Search NASA⌕ Search

SEARCH · Search NASA

Results for “Neural network models”

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 163 records · Page 9
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Neuromorphic intermediate representation: A unified instruction set for interoperable brain-inspired computing

Abstract Spiking neural networks and neuromorphic hardware platforms that simulate neuronal dynamics are getting wide attention and are being applied to many relevant problems using Machine Learning. Despite a well-established mathematical foundation for neural dynamics, there exists numerous software and hardware solutions and stacks whose variability makes it difficult to reproduce findings. Here, we establish a common reference frame for computations in digital neuromorphic systems, titled Neuromorphic Intermediate Representation (NIR). NIR defines a set of computational and composable model primitives as hybrid systems combining continuous-time dynamics and discrete events. By abstracting away assumptions around discretization and hardware constraints, NIR faithfully captures the computational model, while bridging differences between the evaluated implementation and the underlying mathematical formalism. NIR supports an unprecedented number of neuromorphic systems, which we demonstrate by reproducing three spiking neural network models of different complexity across 7 neuromorphic simulators and 4 digital hardware platforms. NIR decouples the development of neuromorphic hardware and software, enabling interoperability between platforms and improving accessibility to multiple neuromorphic technologies. We believe that NIR is a key next step in brain-inspired hardware-software co-evolution, enabling research towards the implementation of energy efficient computational principles of nervous systems. NIR is available atneuroir.org

Science & Technology - Other Topics↗
DOE OSTI2024

Model-form Error Correction using Universal Differential Equations for an Agent-Based Model of Infectious Disease

This report demonstrates universal differential equations (UDEs) as an approach to bridge the gap between ordinary differential equations (ODE) models and agent-based models (ABMs). Using UDE models as surrogates for ABMs allows us to preserve the foundational ODE that represents global disease dynamics while coupling it with a neural network model to approximate functions for the local behaviors of the ABM.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS2022

Lithium-ion Battery Strain Gauge Monitoring and Depth of Discharge Estimation

Lithium-ion battery diagnostics and prognostics rely on measurements of electrical impedance, capacity, and voltage to infer the internal state of the battery. Mechanical changes to the cell structure represent an additional measure of the battery’s state because these changes are related to the overall battery health. As lithium-ion batteries are charged and discharged, lithium ions are inserted or removed from the anode and cathode, a process called intercalation and de-intercalation. As lithium ions intercalate and de-intercalate, they can cause changes to the lattice of the electrode particles, resulting in volumetric changes. These volumetric changes cause mechanical stresses and strains on the lithium-ion battery electrodes, and subsequently, the whole cell’s thickness varies as it is charged and discharged. This paper presents a study on the use of surface-mounted strain gauges for in-situ measurement of structural changes to lithium-ion batteries, along with a characterization of the unit-to-unit differences in strain response. A neural network modeling structure is then used to predict the battery’s depth of discharge under dynamic discharge conditions.

Lithium-ion batteries, depth of discharge, strain ↗
DOE OSTI2025

Leveraging prior mean models for faster Bayesian optimization of particle accelerators

Tuning particle accelerators is a challenging and time-consuming task that can be automated and carried out efficiently using suitable optimization algorithms, such as model-based Bayesian optimization techniques. One of the major advantages of Bayesian algorithms is the ability to incorporate prior information about beam physics and historical behavior into the model used to make control decisions. In this work, we examine incorporating prior accelerator physics information into Bayesian optimization algorithms by utilizing fast executing, neural network models trained on simulated or historical datasets as prior mean functions in Gaussian process models. We show that in ideal cases, this technique substantially increases convergence speed to optimal solutions in high-dimensional tuning parameter spaces. Additionally, we demonstrate that even in non-ideal cases, where prior models of beam dynamics do not exactly match experimental conditions, the use of this technique can still enhance convergence speed. Finally, we demonstrate how these methods can be used to improve optimization in practical applications, such as transferring information gained from beam dynamics simulations to online control of the LCLS injector, and transferring knowledge gained from experimental measurements across different operating modes, such as accelerating different ion species at the ATLAS heavy ion accelerator.

43 PARTICLE ACCELERATORS↗
NASA NTRS2022

Instantaneous Photosynthetically Available Radiation (IPAR) prediction models based on Neural Network for ocean waters.

Instantaneous photosynthetically available radiation (IPAR) at the ocean surface and its vertical profile below the surface play a critical role in models to calculate net primary productivity of marine phytoplankton. In this work, we report two IPAR prediction models based on neural network (NN) approach, one for open ocean and the other for coastal waters. These models are trained, validated, and tested using a large volume of synthetic datasets for open ocean and coastal waters simulated by a radiative transfer model. Our NN models are designed to predict IPAR under a large range of atmospheric and oceanic conditions. The NN models can compute subsurface IPAR profile very accurately up to euphotic zone depth. The root mean square errors associated with the diffuse attenuation coefficient of IPAR are less than 0.011 𝑚−1 and 0.036 𝑚−1 for open ocean and coastal waters respectively. The performance of the NN models is better than presently available semi analytical models, with significant superiority in coastal waters.

PACE↗
NASA NTRSDate not supplied

Usage of ChatGPT for Engineering Design and Analysis Tool Development

ChatGPT, a generative AI large language model, has recently captured significant attention in both the computer science community and the broader public domain. It has demonstrated a wide range of capabilities, from answering simple questions to writing fully functional computer code. This study spotlights both the capabilities and limitations of ChatGPT when addressing engineering problems. The model's capacity to generate practical engineering tools is highlighted through an example of a prompt that leads to an interactive plotting tool, enabling the examination of the fluid boundary layer around a fan blade. Subsequently, the paper also uncovers potential pitfalls in ChatGPT’s application, shown through an unsuccessful attempt to use ChatGPT to automate a process in Ansys Workbench through scripting. The research further investigates ChatGPT's proficiency in addressing inquiries and providing explanations about the functionalities of OpenMDAO, an open-source, multidisciplinary design, analysis, and optimization tool developed at NASA Glenn Research Center. Finally, an optimization methodology, developed with ChatGPT’s help, is applied to the structural optimization of a fan blade. The developed optimization method utilizes T-Blade3 for geometry generation, Ansys Mechanical for meshing and finite element analysis, and sci-kit learn’s MLPRegressor method to generate a trained neural network model of the design space. OpenMDAO is then used to find the optimal point within the design space. The outcome is a significant reduction in stress in the optimized model—less than one-fifth of the stress value in the baseline model.

Design↗
DOE OSTI2026

Machine-learning force-field models for dynamical simulations of metallic magnets

We review recent advances in machine-learning (ML) force-field methods for Landau–Lifshitz–Gilbert simulations of itinerant electron magnets, focusing on their scalability and transferability. Built on the principle of locality, a deep neural-network model is developed to efficiently and accurately predict electron-mediated forces governing spin dynamics. Symmetry-aware descriptors constructed through a group-theoretical approach ensure rigorous incorporation of both lattice and spin-rotation symmetries. The framework is demonstrated using the prototypical s-d exchange model widely employed in spintronics. ML-enabled large-scale simulations reveal novel nonequilibrium phenomena, including anomalous coarsening of tetrahedral spin order on the triangular lattice and the freezing of phase-separation dynamics in lightly hole-doped, strong-coupling square-lattice systems. These results establish ML force-field frameworks as scalable, accurate, and versatile tools for modeling nonequilibrium spin dynamics in itinerant magnets.

Artificial neural networks↗
NASA NTRS2024

Usage of ChatGPT for Engineering Design and Analysis Tool Development

ChatGPT, a generative AI large language model, has recently captured significant attention in both the computer science community and the broader public domain. It has demonstrated a wide range of capabilities, from answering simple questions to writing fully functional computer code. This study spotlights both the capabilities and limitations of ChatGPT when addressing engineering problems. The model's capacity to generate practical engineering tools is highlighted through an example of a prompt that leads to an interactive plotting tool, enabling the examination of the fluid boundary layer around a fan blade. Subsequently, the paper also uncovers potential pitfalls in ChatGPT’s application, shown through an unsuccessful attempt to use ChatGPT to automate a process in Ansys Workbench through scripting. The research further investigates ChatGPT's proficiency in addressing inquiries and providing explanations about the functionalities of OpenMDAO, an open-source, multidisciplinary design, analysis, and optimization tool developed at NASA Glenn Research Center. Finally, an optimization methodology, developed with ChatGPT’s help, is applied to the structural optimization of a fan blade. The developed optimization method utilizes T-Blade3 for geometry generation, Ansys Mechanical for meshing and finite element analysis, and sci-kit learn’s MLPRegressor method to generate a trained neural network model of the design space. OpenMDAO is then used to find the optimal point within the design space. The outcome is a significant reduction in stress in the optimized model—less than one-fifth of the stress value in the baseline model.

Design↗
DOE OSTI2024

Upsampling Monte Carlo Reactor Simulation Tallies in Depleted Sodium-Cooled Fast Reactor Assemblies Using a Convolutional Neural Network

The computational demand of neutron Monte Carlo transport simulations can increase rapidly with the spatial and energy resolution of tallied physical quantities. Convolutional neural networks have been used to increase the resolution of Monte Carlo simulations of light water reactor assemblies while preserving accuracy with negligible additional computational cost. Here, we show that a convolutional neural network can also be used to upsample tally results from Monte Carlo simulations of sodium-cooled fast reactor assemblies, thereby extending the applicability beyond thermal systems. The convolutional neural network model is trained using neutron flux tallies from 300 procedurally generated nuclear reactor assemblies simulated using OpenMC. Validation and test datasets included 16 simulations of procedurally generated assemblies, and a realistic simulation of a European sodium-cooled fast reactor assembly was included in the test dataset. We show the residuals between the high-resolution flux tallies predicted by the neural network and high-resolution Monte Carlo tallies on relative and absolute bases. The network can upsample tallies from simulations of fast reactor assemblies with diverse and heterogeneous materials and geometries by a factor of two in each spatial and energy dimension. The network’s predictions are within the statistical uncertainty of the Monte Carlo tallies in almost all cases. This includes test assemblies for which burnup values and geometric parameters were well outside the ranges of those in assemblies used to train the network.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2025

Anti-symmetric barron functions and their approximation with sums of determinants

A fundamental problem in quantum physics is to encode functions that are completely anti-symmetric under permutations of identical particles. The architecture of neural network models for the electron wave function typically comprises an equivariant component followed by a summation of determinants. The recently introduced Generic Antisymmetric (GA) block is designed to enhance the expressivity of such neural wave functions, and it was found that the 2-layer GA block achieved more accurate energies than the corresponding single-determinant FermiNet architecure, suggesting its promise as a way to improve the expressivity of neural wave functions. In this paper we show how the function expressed by the 2-layer GA block can be decomposed into a sum of determinants. We formalize this result by defining the antisymmetric Barron space as a generalized version of the 2-layer GA block and providing an appromation theorem for this function class. This result can be viewed as a negative result showing that the 2-layer GA block is not more expressive than using multiple determinants.

Abrahamsen, Nilin↗
DOE OSTI2024

Towards interpretable Cryo-EM: disentangling latent spaces of molecular conformations

Molecules are essential building blocks of life and their different conformations (i.e., shapes) crucially determine the functional role that they play in living organisms. Cryogenic Electron Microscopy (cryo-EM) allows for acquisition of large image datasets of individual molecules. Recent advances in computational cryo-EM have made it possible to learn latent variable models of conformation landscapes. However, interpreting these latent spaces remains a challenge as their individual dimensions are often arbitrary. The key message of our work is that this interpretation challenge can be viewed as an Independent Component Analysis (ICA) problem where we seek models that have the property of identifiability. That means, they have an essentially unique solution, representing a conformational latent space that separates the different degrees of freedom a molecule is equipped with in nature. Thus, we aim to advance the computational field of cryo-EM beyond visualizations as we connect it with the theoretical framework of (nonlinear) ICA and discuss the need for identifiable models, improved metrics, and benchmarks. Moving forward, we propose future directions for enhancing the disentanglement of latent spaces in cryo-EM, refining evaluation metrics and exploring techniques that leverage physics-based decoders of biomolecular systems. Moreover, we discuss how future technological developments in time-resolved single particle imaging may enable the application of nonlinear ICA models that can discover the true conformation changes of molecules in nature. The pursuit of interpretable conformational latent spaces will empower researchers to unravel complex biological processes and facilitate targeted interventions. This has significant implications for drug discovery and structural biology more broadly. More generally, latent variable models are deployed widely across many scientific disciplines. Thus, the argument we present in this work has much broader applications in AI for science if we want to move from impressive nonlinear neural network models to mathematically grounded methods that can help us learn something new about nature.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS1992

Incorporation of varying types of temporal data in a neural network

Most neural network models do not specifically deal with temporal data. Handling of these variables is complicated by the different uses to which temporal data are put, depending on the application. Even within the same application, temporal variables are often used in a number of different ways. In this paper, types of temporal data are discussed, along with their implications for approximate reasoning. Methods for integrating approximate temporal reasoning into existing neural network structures are presented. These methods are illustrated in a medical application for diagnosis of graft-versus-host disease which requires the use of several types of temporal data.

Cohen, M. E.↗
NASA NTRS2001

Inferring Instantaneous, Multivariate and Nonlinear Sensitivities for the Analysis of Feedback Processes in a Dynamical System: Lorenz Model Case Study

A new approach is presented for the analysis of feedback processes in a nonlinear dynamical system by observing its variations. The new methodology consists of statistical estimates of the sensitivities between all pairs of variables in the system based on a neural network modeling of the dynamical system. The model can then be used to estimate the instantaneous, multivariate and nonlinear sensitivities, which are shown to be essential for the analysis of the feedbacks processes involved in the dynamical system. The method is described and tested on synthetic data from the low-order Lorenz circulation model where the correct sensitivities can be evaluated analytically.

Aires, Filipe↗
NASA NTRS1993

Modelling intelligent behavior

An introductory discussion of the related concepts of intelligence and consciousness suggests criteria to be met in the modeling of intelligence and the development of intelligent materials. Methods for the modeling of actual structure and activity of the animal cortex have been found, based on present knowledge of the ionic and cellular constitution of the nervous system. These have led to the development of a realistic neural network model, which has been used to study the formation of memory and the process of learning. An account is given of experiments with simple materials which exhibit almost all properties of biological synapses and suggest the possibility of a new type of computer architecture to implement an advanced type of artificial intelligence.

Green, H. S.↗
NASA NTRS2024

Short–Period Variables in TESS Full–Frame Image Light Curves Identified via Convolutional Neural Networks

The Transiting Exoplanet Survey Satellite (TESS) mission measured light from stars in ∼85% of the sky throughout its 2 yr primary mission, resulting in millions of TESS 30-minute-cadence light curves to analyze in the search for transiting exoplanets. To search this vast data set, we aim to provide an approach that is computationally efficient, produces accurate predictions, and minimizes the required human search effort. We present a convolutional neural network that we train to identify short-period variables. To make a prediction for a given light curve, our network requires no prior target parameters identified using other methods. Our network performs inference on a TESS 30-minute-cadence light curve in ∼5 ms on a single GPU, enabling large-scale archival searches. We present a collection of 14,156 short-period variables identified by our network. The majority of our identified variables fall into two prominent populations, one of close-orbit main-sequence binaries and another of δ Scuti stars. Our neural network model and related code are additionally provided as open-source code for public use and extension.

Convolutional neural networks↗
NASA NTRS1992

Real-time antenna fault diagnosis experiments at DSS 13

Experimental results obtained when a previously described fault diagnosis system was run online in real time at the 34-m beam waveguide antenna at Deep Space Station (DSS) 13 are described. Experimental conditions and the quality of results are described. A neural network model and a maximum-likelihood Gaussian classifier are compared with and without a Markov component to model temporal context. At the rate of a state update every 6.4 seconds, over a period of roughly 1 hour, the neural-Markov system had zero errors (incorrect state estimates) while monitoring both faulty and normal operations. The overall results indicate that the neural-Markov combination is the most accurate model and has significant practical potential.

Mellstrom, J.↗
DOE OSTI2025

Even Higher-Level Synthesis: An Exploration of AI Hardware Accelerators using HLS4ML

With the rise of artificial intelligence, the popularization of deep learning, and a constantly evolving industry, the demand for flexible and efficient tools has never been greater. As algorithms grow more complex, their runtime and energy consumption increase exponentially. Customized hardware accelerators, long used for specific mathematical operations, remain essential for managing modern applications' computational and power demands. Hardware accelerators can speed up complex computations by orders of magnitude, but their manual design and verification processes are often challenging and time-consuming. High-Level Synthesis (HLS) provides a solution by transforming high-level algorithm descriptions, typically written in C++ or SystemC, into synthesizable RTL suitable for hardware implementation. This approach reduces development time for RTL engineers while offering flexibility beyond what traditional handwritten RTL can provide. We extended this capability to the machine-learning domain with the open-source framework hls4ml, which allows neural networks trained in Python frameworks like Tensorflow or PyTorch to be synthesized into efficient hardware representations for the traditional FPGA and ASIC flows. This breakthrough addresses the growing need for reduced design turnaround and easy verification of ML hardware accelerators with low latency and power efficiency constraints. During this tutorial, we will demonstrate how Python complements HLS by simplifying the ML design process, bridging the gap between software and hardware development. Attendees will explore how we translate neural networks modeled in Python into fixed-point C++ models suitable for HLS workflows. We will dive into strategies like Value-Range Analysis and Quantization-Aware Training, which optimize these designs for deployment and evaluate their accuracy, power consumption, and energy efficiency. To exemplify these concepts, experts from Fermilab will share their experiences applying this technology to high-energy physics experiments, where real-time, low-latency processing is critical. Over the years, Fermilab engineers have demonstrated how deep neural networks, optimized for hardware using hls4ml, can meet the stringent requirements of trigger systems at the CERN Large Hadron Collider. These systems rely on rapid decision-making to process immense data volumes while retaining only the most relevant events for further analysis. The application of hls4ml has also been extended to innovative technologies like smart pixel arrays. These smart pixels integrate ML inference capabilities directly into sensor devices, enabling localized data processing at the pixel level. This approach drastically reduces the need to transmit raw data to external processing units, significantly decreasing power consumption and latency. By embedding neural networks within the pixel architecture, the smart pixels can identify and prioritize relevant data in real time, providing a highly efficient solution for edge computing in scenarios such as particle detectors and imaging systems. Fermilab's work highlights the potential of hardware-accelerated ML in scenarios where both speed and power efficiency are mission-critical. Through this tutorial, attendees will gain valuable insights into the challenges and solutions of deploying ML in hardware. Understanding how HLS and hls4ml streamline the development of neural network-based hardware accelerators is fundamental for the industry's future. Participants will learn how these technologies are shaping the future of AI and scientific computing.

Di Guglielmo, Giuseppe [Fermilab]↗