collaboration

The design of buildings has become a complex and multidisciplinary problem involving multiple conflicting objectives as architects and designers address competing technical, economic, environmental, and societal concerns. This has been driving research in Architecture, Engineering, and Construction (AEC) toward rigorous multidisciplinary decision-making frameworks that generate and evaluate numerous design alternatives using multi-objective optimization in concert with simulation and analysis models of varying fidelity and computational expense. While such frameworks are well known and widely employed in the aerospace and systems engineering domains, efforts by design professionals and researchers in the AEC field are scattered at best. In this paper, we provide a detailed review of recent developments in optimization frameworks in the AEC field and subsequently highlight how such developments are largely compartmentalized into separate domains such as structural, energy, daylighting, and other performance factors. Herein, we further discuss the technical challenges involved in concurrent coupled multidisciplinary design optimization (MDO) in the AEC field such as interoperability issues between Building Information Modeling (BIM) environments, analysis/simulation tools, and optimization frameworks. We conclude by outlining research needed for more unified modeling and simulation-based optimization frameworks to aid in complex and multidisciplinary building design processes. We also highlight the need for the identification and development of multi-fidelity simulation tools for use across multiple design phases. As such, this paper contributes a novel roadmap to leverage aerospace and systems engineering research in MDO into the field of AEC, along with a call for researchers in the MDO community to seek collaborations in AEC field.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2022

Towards Ultra-high-resolution E3SM Land Modeling on Exascale Computers

Here we present an ultra-high-resolution E3SM land model (uELM) for high-fidelity land simulations targeting new Exascale computers. After considering modeling infrastructure compatibility and ELM software features, we designed a parallel model for the uELM development targeting hybrid architectures of new US Exascale computers. We also described a function unit test framework to expedite the piece-wise code porting (with compiler directives), verification, and global variable management. Furthermore, in this study, we report an early uELM model development using OpenACC within a function unit test framework on a pre-Exascale computer, demonstrate the performance of a uLEM submodel with a 3.0-time speedup, and summarize the code porting experience regarding global variable handling, deepcopy, memory reduction, and parallel loop reconstruction.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Toward Generalized Tensor Algebra for ab initio Quantum Chemistry Methods

The widespread use of tensor operations in describing electronic structure calculations has motivated the design of software frameworks for productive development of scalable optimized tensor-based electronic structure methods. Whereas prior work focused on Cartesian abstractions for dense tensors, we present an algebra to specify and perform tensor operations on a larger class of block-sparse tensors. We illustrate the use of this framework in expressing real-world computational chemistry calculations beyond the reach of existing frameworks.

Mutlu, Erdal↗
DOE OSTI2019

Combining Observational and Computational Uncertainty in Calibration Experiments

Traditional computer experiments using emulators have treat the emulator output as containing no error; however, this is not always feasible, as there exist computational codes which produce stochastic output, and future computations may require a balance between the time or resources to run the code and the accuracy of the output. This paper proposes a framework for incorporating both computational uncertainty and observational uncertainty into an analysis using a Gaussian process regression and calibration, and provides an example using simulated output from the one-dimensional diffusion equation. The particular values of observational and computation uncertainty which minimize a cost function are obtained.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

ERF: Energy Research and Forecasting

The Energy Research and Forecasting (ERF) code is a new model that simulates the mesoscale and microscale dynamics of the atmosphere using the latest high-performance computing architectures. It employs hierarchical parallelism using an MPI+X model, where X may be OpenMP on multicore CPU-only systems, or CUDA, HIP, or SYCL on GPU-accelerated systems. ERF is built on AMReX (Zhang et al., 2019, 2021), a block-structured adaptive mesh refinement (AMR) software framework that provides the underlying performance-portable software infrastructure for block-structured mesh operations. The "energy" aspect of ERF indicates that the software has been developed with renewable energy applications in mind. In addition to being a numerical weather prediction model, ERF is designed to provide a flexible computational framework for the exploration and investigation of different physics parameterizations and numerical strategies, and to characterize the flow field that impacts the ability of wind turbines to extract wind energy. The ERF development is part of a broader effort led by the US Department of Energy's Wind Energy Technologies Office.

17 WIND ENERGY↗