Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Neural network embeddings”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

An image classification deep-learning algorithm for shrapnel detection from ultrasound images

Ultrasound imaging is essential for non-invasively diagnosing injuries where advanced diagnostics may not be possible. However, image interpretation remains a challenge as proper expertise may not be available. In response, artificial intelligence algorithms are being investigated to automate image analysis and diagnosis. Here, we highlight an image classification convolutional neural network for detecting shrapnel in ultrasound images. As an initial application, different shrapnel types and sizes were embedded first in a tissue mimicking phantom and then in swine thigh tissue. The algorithm architecture was optimized stepwise by minimizing validation loss and maximizing F1 score. The final algorithm design trained on tissue phantom image sets had an F1 score of 0.95 and an area under the ROC curve of 0.95. It maintained higher than a 90% accuracy for each of 8 shrapnel types. When trained only on swine image sets, the optimized algorithm format had even higher metrics: F1 and area under the ROC curve of 0.99. Overall, the algorithm developed resulted in strong classification accuracy for both the tissue phantom and animal tissue. This framework can be applied to other trauma relevant imaging applications such as internal bleeding to further simplify trauma medicine when resources and image interpretation are scarce.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Latent code-based fusion: A Volterra neural network approach

We propose a deep structure encoder using Volterra Neural Networks (VNNs) to seek a latent representation of multi-modal data whose features are jointly captured by a union of subspaces. The so-called self-representation embedding of the latent codes leads to a simplified fusion which is driven by a similarly constructed decoding. The Volterra Filter architecture achieved reduction in parameter complexity is primarily due to controlled non-linearities being introduced by the higher-order convolutions in lieu of generalized activation functions. Experimental results on two different datasets have shown a significant improvement in the clustering performance for VNNs auto-encoder over conventional Convolutional Neural Networks (CNNs) auto-encoder. In addition, we also show that the proposed approach demonstrates a much-improved sample complexity over CNN-based auto-encoder with a robust classification performance.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Gray-Box Modeling for Distribution Systems With Inverter-Based Resources: Integrating Physics-Based and Data-Driven Approaches

Here, in this paper, we develop a novel gray-box modeling approach for distribution systems with inverter-based resources (IBRs). The proposed gray-box modeling method aims to improve estimation accuracy by taking advantages of both physics-based (white-box) and data-driven (black-box) modeling approaches. To this end, we utilize partial physical knowledge of the system, including the inverters’ structures and control diagrams, as well as the equivalent network model simplified through Kron reduction. The white-box model containing unknown parameters is then constructed with mathematical equations and an optimization-based method is subsequently employed to identify these unknown parameters within the white-box model. Next, the graybox modeling framework is then constructed by embedding the output variables of the white-box model into the input vector of a black-box model (represented using a neural network). Finally, the black-box section is trained using the collected input-output datasets and the gray-box model is then obtained. Furthermore, case studies demonstrate that our gray-box modeling approach effectively improves estimation accuracy compared to purely physics-based or data-driven methods.

42 ENGINEERING↗
DOE OSTI2024

Deep Koopman Neural Network for Analyzing High-Energy-Density Simulations of Electrical Wire Explosions

Megaampere-scale electrical wire experiments (EWEs) provide a platform for studying magnetohydrodynamic (MHD) instability growth in magneto-inertial fusion (MIF) devices. Even when nonlinear simulations of these experiments can digitally reproduce much of the experimentally observed instability growth, interpreting the results and understanding mode growth and evolution can be non-trivial. As a first step toward providing better interpretation of these simulation features, this work investigates the use of a deep neural network that uses Koopman operator theory to analyze the dynamics of pulsed-power-driven explosions of EWEs. This deep neural network is trained on 1-D resistive MHD simulations of EWEs. This neural network learns to transform the nonlinear data into a lower-dimensional representation where the time dynamics are linear. Layers of this neural network are shown to learn features of the simulations, including the locations of shock waves and different physical regimes of the simulation. Using the learned features, the network can compress a time state of the simulation consisting of 5120 data point into a 36-parameter lower-dimensional latent space embedding. Furthermore, these embeddings are shown to be clustered in the latent space by initial radius and time state.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models

Abstract We evaluate the ability of machine learning to predict whether a hypothetical crystal structure can be synthesized and explain those predictions to scientists. Fine‐tuned large language models (LLMs) trained on a human‐readable text description of the target crystal structure perform comparably to previous bespoke convolutional graph neural network methods, but better prediction quality can be achieved by training a positive‐unlabeled learning model on a text‐embedding representation of the structure. An LLM‐based workflow can then be used to generate human‐readable explanations for the types of factors governing synthesizability, extract the underlying physical rules, and assess the veracity of those rules. These explanations can guide chemists in modifying or optimizing non‐synthesizable hypothetical structures to make them more feasible for materials design.

Kim, Seongmin [Department of Chemical and Biologic↗
DOE OSTI2025

Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models

Abstract We evaluate the ability of machine learning to predict whether a hypothetical crystal structure can be synthesized and explain those predictions to scientists. Fine‐tuned large language models (LLMs) trained on a human‐readable text description of the target crystal structure perform comparably to previous bespoke convolutional graph neural network methods, but better prediction quality can be achieved by training a positive‐unlabeled learning model on a text‐embedding representation of the structure. An LLM‐based workflow can then be used to generate human‐readable explanations for the types of factors governing synthesizability, extract the underlying physical rules, and assess the veracity of those rules. These explanations can guide chemists in modifying or optimizing non‐synthesizable hypothetical structures to make them more feasible for materials design.

Kim, Seongmin [Department of Chemical and Biologic↗
DOE OSTI2023

A deep learning upscaling framework: Reactive transport and mineral precipitation in fracture-matrix systems

Pore-scale modeling has limited applicability at large scales due to its high computational cost. One common approach to upscale pore-scale models is the use of effective medium theories, which homogenize small-scale features in a porous structure and characterize the medium by macroscale properties (e.g., permeability) and equations (e.g., Darcy’s law). However, there are classes of physical processes for which effective medium approximations may become inaccurate, e.g., mineral precipitation and clogging during reactive transport. We have developed a deep learning upscaling framework, in which pore-scale modeling is directly employed in macroscale systems, without relying on effective medium approximations. The upscaling framework is first developed for general multiscale systems and then applied to modeling reactive transport with mineral precipitation in the altered layer in fracture-matrix structures. Solute transport from the fractures to the matrix is modeled as a wall boundary condition for the fractures, which, in turn, is predicted by recurrent neural networks using the concentration histories at the fracture-matrix boundary. Specifically, we consider a meter-scale fracture network embedded in sandstones, where the smallest feature is at the micron scale. Here the proposed framework allows us to span five orders of magnitude in length scales by capturing mineral precipitation in the altered layer of the rock matrix at the pore scale across the entire meter-scale fracture network.

42 ENGINEERING↗
DOE OSTI2025

Simulating Atmospheric Processes in Earth System Models and Quantifying Uncertainties With Deep Learning Multi‐Member and Stochastic Parameterizations

Abstract Deep learning is a powerful tool to represent subgrid processes in climate models, but many application cases have so far used idealized settings and deterministic approaches. Here, we develop stochastic parameterizations with calibrated uncertainty quantification to learn subgrid convective and turbulent processes and surface radiative fluxes of a superparameterization embedded in an Earth System Model (ESM). We explore three methods to construct stochastic parameterizations: (a) a single Deep Neural Network (DNN) with Monte Carlo Dropout; (b) a multi‐member parameterization; and (c) a Variational Encoder Decoder with latent space perturbation. We show that the multi‐member parameterization improves the representation of convective processes, especially in the planetary boundary layer, compared to individual DNNs. The respective uncertainty quantification illustrates that methods (b) and (c) are advantageous compared to a dropout‐based DNN parameterization regarding the spread of convective processes. Hybrid simulations with our best‐performing multi‐member parameterizations remained challenging and crash within the first days. Therefore, we develop a pragmatic partial coupling strategy relying on the superparameterization for condensate emulation. Partial coupling reduces the computational efficiency of hybrid Earth‐like simulations but enables model stability over 5 months with our multi‐member parameterizations. However, our hybrid simulations exhibit biases in thermodynamic fields and differences in precipitation patterns. Despite this, the multi‐member parameterizations enable improvements in reproducing tropical extreme precipitation compared to a traditional convection parameterization. Despite these challenges, our results indicate the potential of a new generation of multi‐member machine learning parameterizations leveraging uncertainty quantification to improve the representation of stochasticity of subgrid effects.

Behrens, Gunnar [Deutsches Zentrum für Luft‐ und R↗
DOE OSTI2026

SBND Shower Reconstruction with SPINE

The Short-Baseline Near Detector (SBND) is a liquid argon time projection chamber (LArTPC) neutrino detector in the Short-Baseline Neutrino (SBN) program at Fermilab. SBND is designed to investigate the Low-Energy Excess (LEE), an unexplained excess of electron-like events observed by previous short-baseline neutrino experiments that may point to physics beyond the Standard Model. In LArTPC detectors, precise shower reconstruction is essential for distinguishing electrons from photons, a key requirement for testing possible explanations of the LEE and improving $\nu_e$ event selection. In this poster, the reconstruction studies using the Scalable Particle Imaging with Neural Embeddings (SPINE), a machine learning based reconstruction framework for particle imaging detectors will be presented. SPINE combines sparse convolutional neural networks (CNN) and graph neural networks (GNN) to enable detailed reconstruction and characterization of neutrino interactions in LArTPC detectors. Shower calorimetry and kinematic reconstruction are performed in dedicated post-processing stages. Strong agreement between data and Monte Carlo simulation will be demonstrated, indicating high-precision detector calibration and reconstruction performance. The agreement between reconstructed and true electron shower energy will also be discussed, emphasizing the robustness of the shower reconstruction performance. These results demonstrate the unprecedented precision achievable with SPINE in SBND, highlighting their potential for future high-resolution neutrino measurements.

Fan, Castaly [Florida U.; Fermilab] (ORCID:0000000↗
DOE OSTI2022

Graph Neural Networks for Charged Particle Tracking on FPGAs

The determination of charged particle trajectories in collisions at the CERN Large Hadron Collider (LHC) is an important but challenging problem, especially in the high interaction density conditions expected during the future high-luminosity phase of the LHC (HL-LHC). Graph neural networks (GNNs) are a type of geometric deep learning algorithm that has successfully been applied to this task by embedding tracker data as a graph—nodes represent hits, while edges represent possible track segments—and classifying the edges as true or fake track segments. However, their study in hardware- or software-based trigger applications has been limited due to their large computational cost. In this paper, we introduce an automated translation workflow, integrated into a broader tool called hls4ml , for converting GNNs into firmware for field-programmable gate arrays (FPGAs). We use this translation tool to implement GNNs for charged particle tracking, trained using the TrackML challenge dataset, on FPGAs with designs targeting different graph sizes, task complexites, and latency/throughput requirements. This work could enable the inclusion of charged particle tracking GNNs at the trigger level for HL-LHC experiments.

Elabd, Abdelrahman↗
DOE OSTI2023

Systems and methods for customizing kernel machines with deep neural networks

A method including receiving an input data set. The input data set can include one of a feature domain set or a kernel matrix. The method also can include constructing dense embeddings using: (i) Nyström approximations on the input data set when the input data set comprises the kernel matrix, and (ii) clustered Nyström approximations on the input data set when the input data set comprises the feature domain set. The method additionally can include performing representation learning on each of the dense embeddings using a multi-layer fully-connected network for each of the dense embeddings to generate latent representations corresponding to each of the dense embeddings. The method further can include applying a fusion layer to the latent representations corresponding to the dense embeddings to generate a combined representation. The method additionally can include performing classification on the combined representation. Other embodiments of related systems and methods are also disclosed.

Song, Huan↗
DOE OSTI2021

Reduced-order modeling of advection-dominated systems with recurrent neural networks and convolutional autoencoders

A common strategy for the dimensionality reduction of nonlinear partial differential equations (PDEs) relies on the use of the proper orthogonal decomposition (POD) to identify a reduced subspace and the Galerkin projection for evolving dynamics in this reduced space. However, advection-dominated PDEs are represented poorly by this methodology since the process of truncation discards important interactions between higher-order modes during time evolution. In this study, we demonstrate that encoding using convolutional autoencoders (CAEs) followed by a reduced-space time evolution by recurrent neural networks overcomes this limitation effectively. We demonstrate that a truncated system of only two latent space dimensions can reproduce a sharp advecting shock profile for the viscous Burgers equation with very low viscosities, and a six-dimensional latent space can recreate the evolution of the inviscid shallow water equations. Additionally, the proposed framework is extended to a parametric reduced-order model by directly embedding parametric information into the latent space to detect trends in system evolution. Furthermore, our results show that these advection-dominated systems are more amenable to low-dimensional encoding and time evolution by a CAE and recurrent neural network combination than the POD-Galerkin technique.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Deep learning of parameterized equations with applications to uncertainty quantification

We propose a learning algorithm for discovering unknown parameterized dynamical systems by using observational data of the state variables. Our method is built upon and extends the recent work of discovering unknown dynamical systems, in particular those using deep neural network (DNN). We propose a DNN structure, largely based upon the residual network (ResNet), to not only learn the unknown form of the governing equation but also take into account the random effect embedded in the system, which is generated by the random parameters. Once the DNN model is successfully constructed, it is able to produce system prediction over longer term and for arbitrary parameter values. For uncertainty quantification, it allows us to conduct uncertainty analysis by evaluating solution statistics over the parameter space.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Graph Convolutional Network-Based Topology Embedded Deep Reinforcement Learning for Voltage Stability Control

Topological variations in power system is a common phenomenon and can impose significant challenges to traditional controllers of power system. Recent study revealed the strength of deep reinforcement learning (DRL) based approaches in power system preventive and corrective control. But topological variations are difficult to capture using classical fully connected neural network (FCN) model and has not been explicitly modeled in previous work. Hence, we develop a Graph Convolutional Network (GCN) based DRL framework to tackle topology changes in control design of power system. The GCN model exploits the graph structure of the power network and helps the DRL agent to embed the topology information during learning process. Our GCN based approach is evaluated using the IEEE-39 bus system and it outperforms the FCN-based DRL scheme in terms of training convergence and control performance considering grid topology changes.

Hossain, Ramij Raja↗
DOE OSTI2021

Physics-informed machine learning of the Lagrangian dynamics of velocity gradient tensor

Reduced models describing the Lagrangian dynamics of the velocity gradient tensor (VGT) in homogeneous isotropic turbulence (HIT) are developed under the physics-informed machine learning (PIML) framework. We consider the VGT at both Kolmogorov scale and coarse-grained scale within the inertial range of HIT. Building reduced models requires resolving the pressure Hessian and subfilter contributions, which is accomplished by constructing them using the integrity bases and invariants of the VGT. The developed models can be expressed using the extended tensor basis neural network (TBNN) introduced by Ling et al. [J. Fluid Mech. 807, 155 (2016)]. Physical constraints, such as Galilean invariance, rotational invariance, and incompressibility condition, are thus embedded in the models explicitly. Our PIML models are trained on the Lagrangian data from a high-Reynolds number direct numerical simulation (DNS). To validate the results, we perform a comprehensive out-of-sample test. We observe that the PIML model provides an improved representation for the magnitude and orientation of the small-scale pressure Hessian contributions. Statistics of the flow, as indicated by the joint PDF of second and third invariants of the VGT, show good agreement with the “ground-truth” DNS data. A number of other important features describing the structure of HIT are reproduced by the model successfully. We have also identified challenges in modeling inertial range dynamics, which indicates that a richer modeling strategy is required. This helps us identify important directions for future research, in particular towards including inertial range geometry into the TBNN.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

Enhancing Network Anomaly Detection Using Graph Neural Networks

In the world of Internet of Things (IoT) networks, where devices are constantly communicating, keeping them secure from cyber threats is critical. This paper introduces a novel approach to detecting unusual and potentially harmful activities in these networks using graph neural networks (GNNs). We combine two specific types of GNNs-GraphSAGE and graph attention networks (GAT)-to create a model that understands and represents the behaviors and interactions in a network. GraphSAGE creates an embedding of network activities by examining local data interactions, while GAT directs the model's focus to the most critical interactions. By integrating these two methods in a single model that considers different types of interactions (both host and flow nodes), we aim to create a system that accurately represents the current state of a network and can also spot anomalies effectively while reducing false positives and negatives. Our innovative approach has demonstrated promising results, achieving an accuracy of 98% on the UNSW-NB15 dataset, significantly outperforming standalone GraphSAGE and GAT models. This underscores its potential as a robust framework for securing IoT networks against cyber threats and anomalies.

Marfo, William↗
DOE OSTI2025

FPGA-accelerated SpeckleNN with SNL for real-time X-ray single-particle imaging

We present the implementation of a specialized version of our previously published unified embedding model, SpeckleNN, for real-time speckle pattern classification in X-ray Single-Particle Imaging (SPI), using the SLAC Neural Network Library (SNL) on an FPGA platform. This hardware realization transitions SpeckleNN from a prototypic model into a practical edge solution, optimized for running inference near the detector in high-throughput X-ray free-electron laser (XFEL) facilities, such as those found at the Linac Coherent Light Source (LCLS). To address the resource constraints inherent in FPGAs, we developed a more specialized version of SpeckleNN. The original model, which was designed for broader classification across multiple biological samples, comprised ~5.6 million parameters. The new implementation, while reducing the parameter count to 64.6K (a 98.8% reduction), focuses on maintaining the model's essential functionality for real-time operation, achieving an accuracy of 90%. Furthermore, we compressed the latent space from 128 to 50 dimensions. This implementation was demonstrated on the KCU1500 FPGA board, utilizing 71% of available DSPs, 75% of LUTs, and 48% of FFs, with an average power consumption of 9.4W according to the Vivado post-implementation report. The FPGA performed inference on a single image with a latency of 45.015 microseconds at a 200 MHz clock rate. In comparison, running the same inference on an NVIDIA A100 GPU resulted in an average power consumption of ~73W and an image processing latency of around 400 microseconds. Our FPGA-accelerated version of SpeckleNN demonstrated significant improvements, achieving an 8.9 × speedup and a 7.8 × reduction in power consumption compared to the GPU implementation. Key advancements include model specialization and dynamic weight loading through SNL, which eliminates the need for time-consuming FPGA design re-synthesis, allowing fast and continuous deployment of models (re)trained online. These innovations enable real-time adaptive classification and efficient vetoing of speckle patterns, making SpeckleNN more suited for deployment in XFEL facilities. This implementation has the potential to significantly accelerate SPI experiments and enhance adaptability to evolving experimental conditions.

47 OTHER INSTRUMENTATION↗
DOE OSTI2023

GRIDS-Net: Inverse shape design and identification of scatterers via geometric regularization and physics-embedded deep learning

This study presents a deep learning based methodology for both remote sensing and design of acoustic scatterers. The ability to determine the shape of a scatterer, either in the context of material design or sensing, plays a critical role in many practical engineering problems. This class of inverse problems is extremely challenging due to their high-dimensional, nonlinear, and ill-posed nature. To overcome these technical hurdles, we introduce a geometric regularization approach for deep neural networks (DNN) based on non-uniform rational B-splines (NURBS) and capable of predicting complex 2D scatterer geometries in a parsimonious dimensional representation. Then, this geometric regularization is combined with physics-embedded learning and integrated within a robust convolutional autoencoder (CAE) architecture to accurately predict the shape of 2D scatterers in the context of identification and inverse design problems. Further, an extensive numerical study is presented in order to showcase the remarkable ability of this approach to handle complex scatterer geometries while generating physically-consistent acoustic fields. The study also assesses and contrasts the role played by the (weakly) embedded physics in the convergence of the DNN predictions to a physically consistent inverse design.

42 ENGINEERING↗