Search NASA⌕ Search

SEARCH · Search NASA

Results for “network optimization”

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 505 records · Page 28
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Application of deep learning to single-shot gas-phase laser-induced breakdown spectroscopy

Single-shot fs laser-induced breakdown spectroscopy (LIBS) has the potential to capture ns-scale electrode desorption phenomena in pulsed power fusion drivers. However, the successful implementation of the diagnostic for this purpose is challenging, as it requires interpreting single-shot measurements collected from low-density gas mixtures. In this work, we demonstrate the efficacy of a Bayesian-optimized convolutional neural network (CNN) to interpret these measurements. We generated 256 distinct measurement conditions at relevant gas pressures ranging from 80–530 mTorr by mixing 100–250 sccm H 2 and 50–200 sccm CH 4 in increments of 10 sccm. Despite the considerable overlap between signals separated by 20 sccm, the CNN is able to predict the H 2 flow rate with a root-mean-square error (RMSE) of 15.9 sccm and the CH 4 flow rate with an RMSE of 12.0 sccm. The average relative prediction error is <9% for each gas and largely remains below or near 10%.

Brown, Nathan Parnell [Sandia National Lab. (SNL-N↗
DOE OSTI2020

Hyper Parameter Tuning in Neural Optical Image Categorizer for the E-log (NOICE

The Fermilab Accelerator Division Electronic logbook (E-log) is a record of all the activities and events in the Division for the past 10 years and more. The Elog search function is a valuable resource and the institutional memory of the accelerator complex. About 300,000 files are stored in the E-log, of which the vast majority are images attached to entries and comments. The visual information contained in the images is not presently searchable. The goal of team NOICE (Neural Optical Image Categorizer for the E-log) was to design a neural network able to produce label categories for these images for use by searches. The group developed a dataset and trained a convolutional neural network (CNN) with optimized hyperparameter. Final performance metrics are presented.

Njekeu, Terence Franck↗
DOE OSTI2021

Financial Assistance for the Inaugural MagNetUS Workshop (Final Technical Report)

The MagNetUS Workshop was primarily scientific in character but also functioned as a joint user group meeting for the WiPPL and BaPSF collaborative user facilities, as well as included a community driven planning component to consider how to optimally form a network of facilities that operate in this general area with the proposed name of MagNetUS.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

HLS4ML Integration with QICK

The QICK (Quantum Instrumentation Control Kit) integration aims to enhance the readout of superconducting qubits by leveraging machine learning (ML) techniques. These techniques offer high accuracy, increased speed, and better state preservation for qubit readouts. The integration process employs neural network algorithms to optimize system performance and achieve high accuracy rates. The QICK board uses an FPGA, allowing for efficient real-time processing and high-performance execution of machine learning algorithms.

Ali, Mohamud↗
DOE OSTI2024

EFIT‐AI: Machine Learning and Artificial Intelligence Assisted Equilibrium Reconstruction for Tokamak Experiments and Burning Plasmas (Final Report)

The EFIT-AI project is creating a modern advanced equilibrium reconstruction code suitable for tokamak experiments of burning plasmas. EFIT [1,2] was the first and is the most extensively used equilibrium reconstruction code in the world. This project builds on the production-level experience and adds key elements as follows. 1. A Model Order Reduction (MOR) version of the two-dimensional (2D) Grad-Shafranov equation solver (EFIT-MORNN) using physics-informed neural networks. 2. Improved optimization and data analysis capabilities using a Bayesian framework enhanced with machine learning. 3. A MOR version of the three-dimensional (3D) perturbed equilibrium reconstruction tool.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Surrogate Neural Architecture Codesign Package (SNAC-Pack)

Neural Architecture Search is a powerful approach for automating model design, but existing methods struggle to accurately optimize for real hardware performance, often relying on proxy metrics such as bit operations. We present Surrogate Neural Architecture Codesign Package (SNAC-Pack), an integrated framework that automates the discovery and optimization of neural networks focusing on FPGA deployment.

Weitz, Jason [UC, San Diego]↗
DOE OSTI2025

Single Photon Searches at ICARUS with SPINE

The MiniBooNE Low-Energy Excess (LEE) of electron-like events from the Booster Neutrino Beam (BNB) has puzzled neutrino physicists for decades. One possible explanation has been an unpredicted excess of neutral current (NC) $\Delta$ resonance interactions with a subsequent radiative decay. An increase in the rate of NC $\Delta\rightarrow N\gamma$ events by a factor of 3.18 could explain the LEE seen by MiniBooNE. ICARUS also sees neutrinos from the BNB and can check the rate of these single photon events. The SPINE particle physics reconstruction suite leverages deep neural networks (DNNs) to optimize particle reconstruction and identification in Liquid Argon Time Projection Chambers (LArTPCs). I present preliminary findings on the effectiveness of these machine learnign (ML) techniques for single photon event reconstruction at ICARUS.

Hausner, Harry [Fermilab] (ORCID:0000000188932280)↗
DOE OSTI2019

Surrogate-Based Robust Design for a Non-Smooth Radiation Source Detection Problem

In this paper, we develop and numerically illustrate a robust sensor network design to optimally detect a radiation source in an urban environment. This problem exhibits several challenges: penalty functionals are non-smooth due to the presence of buildings, radiation transport models are often computationally expensive, sensor locations are not limited to a discrete number of points, and source intensity and location responses, based on a fixed number of sensors, are not unique. We consider a radiation source located in a prototypical 250 m × 180 m urban setting. To address the non-smooth properties of the model and computationally expensive simulation codes, we employ a verified surrogate model based on radial basis functions. Using this surrogate, we formulate and solve a robust design problem that is optimal in an average sense for detecting source location and intensity with minimized uncertainty.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2018

167-PFlops deep learning for electron microscopy: from learning physics to atomic manipulation

An artificial intelligence system called MENNDL, which used 25,200 NVIDIA Volta GPUs on Oak Ridge National Laboratory's Summit machine, automatically designed an optimal deep learning network in order to extract structural information from raw atomic-resolution microscopy data. In a few hours, MENNDL creates and evaluates millions of networks using a scalable, parallel, asynchronous genetic algorithm augmented with a support vector machine to automatically find a superior deep learning network topology and hyper-parameter set than a human expert can find in months. For the application of electron microscopy, the system furthers the goal of improving our understanding of the electron-beam-matter interactions and real-time image-based feedback, which enables a huge step beyond human capacity towards nanofabricating materials automatically. MENNDL has been scaled to the 4,200 available nodes of Summit achieving a measured 152.5 PFlops, with an estimated sustained performance of 167 PFlops when the entire machine is available.

Patton, Robert↗
DOE OSTI2020

(NOICE) Neural Optical Image Categorizer for the E-log

The Fermilab Accelerator Division Electronic logbook (E-log) is a record of all the activities and events in the Division for the past 10 years and more. The E-log search function is a valuable resource and the institutional memory of the accelerator complex. About 300,000 files are stored in the E-log, of which the vast majority are images attached to entries and comments. The visual information contained in the images is not presently searchable. The goal of team NOICE (Neural Optical Image Categorizer for the E-log) was to design a neural network able to produce label categories for these images for use by searches. The group developed a dataset and trained a convolutional neural network (CNN) with optimized hyperparameter

43 PARTICLE ACCELERATORS↗
DOE OSTI2021

Machine-Learned Manifold-Based Models for Large Eddy Simulation of Turbulent Combustion

Reduced-order manifold combustion models are commonly used to lower the cost of reacting Large Eddy Simulations (LES) and may be derived either from physical principles as in flamelet models or from data-driven methods like Principal Component Analysis (PCA). In either case, neural networks are increasingly used as part of these models to provide a nonlinear mapping between a small set of pre-defined variables that parameterize the manifold and outputs of interest, such as reaction rates. In this work, we propose a new manifold-based modeling approach that combines the definition of the manifold-parameterizing variables (linear combinations of species), the nonlinear mapping to the outputs, and closure of filtered quantities for LES into the structure of a single neural network. This allows the process used to train the neural network to simultaneously optimize both the functional form of the model and the identities of the inputs to the model. The new approach can flexibly incorporate thermochemical data from any combustion system; if trained on data from 1D flames it can be interpreted as an optimized flamelet model, but it can also be used to learn models from data from more complex configurations. This work presents a priori evaluations of the new approach in both contexts. Evaluation using data from 1D premixed flames demonstrates the physical interpretability of the manifold variables generated by the new approach. Evaluation using data from direct numerical simulations of turbulent flames shows improved predictions relative to either flamelet or PCA-based models in a more complex configuration.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Integrating Analytical Solutions and U-Net Model for Predicting Groundwater Contaminant Plumes in Pump-and-Treat Systems

Pump-and-treat (P&T) is a common technique for groundwater remediation involving the extraction and treatment of contaminated water above ground. Optimizing the design and operation of the P&T well network is essential for maximizing the system’s effectiveness and efficiency. However, this optimization often necessitates many model evaluations, leading to computationally demanding tasks. This study introduces a novel approach that integrates analytical solutions for groundwater dynamics with the U-Net (Ronneberger et al., 2015) deep learning framework to predict groundwater contaminant plume migration under dynamic pumping conditions. By incorporating the Thiem equation (Thiem, 1906) into the input preprocessing, the U-Net model transforms sparse well data into a continuous spatial field that captures the hydraulic impacts of pumping activities. This integration enables the model to leverage both deep learning capabilities and classical physics-based groundwater theories, enhancing prediction accuracy and computational efficiency. These advancements can facilitate rapid, large-scale evaluations of P&T optimization simulations, allowing for timely and effective decision-making in well placement and system management. We demonstrate the model's robust performance across both simplified transient 2D models and a more complex 3D heterogeneous site model at the 200 West P&T facility at the Hanford Site. The U-Net-based model offers substantial computational advantages, reducing simulation times significantly compared to full physics-based models and providing a powerful tool for rapid site evaluation and P&T system optimization, such as evaluating alternative P&T well network designs. Our findings highlight the potential of advanced machine learning models to significantly enhance the efficiency and sustainability of groundwater remediation efforts, offering a novel application of U-Net architecture in environmental science.

Pump-and-treat↗
DOE OSTI2023

Improved coarse-graining methods for two dimensional tensor networks including fermions

We show how to apply renormalization group algorithms incorporating entanglement filtering methods and a loop optimization to a tensor network which includes Grassmann variables which represent fermions in an underlying lattice field theory. As a numerical test a variety of quantities are calculated for two dimensional Wilson-Majorana fermions and for the two flavor Gross-Neveu model. The improved algorithms show much better accuracy for quantities such as the free energy and the determination of Fisher’s zeros.

Phase transitions↗
DOE OSTI2018

Optimal Power Scheduling for a Medium Voltage AC/DC Hybrid Distribution Network

With the great increase of renewable generation as well as the DC loads in the distribution network; DC distribution technology is receiving more attention; since the DC distribution network can improve operating efficiency and power quality by reducing the energy conversion stages. This paper presents a new architecture for the medium voltage AC/DC hybrid distribution network; where the AC and DC subgrids are looped by normally closed AC soft open point (ACSOP) and DC soft open point (DCSOP); respectively. The proposed AC/DC hybrid distribution systems contain renewable generation (i.e., wind power and photovoltaic (PV) generation); energy storage systems (ESSs); soft open points (SOPs); and both AC and DC flexible demands. An energy management strategy for the hybrid system is presented based on the dynamic optimal power flow (DOPF) method. The main objective of the proposed power scheduling strategy is to minimize the operating cost and reduce the curtailment of renewable generation while meeting operational and technical constraints. The proposed approach is verified in five scenarios. The five scenarios are classified as pure AC system; hybrid AC/DC system; hybrid system with interlinking converter; hybrid system with DC flexible demand; and hybrid system with SOPs. Results show that the proposed scheduling method can successfully dispatch the controllable elements; and that the presented architecture for the AC/DC hybrid distribution system is beneficial for reducing operating cost and renewable generation curtailment.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2018

Dynamic ADMM for Real-Time Optimal Power Flow

This paper considers distribution networks featuring distributed energy resources (DERs), and develops a dynamic optimization method to maximize given operational objectives in real time while adhering to relevant network constraints. The design of the dynamic algorithm is based on suitable linearization of the AC power flow equations, and it leverages the so-called alternating direction method of multipliers (ADMM). The steps of the ADMM, however, are suitably modified to accommodate appropriate measurements from the distribution network and the DERs. With the aid of these measurements, the resultant algorithm can enforce given operational constraints in spite of inaccuracies in the representation of the AC power flows, and it avoids ubiquitous metering to gather the state of noncontrollable resources. Optimality and convergence of the proposed algorithm are established in terms of tracking of the solution of a convex surrogate of the AC optimal power flow problem.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2018

Sparse Hardware Embedding of Spiking Neuron Systems for Community Detection

We study the applicability of spiking neural networks and neuromorphic hardware for solving general optimization problems without the use of adaptive training or learning algorithms. We leverage the dynamics of Hopfield networks and spin-glass systems to construct a fully connected spiking neural system to generate synchronous spike responses indicative of the underlying community structure in an undirected, unweighted graph. Mapping this fully connected system to current generation neuromorphic hardware is done by embedding sparse tree graphs to generate only the leading-order spiking dynamics. Here, we demonstrate that for a chosen set of benchmark graphs, the spike responses generated on a current generation neuromorphic processor can improve the stability of graph partitions and non-overlapping communities can be identified even with the loss of higher-order spiking behavior if the graphs are sufficiently dense. For sparse graphs, the loss of higher-order spiking behavior improves the stability of certain graph partitions but does not retrieve the known community memberships.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Real-Time Optimal Power Flow

Future power networks are expected to incorporate a large number of distributed energy resources, which introduce randomness and fluctuations as well as fast control capabilities. But traditional optimal power flow methods are only appropriate for applications that operate on a slow timescale. In this paper, we build on recent work to develop a real-time algorithm for AC optimal power flow, based on quasi-Newton methods. The algorithm uses second-order information to provide suboptimal solutions on a fast timescale, and can be shown to track the optimal power flow solution when the estimated second-order information is sufficiently accurate. We also give a specific implementation based on L-BFGS-B method, and show by simulation that the proposed algorithm has good performance and is computationally efficient.

Tang, Yujie↗
DOE OSTI2024

Improving ADAM through an implicit-explicit (IMEX) time-stepping approach

The ADAM optimizer, often used in machine learning for neural network training, corresponds to an underlying ordinary differential equation (ODE) in the limit of very small learning rates. Here, this work shows that the classical ADAM algorithm is a first-order implicit-explicit (IMEX) Euler discretization of the underlying ODE. Employing the time discretization point of view, we propose new extensions of the ADAM scheme obtained by using higher-order IMEX methods to solve the ODE. Based on this approach, we derive a new optimization algorithm for neural network training that performs better than classical ADAM on several regression and classification problems.

97 MATHEMATICS AND COMPUTING↗