Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Scalable deep learning”

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 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Using a physics-informed neural network and fault zone acoustic monitoring to predict lab earthquakes

Abstract Predicting failure in solids has broad applications including earthquake prediction which remains an unattainable goal. However, recent machine learning work shows that laboratory earthquakes can be predicted using micro-failure events and temporal evolution of fault zone elastic properties. Remarkably, these results come from purely data-driven models trained with large datasets. Such data are equivalent to centuries of fault motion rendering application to tectonic faulting unclear. In addition, the underlying physics of such predictions is poorly understood. Here, we address scalability using a novel Physics-Informed Neural Network (PINN). Our model encodes fault physics in the deep learning loss function using time-lapse ultrasonic data. PINN models outperform data-driven models and significantly improve transfer learning for small training datasets and conditions outside those used in training. Our work suggests that PINN offers a promising path for machine learning-based failure prediction and, ultimately for improving our understanding of earthquake physics and prediction.

42 ENGINEERING↗
DOE OSTI2024

Astronomaly at scale: searching for anomalies amongst 4 million galaxies

ABSTRACT Modern astronomical surveys are producing data sets of unprecedented size and richness, increasing the potential for high-impact scientific discovery. This possibility, coupled with the challenge of exploring a large number of sources, has led to the development of novel machine-learning-based anomaly detection approaches, such as astronomaly. For the first time, we test the scalability of astronomaly by applying it to almost 4 million images of galaxies from the Dark Energy Camera Legacy Survey. We use a trained deep learning algorithm to learn useful representations of the images and pass these to the anomaly detection algorithm isolation forest, coupled with astronomaly’s active learning method, to discover interesting sources. We find that data selection criteria have a significant impact on the trade-off between finding rare sources such as strong lenses and introducing artefacts into the data set. We demonstrate that active learning is required to identify the most interesting sources and reduce artefacts, while anomaly detection methods alone are insufficient. Using astronomaly, we find 1635 anomalies among the top 2000 sources in the data set after applying active learning, including eight strong gravitational lens candidates, 1609 galaxy merger candidates, and 18 previously unidentified sources exhibiting highly unusual morphology. Our results show that by leveraging the human–machine interface, astronomaly is able to rapidly identify sources of scientific interest even in large data sets.

Astronomy & Astrophysics↗
DOE OSTI2024

MDLoader: A Hybrid Model-Driven Data Loader for Distributed Graph Neural Network Training

Scalable data management is essential for processing large scientific dataset on HPC platforms for distributed deep learning. In-memory distributed storage is preferred for its speed, enabling rapid, random, and frequent data access required by stochastic optimizers. Processes use one-sided or collective communication to fetch remote data, with optimal performance depending on (i) dataset characteristics, (ii) training scale, and (iii) interconnection network. Empirical analysis shows collective communication excels with larger mini-batch sizes and/or fewer processes, whereas one-sided communication outperforms at larger scales. We propose MDLoader, a hybrid in-memory data loader for distributed graph neural network training. MDLoader features a model-driven performance estimator that dynamically selects between one-sided and collective communication at the beginning of training using Tree of Parzen Estimators (TPE). Evaluations on NERSC Perlmutter and OLCF Summit show MDLoader outperforms single-backend loaders by up to 2.83 × and predicts the suitable communication method with 96.3% (Perlmutter) and 94.3% (Summit) success rate.

Bae, Jonghyun↗
DOE OSTI2023

DDStore: Distributed Data Store for Scalable Training of Graph Neural Networks on Large Atomistic Modeling Datasets

Graph neural networks (GNNs) are a class of Deep Learning models used in designing atomistic materials for effective screening of large chemical spaces. To ensure robust prediction, GNN models must be trained on large volumes of atomistic data on leadership class supercomputers. Even with the advent of modern architectures that consist of multiple storage layers that include node-local NVMe devices in addition to device memory for caching large datasets, extreme-scale model training faces I/O challenges at scale.We present DDStore, an in-memory distributed data store designed for GNN training on large-scale graph data. DDStore provides a hierarchical, distributed, data caching technique that combines data chunking, replication, low-latency random access, and high throughput communication. DDStore achieves near-linear scaling for training a GNN model using up to 1000 GPUs on the Summit and Perlmutter supercomputers, and reaches up to a 6.15x reduction in GNN training time compared to state-of-the-art methodologies.

Choi, Jong Youl↗
DOE OSTI2021

InversionNet3D: Efficient and Scalable Learning for 3-D Full-Waveform Inversion

Seismic full-waveform inversion (FWI) techniques aim to find a high-resolution subsurface geophysical model provided with waveform data. Some recent effort in data-driven FWI has shown some encouraging results in obtaining 2-D velocity maps. However, due to high computational complexity and large memory consumption, the reconstruction of 3-D high-resolution velocity maps via deep networks is still a great challenge. Here, in this article, we present InversionNet3D (InvNet3D), an efficient and scalable encoder–decoder network for 3-D FWI. The proposed method employs group convolution in the encoder to establish an effective hierarchy for learning information from multiple sources while cutting down unnecessary parameters and operations at the same time. The introduction of invertible layers further reduces the memory consumption of intermediate features during training and, thus, enables the development of deeper networks with more layers and higher capacity as required by different application scenarios. Experiments on the 3-D Kimberlina dataset demonstrate that InvNet3D achieves state-of-the-art reconstruction performance with lower computational cost and lower memory footprint compared to the baseline.

58 GEOSCIENCES↗
DOE OSTI2025

HAMscope: a snapshot Hyperspectral Autofluorescence Miniscope for real-time molecular imaging

We introduce HAMscope, a compact, snapshot hyperspectral autofluorescence miniscope that enables real-time, label-free molecular imaging in a wide range of biological systems. By integrating a thin polymer diffuser into a widefield miniscope, HAMscope spectrally encodes each frame and employs a probabilistic deep learning framework to reconstruct 30-channel hyperspectral stacks (452-703 nm) or directly infer molecular composition maps from single images. A scalable multi-pass U-Net architecture with transformer-based attention and per pixel uncertainty estimation enables high spatio-spectral fidelity (mean absolute error ∼0.0048) at video rates. While initially demonstrated in plant systems, including lignin, chlorophyll, and suberin imaging in intact poplar and cork tissues, the platform is readily adaptable to other applications such as neural activity mapping, metabolic profiling, and histopathology. We show that the system generalizes to out-of-distribution tissue types and supports direct molecular mapping without the need for spectral unmixing. HAMscope establishes a general framework for compact, uncertainty-aware spectral imaging that combines minimal optics with advanced deep learning, offering broad utility for real-time biochemical imaging across neuroscience, environmental monitoring, and biomedicine.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Resilience Enhancements through Deep Learning Yields

This report documents the Resilience Enhancements through Deep Learning Yields (REDLY) project, a three-year effort to improve electrical grid resilience by developing scalable methods for system operators to protect the grid against threats leading to interrupted service or physical damage. The computational complexity and uncertain nature of current real-world contingency analysis presents significant barriers to automated, real-time monitoring. While there has been a significant push to explore the use of accurate, high-performance machine learning (ML) model surrogates to address this gap, their reliability is unclear when deployed in high-consequence applications such as power grid systems. Contemporary optimization techniques used to validate surrogate performance can exploit ML model prediction errors, which necessitates the verification of worst-case performance for the models.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

PWR loading pattern optimization with reinforcement learning

The core loading pattern optimization problem belongs to the class of combinatorial optimization problem and has been studied since the dawn of commercial nuclear energy industry. It is characterized by multiple objectives and constraints, with a very high number of candidate patterns, which makes it impossible to solve explicitly. Stochastic optimization methodologies including Genetic Algorithms and Simulated Annealing are used by different nuclear utilities and vendors to perform fuel cycle reload design. Nevertheless, hand-designed solutions continue to be the prevalent method in the industry. To improve the state-of-the-art core reload patterns, we aim to create a method as scalable as possible, that agrees with the designer's goal of performance and safety. To help in this task Deep Reinforcement Learning (DRL), in particular Proximal Policy Optimization is leveraged. DRL has recently experienced a strong impetus from its successes applied to games, sometimes even reaching 'super-human' performances. This paper lays out the foundation of this method and proposes to study the behavior of several hyper-parameters that influence the DRL algorithm. The algorithm is highly dependent on multiple factors such as an exploration/exploitation trade-off that manifests through different parameters such as the number of loading patterns seen and the number of samples collected before a policy update, but also the shape of the objective function derived for the core design. Experimental results also demonstrate the effectiveness of the method in finding high-quality solutions from scratch within a reasonable amount of time. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2024

Fault-Tolerant Deep Learning Cache with Hash Ring for Load Balancing in HPC Systems

Large-scale DL on HPC systems like Frontier and Summit uses distributed node-local caching to address scalability and performance challenges. However, as these systems grow more complex, the risk of node failures increases, and current caching approaches lack fault tolerance, jeopardizing large-scale training jobs. We analyzed six months of SLURM job logs from Frontier and found that over 30% of jobs failed after an average of 75 minutes. To address this, we propose fault-tolerance strategies that recache data lost from failed nodes using a hash ring technique for balanced data recaching in the distributed node-local caching, reducing reliance on the PFS. Our extensive evaluations on Frontier showed that the hash ring-based recaching approach reduced training time by approximately 25% compared to the approach that redirects I/O to the PFS after node failures and demonstrated effective load balancing of training data across nodes.

Lee, Seoyeong↗
DOE OSTI2025

FIRE: A Failure-Adaptive RL Framework for Edge Computing Migrations

In edge computing, users' service profiles are migrated between edge servers due to user mobility. Reinforcement Learning (RL) frameworks have been proposed to do so, often trained on simulated data. However, existing RL frameworks overlook occasional server failures, which although rare, impact latency-sensitive applications like AR/VR and real- time obstacle detection. These rare failures, being not adequately represented in historical training data, pose a challenge for data-driven RL algorithms. We introduce FIRE, a framework that adapts to rare events by training a RL policy in an edge computing digital twin environment. We propose FIRE-ImRE, an importance sampling-based Q-learning algorithm, which samples rare events proportionally to their impact on the value function. FIRE considers delay, migration, failure, and backup placement costs across individual and shared service profiles. We prove FIRE-ImRE's boundedness and convergence to optimality. Next, we introduce novel deep Q-learning (FIRE-ImDQL) and actor critic (FIRE-ImACRE) versions of our algorithm to enhance scalability. Here, we extend our framework to accommodate users with varying risk tolerances of rare failure events. Through trace-driven experiments, we show that FIRE reduces edge computing costs compared to vanilla RL and the greedy baseline in the event of failures.

Edge computing↗
DOE OSTI2022

Scalable training of graph convolutional neural networks for fast and accurate predictions of HOMO-LUMO gap in molecules

Abstract Graph Convolutional Neural Network (GCNN) is a popular class of deep learning (DL) models in material science to predict material properties from the graph representation of molecular structures. Training an accurate and comprehensive GCNN surrogate for molecular design requires large-scale graph datasets and is usually a time-consuming process. Recent advances in GPUs and distributed computing open a path to reduce the computational cost for GCNN training effectively. However, efficient utilization of high performance computing (HPC) resources for training requires simultaneously optimizing large-scale data management and scalable stochastic batched optimization techniques. In this work, we focus on building GCNN models on HPC systems to predict material properties of millions of molecules. We use HydraGNN, our in-house library for large-scale GCNN training, leveraging distributed data parallelism in PyTorch. We use ADIOS, a high-performance data management framework for efficient storage and reading of large molecular graph data. We perform parallel training on two open-source large-scale graph datasets to build a GCNN predictor for an important quantum property known as the HOMO-LUMO gap. We measure the scalability, accuracy, and convergence of our approach on two DOE supercomputers: the Summit supercomputer at the Oak Ridge Leadership Computing Facility (OLCF) and the Perlmutter system at the National Energy Research Scientific Computing Center (NERSC). We present our experimental results with HydraGNN showing (i) reduction of data loading time up to 4.2 times compared with a conventional method and (ii) linear scaling performance for training up to 1024 GPUs on both Summit and Perlmutter.

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

Operator learning for energy-efficient building ventilation control with computational fluid dynamics simulation of a real-world classroom

Energy-efficient ventilation control plays an important role in reducing building energy consumption while ensuring occupant health and comfort. While Computational Fluid Dynamics (CFD) simulations provide detailed and physically accurate representations of indoor airflow, their high computational cost limits their use in real-time building control. In this work, we present a neural operator learning framework that combines the physical accuracy of CFD with the computational efficiency of machine learning to enable building ventilation control with the high-fidelity fluid dynamics models. Our method jointly optimizes the airflow supply rates and vent angles to reduce energy use and adhere to air quality constraints. We train an ensemble of neural operator transformer models to learn the mapping from building control actions to airflow fields using high-resolution CFD data. This learned neural operator is then embedded in an optimization-based control framework for building ventilation control. Experimental results show that our approach achieves significant energy savings compared to maximum airflow rate control, rule-based control, as well as data-driven control methods using spatially averaged CO 2 prediction and deep learning–based reduced-order models, while consistently maintaining safe indoor air quality. These results highlight the practicality and scalability of our method in maintaining energy efficiency and indoor air quality in real-world buildings.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Digital Real-Time Simulation and Power Quality Analysis of a Hydrogen-Generating Nuclear-Renewable Integrated Energy System

This paper investigates the challenges and solutions associated with integrating a hydrogen-generating nuclear-renewable integrated energy system (NR-IES) under a transactive energy framework. The proposed system directs excess nuclear power to hydrogen production during periods of low grid demand while utilizing renewables to maintain grid stability. Using digital real-time simulation (DRTS) in the Typhoon HIL 404 model, the dynamic interactions between nuclear power plants, electrolyzers, and power grids are analyzed to mitigate issues such as harmonic distortion, power quality degradation, and low power factor caused by large non-linear loads. A three-phase power conversion system is modeled using the Typhoon HIL 404 model and includes a generator, a variable load, an electrolyzer, and power filters. Active harmonic filters (AHFs) and hybrid active power filters (HAPFs) are implemented to address harmonic mitigation and reactive power compensation. The results reveal that the HAPF topology effectively balances cost efficiency and performance and significantly reduces active filter current requirements compared to AHF-only systems. During maximum electrolyzer operation at 4 MW, the grid frequency dropped below 59.3 Hz without filtering; however, the implementation of power filters successfully restored the frequency to 59.9 Hz, demonstrating its effectiveness in maintaining grid stability. Future work will focus on integrating a deep reinforcement learning (DRL) framework with real-time simulation and optimizing real-time power dispatch, thus enabling a scalable, efficient NR-IES for sustainable energy markets.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2026

NanoPSD: A software for automatic detection of Nano-Particle Shape Distribution in electron microscopy images

Accurate quantification of the size and morphology of nanoparticles from electron microscopy (EM) images is essential to understand growth mechanisms, surface reactivity, and functional behavior in nanoscale materials. Manual analysis remains slow, subjective, and difficult to reproduce in large datasets. We introduce NanoPSD (Nano-Particle Shape Distribution), an open-source and fully automated framework for quantitative particle detection and morphology analysis from EM images. NanoPSD integrates adaptive contrast enhancement, polarity-agnostic scale-bar detection, Optical Character Recognition (OCR)-based calibration, and classical segmentation via Otsu thresholding with morphological refinement. Particle contours are used to extract geometric descriptors, including equivalent circular diameter, aspect ratio, circularity, and solidity, enabling automated classification into spherical, rod-like, and aggregate morphologies. The framework supports both single-image and batch processing, generating publication-quality visualizations, LaTeX-ready tables, and structured comma-separated values (CSV) datasets. As a demonstration, we applied NanoPSD to plasma-synthesized nanoparticle samples diagnosed via transmission electron microscopy (TEM). The code produced statistically robust size and morphology distributions spanning a few to tens of nanometers with minimal user supervision. The pipeline demonstrates high reproducibility and scalability, processing large image collections with consistent calibration and output formatting. Its modular design enables seamless integration of future deep-learning-based segmentation models, providing a pathway toward intelligent, data-driven electron microscopy analysis.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Deep Learning and Photogrammetric Reconstruction for Automated Crack Detection and Dimensional Measurement in Mining Operations

Surface crack detection and dimensional measurement at active mining sites present significant safety and operational challenges. Manual inspection methods are labor-intensive, spatially incomplete, and expose personnel to hazardous environments, while existing automated approaches have been developed primarily for concrete civil infrastructure and have not been validated on the complex, variable surfaces characteristic of mining environments. This dissertation presents an automated pipeline that integrates deep learning semantic segmentation with Structure-from-Motion photogrammetry to detect surface cracks and measure their aperture, length, and vertical displacement from standard RGB imagery acquired during routine Uncrewed Aerial Vehicle (UAV) survey operations, without requiring additional sensor hardware or manual measurement. The pipeline combines a U-Net architecture with an EfficientNet-B0 encoder, pretrained on the SDNET2018 concrete crack dataset and fine-tuned on a mining-specific dataset spanning laboratory concrete specimens, coal refuse impoundment embankments, and post-blast limestone quarry benches. Photogrammetric reconstruction is performed using COLMAP Structure-from-Motion and Multi-View Stereo, with crack segmentation masks projected into the reconstructed point cloud to enable three-dimensional vertical displacement measurement through local plane fitting and bimodal surface detection. The pipeline was validated across 36 controlled laboratory specimens at three imaging distances and four vertical displacement levels, achieving aperture measurement RMSE of 0.047 cm and R² of 0.954, and vertical displacement RMSE of 0.140 cm and R² of 0.966, against independent caliper measurements. Field application at a coal refuse impoundment in southwestern Pennsylvania detected 71 crack components across the embankment crest, with a dominant longitudinal crack exhibiting aperture values reaching 28 cm and a 95th percentile vertical displacement of 35.53 cm, consistent in magnitude and spatial distribution with simultaneously acquired LiDAR-derived estimates. Application across four post-blast limestone quarry bench datasets in California successfully characterized blast-induced fracture networks at ground sampling distances ranging from 0.59 to 1.23 cm/pixel, with detected crack geometries physically consistent with observable surface conditions at each site. The results demonstrate that deep learning-based crack detection and photogrammetric measurement can be integrated into routine UAV inspection workflows at mining sites, providing repeatable, scalable, and quantitative crack characterization across surface types, crack scales, and displacement magnitudes not previously addressed in the literature. The pipeline requires no dedicated surveying equipment beyond the UAV platforms already deployed at mine sites for survey and monitoring purposes, supporting practical adoption within existing operational workflows.

Crack detection, Dimensional Measurement↗
DOE OSTI2026

Dark Energy Survey Year 3 results: Simulation-based 𝑤CDM inference from weak lensing and galaxy clustering maps with deep learning: Analysis design

Data-driven approaches using deep learning are emerging as powerful techniques to extract non-Gaussian information from cosmological large-scale structure. Here, this work presents the first simulation-based inference (SBI) pipeline that combines weak lensing and galaxy clustering maps in a realistic Dark Energy Survey Year 3 (DES Y3) configuration and serves as preparation for a forthcoming analysis of the survey data. We develop a scalable forward model based on the CosmoGridV1 suite of N-body simulations to generate over one million self-consistent mock realizations of DES Y3 at the map level. Leveraging this large dataset, we train deep graph convolutional neural networks on the full survey footprint in spherical geometry to learn low-dimensional features that approximately maximize mutual information with target parameters. These learned compressions enable neural density estimation of the implicit likelihood via normalizing flows in a ten-dimensional parameter space spanning cosmological 𝑤CDM, intrinsic alignment, and linear galaxy bias parameters, while marginalizing over baryonic, photometric redshift, and shear bias nuisances. To ensure robustness, we extensively validate our inference pipeline using synthetic observations derived from both systematic contaminations in our forward model and independent Buzzard galaxy catalogs. Our forecasts yield significant improvements in cosmological parameter constraints, achieving 2−3× higher figures of merit in the 𝛺 𝑚 − 𝑆 8 plane relative to our implementation of baseline two-point statistics and effectively breaking parameter degeneracies through probe combination. These results demonstrate the potential of SBI analyses powered by deep learning for upcoming Stage-IV wide-field imaging surveys.

Thomsen, A. [Zurich, ETH] (ORCID:0000000203099021)↗
DOE OSTI2022

Introduction to Special Section: Machine Learning for Image-based Geologic Interpretation

Image-based geological interpretation has been a labor-intensive and time-consuming process because it requires well-trained geoscientists to identify geological structures, features, and textures from various types of images. These images include scanning electron microscopic images, optical microscopic images, optical photos, resistivity images, seismic volumes, remote-sensing images, etc. With fast-evolving machine learning (ML) technology and computing power in recent decades, computers can achieve nearhuman-level to super-human-level performance with scalable high efficiency in the computer vision field. These technological revolutions facilitated image-based geological interpretation in petroleum exploration and production. For example, a fault picking method applied to 3-D seismic volume data using deep learning can achieve superior performance in comparison to conventional auto-picking methods. In addition, under the new normal of low oil prices, the petroleum industry seeks cost-effective strategies such as automating traditionally labor-intensive processes. Nevertheless, the potential of applying ML to geological image interpretation is still facing a few key challenges including data scarcity, data distribution, poor data and/or label quality, data leakage, learning algorithms, model architecture, training methodologies, testing and evaluation metrics, hyper-parameters optimization, model drift, production deployment, and the like.

58 GEOSCIENCES↗