Results for “deep generative models” — Page 7

DOE OSTI2024

Exploiting Multi-Domain Features for Detection of Unclassified Electromagnetic Signals (Presentation)

Deep Learning based classification techniques have shown excellent performance in static environments, where the training and testing samples are drawn from the same distribution. However, real world scenarios often present samples that do not belong to the known set of classes chosen during training. This is quite common for electromagnetic signals, where it is impractical to assume that all possible waveforms are known a-priori, specially in scenarios like warfare. To address this problem, we propose a deep learning based adversarial model where the generator learns to generate waveform features that can deceive the discriminator model as true samples. We introduce domain knowledge of wireless signals by decomposing the signal into a lower dimensional unique feature set, which is used for classifying known versus unknown signals. We further introduce multiple domain representations of the signal to extract features and combine them together to accurately classify new waveforms as an unknown class. Our results show that combined features from multiple domains outperform any single domain representation, especially at low SNR regimes with fewer number of samples to classify.

99 - GENERAL AND MISCELLANEOUS↗

DOE OSTI2024

Deep operator network surrogate for phase-field modeling of metal grain growth during solidification

A deep operator network (DeepONet) has been constructed that generates accurate representations of phase-field model simulations for evolving two dimensional metal grain morphology growing from melt. These representations serve as lower resolution, computationally efficient stand-ins for quick parameter space exploration of solutions to the the Allen-Cahn equations that dictate the phase-field model simulations. The experimental target for the phase-field model is a uranium casting system cooling a 434 g uranium charge from a maximum temperature of 1400° C at an average rate of 30° C / min , traversing the crystallographic phases of the pure metal. Experimental parameters inform the phase-field model, whose higher resolution computational model solutions are used to train the DeepONet in a given parameter space with the aim of developing a faster, more efficient method for predicting the solidifying metal's microstructure at different potential experimental values. The final DeepONet generates high accuracy, lower resolution predictions with cumulative relative approximation error over all timesteps of less than 0.5%, while ensuring solutions remain within physically feasible ranges. Further, these relative error values are comparable with other state-of-the-art DeepONet models for microstructure evolution, while significantly reducing the amount of training data required. Training a convolutional neural network simultaneously with the DeepONet, enforcing realistic values at the complex metal grain boundaries, and mathematically encoding boundary conditions into the structure of the DeepONet improved prediction accuracy and computational efficiency over a standard DeepONet model.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Predicting fusion ignition at the National Ignition Facility with physics-informed deep learning

Here, an inertial confinement fusion experiment, carried out at the National Ignition Facility, has achieved ignition by generating fusion energy exceeding the laser energy that drove the experiment. Prior to the experiment, a generative machine learning model that combines radiation hydrodynamics simulations, deep learning, experimental data, and Bayesian statistics was used to predict, with a probability greater than 70%, that ignition was the most likely outcome for this shot.

Spears, Brian K. [Lawrence Livermore National Labo↗

NASA NTRS2013

Next Generation Lightweight Mirror Modeling Software

The advances in manufacturing techniques for lightweight mirrors, such as EXELSIS deep core low temperature fusion, Corning's continued improvements in the Frit bonding process and the ability to cast large complex designs, combined with water-jet and conventional diamond machining of glasses and ceramics has created the need for more efficient means of generating finite element models of these structures. Traditional methods of assembling 400,000 + element models can take weeks of effort, severely limiting the range of possible optimization variables. This paper will introduce model generation software developed under NASA sponsorship for the design of both terrestrial and space based mirrors. The software deals with any current mirror manufacturing technique, single substrates, multiple arrays of substrates, as well as the ability to merge submodels into a single large model. The modeler generates both mirror and suspension system elements, suspensions can be created either for each individual petal or the whole mirror. A typical model generation of 250,000 nodes and 450,000 elements only takes 5-10 minutes, much of that time being variable input time. The program can create input decks for ANSYS, ABAQUS and NASTRAN. An archive/retrieval system permits creation of complete trade studies, varying cell size, depth, and petal size, suspension geometry with the ability to recall a particular set of parameters and make small or large changes with ease. The input decks created by the modeler are text files which can be modified by any editor, all the key shell thickness parameters are accessible and comments in deck identify which groups of elements are associated with these parameters. This again makes optimization easier. With ANSYS decks, the nodes representing support attachments are grouped into components; in ABAQUS these are SETS and in NASTRAN as GRIDPOINT SETS, this make integration of these models into large telescope or satellite models possible.

Arnold, William↗

NASA NTRS2013

Next-Generation Lightweight Mirror Modeling Software

The advances in manufacturing techniques for lightweight mirrors, such as EXELSIS deep core low temperature fusion, Corning's continued improvements in the Frit bonding process and the ability to cast large complex designs, combined with water-jet and conventional diamond machining of glasses and ceramics has created the need for more efficient means of generating finite element models of these structures. Traditional methods of assembling 400,000 + element models can take weeks of effort, severely limiting the range of possible optimization variables. This paper will introduce model generation software developed under NASA sponsorship for the design of both terrestrial and space based mirrors. The software deals with any current mirror manufacturing technique, single substrates, multiple arrays of substrates, as well as the ability to merge submodels into a single large model. The modeler generates both mirror and suspension system elements, suspensions can be created either for each individual petal or the whole mirror. A typical model generation of 250,000 nodes and 450,000 elements only takes 5-10 minutes, much of that time being variable input time. The program can create input decks for ANSYS, ABAQUS and NASTRAN. An archive/retrieval system permits creation of complete trade studies, varying cell size, depth, and petal size, suspension geometry with the ability to recall a particular set of parameters and make small or large changes with ease. The input decks created by the modeler are text files which can be modified by any editor, all the key shell thickness parameters are accessible and comments in deck identify which groups of elements are associated with these parameters. This again makes optimization easier. With ANSYS decks, the nodes representing support attachments are grouped into components; in ABAQUS these are SETS and in NASTRAN as GRIDPOINT SETS, this make integration of these models into large telescope or satellite models possible

Arnold, William R., Sr.↗

NASA NTRS2013

Next Generation Lightweight Mirror Modeling Software

The advances in manufacturing techniques for lightweight mirrors, such as EXELSIS deep core low temperature fusion, Corning's continued improvements in the Frit bonding process and the ability to cast large complex designs, combined with water-jet and conventional diamond machining of glasses and ceramics has created the need for more efficient means of generating finite element models of these structures. Traditional methods of assembling 400,000 + element models can take weeks of effort, severely limiting the range of possible optimization variables. This paper will introduce model generation software developed under NASA sponsorship for the design of both terrestrial and space based mirrors. The software deals with any current mirror manufacturing technique, single substrates, multiple arrays of substrates, as well as the ability to merge submodels into a single large model. The modeler generates both mirror and suspension system elements, suspensions can be created either for each individual petal or the whole mirror. A typical model generation of 250,000 nodes and 450,000 elements only takes 5-10 minutes, much of that time being variable input time. The program can create input decks for ANSYS, ABAQUS and NASTRAN. An archive/retrieval system permits creation of complete trade studies, varying cell size, depth, and petal size, suspension geometry with the ability to recall a particular set of parameters and make small or large changes with ease. The input decks created by the modeler are text files which can be modified by any editor, all the key shell thickness parameters are accessible and comments in deck identify which groups of elements are associated with these parameters. This again makes optimization easier. With ANSYS decks, the nodes representing support attachments are grouped into components; in ABAQUS these are SETS and in NASTRAN as GRIDPOINT SETS, this make integration of these models into large telescope or satellite models easier.

Arnold, William R., Sr.↗

DOE OSTI2023

Large-scale physically accurate modelling of real proton exchange membrane fuel cell with deep learning

Proton exchange membrane fuel cells, consuming hydrogen and oxygen to generate clean electricity and water, suffer acute liquid water challenges. Accurate liquid water modelling is inherently challenging due to the multi-phase, multi-component, reactive dynamics within multi-scale, multi-layered porous media. In addition, currently inadequate imaging and modelling capabilities are limiting simulations to small areas (<1 mm 2 ) or simplified architectures. Herein, an advancement in water modelling is achieved using X-ray micro-computed tomography, deep learned super-resolution, multi-label segmentation, and direct multi-phase simulation. The resulting image is the most resolved domain (16 mm 2 with 700 nm voxel resolution) and the largest direct multi-phase flow simulation of a fuel cell. This generalisable approach unveils multi-scale water clustering and transport mechanisms over large dry and flooded areas in the gas diffusion layer and flow fields, paving the way for next generation proton exchange membrane fuel cells with optimised structures and wettabilities.

25 ENERGY STORAGE↗

DOE OSTI2022

Speaker-targeted Synthetic Speech Detection

Text-to-speech technologies are evolving quickly towards realistic-sounding human-like voices. As this technology improves, so does the opportunity for malpractice in speaker identification (SID) via spoofing, the process of impersonating a voice biometric via synthesis. More data typically equates to a more realistic voice model, which poses an issue for well-known subjects, such as politicians and celebrities, who have vast amounts of multimedia available online. Detection of synthetic speech has relied on signal processing techniques that focus on the generation of new acoustic features and train deep learning models to detect when an audio file has been manipulated through the characterization of unnatural changes or artifacts. However, these techniques do not use any information from the speaker they are evaluating. This paper proposes to incorporate information from the speaker-of-interest (SoI) into the models to avoid specific spoofing attacks for certain vulnerable people. The wealth of data for well-known people can also be used to train a speaker-specific spoofing detector with a higher level of accuracy than a speaker-independent model. The paper proposes a new xResNet-PLDA system and compares it to three different baseline systems: a state-of-the-art speaker identification system, an xResNet system trained to discriminate between bona fide and fake speech, and a speaker identification system in which the PLDA and calibration models were trained with bona fide and fake speech. We evaluated the systems in two different scenarios — a cross-validation scenario and a hold-out scenario — with three different databases. We show how the proposed system outperforms dramatically the baseline systems in each scenario and for each database. Finally, we show how using a small amount of the SoI’s speech to adapt global calibration parameters improves the performance of the system, especially in unseen conditions.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

NASA NTRS1999

The 13th Technology of Deep Space One

On October 24th, 1998, the Deep Space One (DS-1) spacecraft launched aboard a Delta II rocket as the first step towards the bold task of testing and validating 12 new technologies for future missions. This launch also represented yet another thrilling event; namely, the successful test and validation of a 13th heretofore undisclosed technology: model-base-code-generation of the spacecraft's system-level fault-protection (FP) software from behavioral state diagrams and structural models.

model-based-code-generation↗

DOE OSTI2025

Downscaled Earth System Model Data for Resilient Energy System Planning

The second-generation Sup3rCC dataset provides high-resolution meteorological data generated through the downscaling of multiple earth system models (ESMs) from the Coupled Model Intercomparison Project Phase 6 (CMIP6). This downscaling is performed through application of a generative machine learning approach called Super-Resolution for Renewable Resource Data (sup3r). This dataset builds on the first-generation Sup3rCC data by applying improved bias correction methods and adding downscaled precipitation to the output variables. In this presentation, we explore the output characteristics of the dataset and various validation analyses. We also present and discuss plans for the integration of this data into power system planning models using a decision-making under deep uncertainty (DMDU) methodology.

97 MATHEMATICS AND COMPUTING↗

NASA NTRSDate not supplied

Recent Developments to the Porous Microstructure Analysis (PuMA) Software

Introduction The Porous Microstructure Analysis (PuMA) software is an open source framework for image-based simulation, primarily used to determine effective properties based on material microstructure. PuMA was originally developed for the study of NASA thermal protection materials; however, many of the solvers in PuMA have applicability to a broad range of materials science applications. PuMA version 3.2 computes material surface area, pore diameters, effective thermal conductivity, continuum and rarefied tortuosity, and permeability. For anisotropic materials, PuMA can estimate material orientation and compute anisotropic thermal conductivity and elasticity. In this talk, a brief overview of the PuMA software and underlying methods will be presented, as well as some recent and ongoing developments, including the use of immersed boundary methods for image-based simulation and the development of a new weave segmentation tool, called TomoSAM. Cut-Cell method for heat and mass transfer For simulations on complex microstructures, traditional unstructured meshing techniques often prove to be difficult and time-intensive. Voxel-based solvers, which represent the surface as a staircase structure, are relatively simple to implement but can lose accuracy when feature resolution is poor. In this work, we present a novel 3D cut-cell method for solving the variable coefficient Poisson equation on complex microstructures, suitable for the determination of effective thermal conductivity or tortuosity of a material. The method uses a Marching Cubes/Marching Squares surface reconstruction to create cut-cells and determine geometric quantities. A flux-correction method is extended to 3D, with least squares gradient reconstruction, to solve for the boundary fluxes in the cut-cells. Verification cases show the solver achieves globally 2nd order accuracy on complex microstructures. TomoSAM TomoSAM, a module of the PuMA software, has been developed as a plugin for 3D Slicer, a software platform used for 3D image processing and visualization. It utilizes the Segment Anything Model (SAM), a deep learning model capable of identifying objects and generating image masks based on minimal user input. This feature enables efficient segmentation of complex 3D datasets, particularly of woven materials, from tomography or similar imaging methods, reducing the need for manual segmentation.

Tomography↗

NASA NTRSDate not supplied

Recent Developments to the Porous Microstructure Analysis (PuMA) Software

The Porous Microstructure Analysis (PuMA) software is an open source framework for image-based simulation, primarily used to determine effective properties based on material microstructure. PuMA was originally developed for the study of NASA thermal protection materials; however, many of the solvers in PuMA have applicability to a broad range of materials science applications. PuMA version 3.2 computes material surface area, pore diameters, effective thermal conductivity, continuum and rarefied tortuosity, and permeability. For anisotropic materials, PuMA can estimate material orientation and compute anisotropic thermal conductivity and elasticity. In this talk, a brief overview of the PuMA software and underlying methods will be presented, as well as some recent and ongoing developments, including the use of immersed boundary methods for image-based simulation and the development of a new weave segmentation tool, called TomoSAM. Cut-Cell method for heat and mass transfer For simulations on complex microstructures, traditional unstructured meshing techniques often prove to be difficult and time-intensive. Voxel-based solvers, which represent the surface as a staircase structure, are relatively simple to implement but can lose accuracy when feature resolution is poor. In this work, we present a novel 3D cut-cell method for solving the variable coefficient Poisson equation on complex microstructures, suitable for the determination of effective thermal conductivity or tortuosity of a material. The method uses a Marching Cubes/Marching Squares surface reconstruction to create cut-cells and determine geometric quantities. A flux-correction method is extended to 3D, with least squares gradient reconstruction, to solve for the boundary fluxes in the cut-cells. Verification cases show the solver achieves globally 2nd order accuracy on complex microstructures. TomoSAM TomoSAM, a module of the PuMA software, has been developed as a plugin for 3D Slicer, a software platform used for 3D image processing and visualization. It utilizes the Segment Anything Model (SAM), a deep learning model capable of identifying objects and generating image masks based on minimal user input. This feature enables efficient segmentation of complex 3D datasets, particularly of woven materials, from tomography or similar imaging methods, reducing the need for manual segmentation.

Tomography↗

DOE OSTI2024

ESM data downscaling: a comparison of super-resolution deep learning models

Abstract Climate projections at fine spatial resolutions are required to conduct accurate risk assessment for critical infrastructure and design adaptation planning. Generating these projections using advanced Earth system models (ESM) requires significant computational resources. To address this issue, various statistical downscaling techniques have been introduced to generate fine-resolution data from coarse-resolution simulations. In this study, we evaluate and compare five deep learning-based downscaling techniques, namely, super-resolution convolutional neural networks, fast super-resolution convolutional neural network ESM, efficient sub-pixel convolutional neural network, enhanced deep residual network (EDRN), and super-resolution generative adversarial network (SRGAN). These techniques are applied to a dataset generated by the Energy Exascale Earth System Model (E3SM), focusing on key surface variables such as surface temperature, shortwave heat flux, and longwave heat flux. Models are trained and validated using paired fine-resolution (0.25 $$^{\circ }$$ ∘ ) and coarse-resolution (1 $$^{\circ }$$ ∘ ) monthly data obtained from a 9-year simulation. Next, blind testing is performed using monthly data obtained from two different years outside of the training and validation set. To evaluate the efficiency of each technique, different statistical metrics are used, including mean squared error (MSE), peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and learned perceptual image patch similarity (LPIPS). The results show that EDRN outperforms other algorithms in terms of PSNR, SSIM, and MSE, but struggles to capture fine-scale features in the data. In contrast, SRGAN, a generative model that uses perceptual loss, excels in capturing fine details at boundaries and internal structures, resulting in lower LPIPS than other methods.

Pawar, Nikhil M. (ORCID:0000000211613289)↗

DOE OSTI2025

An analysis of physics limited dispatch of nuclear renewable integrated energy systems using deep reinforcement learning and dynamic modeling

Previous approaches to dispatching nuclear integrated energy systems (NIES) have focused on the profitability and flexibility of these systems to operate on energy grids with highly variable pricing. However, due to the complexity involved in modeling and designing these systems, there has been less emphasis on ensuring that these dispatch strategies are physically achievable. It is imperative to develop methods that allow the system to remain within the desired NIES operating conditions and perform this based on realistic limited forecasted information. This research employs next generation artificial intelligence, namely deep reinforcement learning (DRL), and a dynamic system model written in Modelica to find a safe and profitable dispatch strategy for a solar nuclear hybrid design. The DRL agent is shown to find a novel dispatch strategy that manages both power ramping and power levels while respecting operational limits. This DRL-based dispatch is compared to other dispatching strategies including an optimal design solution from mixed integer linear programming (MILP). It is found that incorporating the physics of such a tightly coupled NIES limits the profitability of the MILP-based dispatch strategy. As a result, the MILP solution overestimates the design’s generated revenue. In contrast, DRL significantly reduces the number of breaches of safe operational conditions during energy arbitrage while maintaining profitability. Furthermore, this work paves the way for a more detailed assessment of NIES profitability and could be used to aid operator decisions on future NIES projects.

14 - SOLAR ENERGY↗

DOE OSTI2025

Dataset, Code, and Models for Training Deep Learning Potentials for Low Temperature Plasma-Surface Interactions

This repository contains datasets, training scripts, and finished models, and test simulations used in the development of DeepREBO— a machine-learned interatomic potential trained to emulate the REBO2 empirical potential. The data was generated to study deep potential development for simulations of plasma-surface interactions. It uses an active learning framework, starting from a minimal dataset and iteratively expanding it. Included are those generated datasets, the trained models, and simulations used to evaluate the performance of the training process. This resource supports reproducibility and provides a reference framework for training deep potentials in plasma-surface interaction studies.

active learning↗

DOE OSTI2026

AI-Driven Frameworks for Characterizing Urban Energy Systems

We develop AI-driven frameworks to characterize urban energy systems with the goal of transforming planning by reducing the labor of model generation, scaling scenario exploration, and improving accuracy for localized analysis. The approach integrates top-down and bottom-up data to train different AI models that predict missing information and generate inputs and targeted scenarios for district-scale models. The result is a scalable framework that provides actionable insights for reliable and efficient planning.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

DOE OSTI2024

Cryo2StructData: A Large Labeled Cryo-EM Density Map Dataset for AI-based Modeling of Protein Structures

The advent of single-particle cryo-electron microscopy (cryo-EM) has brought forth a new era of structural biology, enabling the routine determination of large biological molecules and their complexes at atomic resolution. The high-resolution structures of biological macromolecules and their complexes significantly expedite biomedical research and drug discovery. However, automatically and accurately building atomic models from high-resolution cryo-EM density maps is still time-consuming and challenging when template-based models are unavailable. Artificial intelligence (AI) methods such as deep learning trained on limited amount of labeled cryo-EM density maps generate inaccurate atomic models. To address this issue, we created a dataset called Cryo2StructData consisting of 7,600 preprocessed cryo-EM density maps whose voxels are labelled according to their corresponding known atomic structures for training and testing AI methods to build atomic models from cryo-EM density maps. Cryo2StructData is larger than existing, publicly available datasets for training AI methods to build atomic protein structures from cryo-EM density maps. We trained and tested deep learning models on Cryo2StructData to validate its quality showing that it is ready for being used to train and test AI methods for building atomic models.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2022

Data-driven modeling of power generation for a coal power plant under cycling

Increased penetration of renewables for power generation has negatively impacted the dynamics of conventional fossil fuel-based power plants. The power plants operating on the base load are forced to cycle, to adjust to the fluctuating power demands. This results in an inefficient operation of the coal power plants, which leads up to higher operating losses. To overcome such operational challenge associated with cycling and to develop an optimal process control, this work analyzes a set of models for predicting power generation. Moreover, the power generation is intrinsically affected by the state of the power plant components, and therefore our model development also incorporates additional power plant process variables while forecasting the power generation. We present and compare multiple state-of-the-art forecasting data-driven methods for power generation to determine the most adequate and accurate model. We also develop an interpretable attention-based transformer model to explain the importance of process variables during training and forecasting. The trained deep neural network (DNN) LSTM model has good accuracy in predicting gross power generation under various prediction horizons with/without cycling events and outperforms the other models for long-term forecasting. The DNN memory-based models show significant superiority over other state-of-the-art machine learning models for short, medium and long range predictions. The transformer-based model with attention enhances the selection of historical data for multi-horizon forecasting, and also allows to interpret the significance of internal power plant components on the power generation. This newly gained insights can be used by operation engineers to anticipate and monitor the health of power plant equipment during high cycling periods.

01 COAL, LIGNITE, AND PEAT↗