Search NASA⌕ Search

SEARCH · Search NASA

Results for “Neural Networks Research”

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 217 records · Page 12
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Position Papers for the 2024 ASCR Workshop on Neuromorphic Computing for Science

Engineering novel neuromorphic computing systems with functionalities, capabilities, and energy efficiency similar to biological brains is one of the most exciting and challenging scientific endeavors of our time. This workshop aims to identify key research needs, challenges, and next steps necessary to develop biologically realistic neuromorphic circuits primitives that capture the functionality of neural systems found in nature. Moreover, simulating neuromorphic computing primitives integrated into networks will be key to under standing their behavior at scale, particularly for those computing architectures where full-scale commercial fabrication is not yet readily accessible. Appropriate neuroscience datasets and metrics will have to be established to vet proposed neuromorphic circuits.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Quantifying Atomically Dispersed Catalysts Using Deep Learning Assisted Microscopy

The catalytic performance of atomically dispersed catalysts (ADCs) is greatly influenced by their atomic configurations, such as atom–atom distances, clustering of atoms into dimers and trimers, and their distributions. Scanning transmission electron microscopy (STEM) is a powerful technique for imaging ADCs at the atomic scale; however, most STEM analyses of ADCs thus far have relied on human labeling, making it difficult to analyze large data sets. Here, we introduce a convolutional neural network (CNN)-based algorithm capable of quantifying the spatial arrangement of different adatom configurations. The algorithm was tested on different ADCs with varying support crystallinity and homogeneity. Results show that our algorithm can accurately identify atom positions and effectively analyze large data sets. Here, this work provides a robust method to overcome a major bottleneck in STEM analysis for ADC catalyst research. We highlight the potential of this method to serve as an on-the-fly analysis tool for catalysts in future in situ microscopy experiments.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Numerical Investigation of a Central Fuel Property Hypothesis under Boosted Spark-ignition Conditions

In the present work, a central fuel property hypothesis (CFPH), which states that fuel properties are sufficient to provide an indication of a fuel's performance irrespective of its chemical composition, was numerically investigated In particular, the objective of the study was to determine whether Research Octane Number (RON) and Motor Octane Number (MON), as fuel properties, are sufficient to describe a fuel's knock-limited performance under boosted spark-ignition (SI) conditions within the framework of CFPH. To this end, four TPRF-bioblendstock surrogates having different compositions but matched RON (= 98) and MON (= 90), were first generated using a non-linear regression model based on artificial neural network (ANN). Three unconventional bioblendstocks were included in the analysis: Di-isobutylene (DIB), Isobutanol and Anisole. Skeletal reaction mechanisms were generated for the TPRF-DIB, TPRF-isobutanol and TPRF-anisole blends from a detailed kinetic mechanism. Thereafter, numerical simulations were performed for the fuel surrogates using the skeletal mechanisms and a virtual cooperative fuel research (CFR) engine model, under a representative boosted operating condition. In the computational fluid dynamics (CFD) model, the G-equation approach was employed to track the turbulent flame front and the well-stirred reactor model combined with multi-zone binning strategy was used to capture auto-ignition in the end-gas. In addition, laminar flame speed was tabulated for each blend as a function of pressure, temperature and equivalence ratio a priori, and the lookup tables were used to prescribe laminar flame speed as an input to the G-equation model. Parametric spark timing sweeps were performed for each fuel blend to determine the corresponding knock-limited spark advance (KLSA) and 50% burn point (CA50) at the respective KLSA timing. It was observed that despite same RON, MON and engine operating conditions, the TPRF-Anisole blend exhibited markedly different knock-limited performance from the other three blends. This deviation from the octane index (DI) expectation was shown to be caused by differences in laminar flame speed (LFS). However; it was found that relatively large fuel-specific differences in LFS (> 20%) would have to be present to cause any appreciable deviation from the 01 framework. Otherwise, RON and MON would still be robust enough to predict a fuel's knock-limited performance.

Pal, Pinaki↗
DOE OSTI2025

Generalizable, fast, and accurate DeepQSPR with fastprop

Abstract Quantitative Structure–Property Relationship studies (QSPR), often referred to interchangeably as QSAR, seek to establish a mapping between molecular structure and an arbitrary target property. Historically this was done on a target-by-target basis with new descriptors being devised to specifically map to a given target. Today software packages exist that calculate thousands of these descriptors, enabling general modeling typically with classical and machine learning methods. Also present today are learned representation methods in which deep learning models generate a target-specific representation during training. The former requires less training data and offers improved speed and interpretability while the latter offers excellent generality, while the intersection of the two remains under-explored. This paper introduces , a software package and general Deep-QSPR framework that combines a cogent set of molecular descriptors with deep learning to achieve state-of-the-art performance on datasets ranging from tens to tens of thousands of molecules. provides both a user-friendly Command Line Interface and highly interoperable set of Python modules for the training and deployment of feedforward neural networks for property prediction. This approach yields improvements in speed and interpretability over existing methods while statistically equaling or exceeding their performance across most of the tested benchmarks. is designed with Research Software Engineering best practices and is free and open source, hosted at github.com/jacksonburns/fastprop.

Burns, Jackson W. (ORCID:0000000206579426)↗
DOE OSTI2020

Numerical Investigation of a Central Fuel Property Hypothesis Under Boosted Spark-Ignition Conditions

In the present work, a central fuel property hypothesis (CFPH), which states that fuel properties are sufficient to provide an indication of a fuel’s performance irrespective of its chemical composition, was numerically investigated. In particular, the objective of the study was to determine whether Research Octane Number (RON) and Motor Octane Number (MON), as fuel properties, are sufficient to describe a fuel’s knock-limited performance under boosted spark-ignition (SI) conditions within the framework of CFPH. To this end, four TPRF-bioblendstock surrogates having different compositions but matched RON (=98) and MON (=90), were first generated using a non-linear regression model based on artificial neural network (ANN). Additionally, three unconventional bioblendstocks were included in the analysis: di-isobutylene (DIB), isobutanol, and Anisole. Skeletal reaction mechanisms were generated for the TPRF-DIB, TPRF-isobutanol, and TPRF-anisole blends from a detailed kinetic mechanism. Thereafter, numerical simulations were performed for the fuel surrogates using the skeletal mechanisms and a virtual cooperative fuel research (CFR) engine model, under a representative boosted operating condition. In the computational fluid dynamics (CFD) model, the G-equation approach was employed to track the turbulent flame front and the well-stirred reactor model combined with the multi-zone binning strategy was used to capture auto-ignition in the end-gas. In addition, laminar flame speed (LFS) was tabulated for each blend as a function of pressure, temperature, and equivalence ratio a priori, and the lookup tables were used to prescribe laminar flame speed as an input to the G-equation model. Parametric spark timing sweeps were performed for each fuel blend to determine the corresponding knock-limited spark advance (KLSA) and 50% burn point (CA50) at the respective KLSA timing. It was observed that despite same RON, MON, and engine operating conditions, the TPRF-anisole blend exhibited markedly different knock-limited performance from the other three blends. This deviation from the octane index (OI) expectation was shown to be caused by differences in laminar flame speed. However, it was found that relatively large fuel-specific differences in LFS (>20%) would have to be present to cause any appreciable deviation from the OI framework. Otherwise, RON and MON would still be robust enough to predict a fuel’s knock-limited performance.

42 ENGINEERING↗
DOE OSTI2024

Deep Learning-Based Failure Prognostic Model for PV Inverter Using Field Measurements

Here, this study presents a novel approach for the precise monitoring and prognosis of photovoltaic (PV) inverter status, which is crucial for the proactive maintenance of PV systems. It addresses the gaps in traditional model-based methods, which tend to neglect the overall reliability of inverters, and the limitations of data-driven approaches that largely depend on simulated data. This research presents a robust solution applicable to real-world scenarios. The proposed data-driven model for PV inverter failure prognosis employs actual inverter measurements, integrating various operational and weather-related factors based on domain knowledge. This approach effectively represents inverter stressors and operational status. Utilizing an Enhanced Siamese Convolutional Neural Network (ESCNN), the model merges operational data with domain knowledge features, redefining the prognosis challenge as a classification task. Furthermore, the paper discusses an ESCNN-based real-time inverter failure monitoring method developed on the well-trained model. The proposed models are rigorously trained and tested with real inverter data and a novel filtering method is included to address accidental failures in practical scenarios. The results validate the model's efficacy, and the directions for future research are also outlined.

42 ENGINEERING↗
DOE OSTI2025

Precision calibration of calorimeter signals in the ATLAS experiment using an uncertainty-aware neural network

The ATLAS experiment at the Large Hadron Collider explores the use of modern neural networks for a multi-dimensional calibration of its calorimeter signal defined by clusters of topologically connected cells (topo-clusters). The Bayesian neural network (BNN) approach not only yields a continuous and smooth calibration function that improves performance relative to the standard calibration but also provides uncertainties on the calibrated energies for each topo-cluster. The results obtained by using a trained BNN are compared to the standard local hadronic calibration and to a calibration provided by training a deep neural network. The uncertainties predicted by the BNN are interpreted in the context of a fractional contribution to the systematic uncertainties of the trained calibration. They are also compared to uncertainty predictions obtained from an alternative estimator employing repulsive ensembles.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Inversion of Time-Lapse Seismic Reservoir Monitoring Data Using CycleGAN: A Deep Learning-Based Approach for Estimating Dynamic Reservoir Property Changes

Carbon capture and storage is being pursued globally as a geoengineering measure for reducing the emission of anthropogenic CO 2 the atmosphere. Comprehensive monitoring, verification, and accounting programs must be established for demonstrating the safe storage of injected CO 2 . One of the most commonly deployed monitoring techniques is time-lapse seismic reservoir monitoring (also known as 4-D seismic), which involves comparing 3-D seismic survey data taken at the same study site but over different times. Analyses of 4-D seismic data volumes can help improve the quality of storage reservoir characterization, track the movement of injected CO 2 plume, and identify potential CO 2 spillover/leakage from the storage reservoirblue. However, the derivation of high-resolution CO 2 saturation maps from 4-D seismic data is a highly nonlinear and ill-posed inverse problem, often requiring significant computational effort. In this research, we apply a physics-based deep learning method to facilitate the solution of both the forward and inverse problems in seismic inversion while honoring physical constraints. A cycle generative adversarial neural network (CycleGAN) model is trained to learn the bidirectional functional mappings between the reservoir dynamic property changes and seismic attribute changes, such that both forward and inverse solutions can be obtained efficiently from the trained model. We show that our CycleGAN-based approach not only improves the reliability of 4-D seismic inversion but also expedites the quantitative interpretation. Our deep learning-based workflow is generic and can be readily used for reservoir characterization and reservoir model updates involving the use of 4-D seismic data.

58 GEOSCIENCES↗
DOE OSTI2024

Improved Subseasonal Forecasting of Extreme Polar Vortices Using Machine Learning

Our research was focused on forecasting the position and shape of the winter stratospheric polar vortex at a subseasonal timescale of 15 days in advance. To achieve this, we employed both statistical and neural network machine learning techniques. The analysis was performed on 42 winter seasons of reanalysis data provided by NASA giving us a total of 6,342 days of data. The state of the polar vortex for determined by using geometric moments to calculate the centroid latitude and the aspect ratio of an ellipse fit onto the vortex. Timeseries for thirty additional precursors were calculated to help improve the predictive capabilities of the algorithm. Feature importance of these precursors was performed using random forest to measure the predictive importance and the ideal number of precursors. Then, using the precursors identified as important, various statistical methods were tested for predictive accuracy with random forest and nearest neighbor performing the best. An echo state network, a type of recurrent neural network that features sparsely connected hidden layer and a reduced number of trainable parameters that allows for rapid training and testing, was also implemented for the forecasting problem. Hyperparameter tuning was performed for each methods using a subset of the training data. The algorithms were trained and tuned on the first 41 years of data, then tested for accuracy on the final year. In general, the centroid latitude of the polar vortex proved easier to predict than the aspect ratio across all algorithms. Random forest outperformed other statistical forecasting algorithms overall but struggled to predict extreme values. Forecasting from echo state network suggested a strong predictive capability past 15 days, but further work is required to fully realize the potential of recurrent neural network approaches.

54 ENVIRONMENTAL SCIENCES↗
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 OSTI2021

Learning the structure of wind: A data-driven nonlocal turbulence model for the atmospheric boundary layer

In this work, we develop a novel data-driven approach to modeling the atmospheric boundary layer. This approach leads to a nonlocal, anisotropic synthetic turbulence model which we refer to as the deep rapid distortion (DRD) model. Our approach relies on an operator regression problem that characterizes the best fitting candidate in a general family of nonlocal covariance kernels parameterized in part by a neural network. This family of covariance kernels is expressed in Fourier space and is obtained from approximate solutions to the Navier–Stokes equations at very high Reynolds numbers. Each member of the family incorporates important physical properties such as mass conservation and a realistic energy cascade. The DRD model can be calibrated with noisy data from field experiments. After calibration, the model can be used to generate synthetic turbulent velocity fields. To this end, we provide a new numerical method based on domain decomposition which delivers scalable, memory-efficient turbulence generation with the DRD model as well as others. We demonstrate the robustness of our approach with both filtered and noisy data coming from the 1968 Air Force Cambridge Research Laboratory Kansas experiments. Using these data, we witness exceptional accuracy with the DRD model, especially when compared to the International Electrotechnical Commission standard.

17 WIND ENERGY↗
DOE OSTI2021

Evaluation of the PhaseNet Model Applied to the IMS Seismic Network

Producing a complete and accurate set of signal detections is essential for automatically building and characterizing seismic events of interest for nuclear explosion monitoring. Signal detection algorithms have been an area of research for decades, but still produce large quantities of false detections and misidentify real signals that must be detected to produce a complete global catalog of events of interest. Deep learning methods have shown promising capabilities in effectively characterizing seismic signals for complex tasks such as identifying phase arrival times. We use the PhaseNet model, a UNet-based Neural Network, trained on local distance data from northern California to predict seismic arrivals on data from the International Monitoring System (IMS) global network. We use an analyst-curated bulletin generated from this data set to compare the performance of PhaseNet to that of the Short-Term Average/Long-Term Average (STA/LTA) algorithm. We find that PhaseNet has the potential of outperforming traditional processing methods and recommend the training of a new model with the IMS data to achieve optimal performance.

58 GEOSCIENCES↗
DOE OSTI2024

Uncertainty-Informed Volume Visualization using Implicit Neural Representation

The increasing adoption of Deep Neural Networks (DNNs) has led to their application in many challenging scientific visualization tasks. While advanced DNNs offer impressive generalization capabilities, understanding factors such as model prediction quality, robustness, and uncertainty is crucial. These insights can enable domain scientists to make informed decisions about their data. However, DNNs inherently lack ability to estimate prediction uncertainty, necessitating new research to construct robust uncertainty-aware visualization techniques tailored for various visualization tasks. In this work, we propose uncertainty-aware implicit neural representations to model scalar field data sets effectively and comprehensively study the efficacy and benefits of estimated uncertainty information for volume visualization tasks. We evaluate the effectiveness of two principled deep uncertainty estimation techniques: (1) Deep Ensemble and (2) Monte Carlo Dropout (MC-Dropout). These techniques enable uncertainty-informed volume visualization in scalar field data sets. Our extensive exploration across multiple data sets demonstrates that uncertainty-aware models produce informative volume visualization results. Moreover, integrating prediction uncertainty enhances the trustworthiness of our DNN model, making it suitable for robustly analyzing and visualizing real-world scientific volumetric data sets.

Saklani, Shanu↗
DOE OSTI2025

Monitoring Fracture Hydromechanical Evolution in the Lab and Field Using Unsupervised Metric Learning

Fractures evolve in time through thermal‐hydraulic‐mechanical‐chemical (THMC) processes that alter their long‐range hydraulic transport properties and modify subsurface behavior and activities. The location of subsurface fractures makes it necessary to use remote sensing techniques such as passive or active seismic monitoring for fracture characterization. In this paper, we develop a machine learning approach to monitor the evolution of fracture properties using passive seismic sources in a laboratory setting and using active seismic monitoring from the Sanford Underground Research Facility in Lead, South Dakota, at a depth of 1.25 km in amphibolite rock during stimulation of natural fractures as well as during induced fracturing. The unsupervised metric learning technique applies tandem neural networks (twin (Siamese) or triplet) with contrastive loss and adaptive margins to track slowly varying systems for which class or similarity labels are not available. The approach adopts locality‐sensitive hashing to divide time‐ordered contiguous data into an arbitrary number of pseudo‐classes. Contrastive‐loss training with many hash bins generates an evolving latent‐space trajectory. This approach enables unsupervised metric learning for seismic data stacks under the condition of contiguous state sampling and slowly varying fracture properties. The displacement discontinuity theory provides a mechanistic foundation for the fracture‐dependent trajectories that are related to relaxation of fractures with time‐dependent specific stiffness responding to changes in stress or fluid saturation.

02 PETROLEUM↗
DOE OSTI2023

A Hardware and Software Co-design Framework for Energy Efficient Neuromorphic Systems

Neuromorphic systems can be realized by a variety of algorithms and architectures. A common understanding is that spiking neuromorphic designs, which encode information into spatio-temporal spiking events, are both a biologically-accurate and efficient way of processing information. However, representing the information through timing relationships induces sophisticated circuit designs in traditional CMOS-based implementations. In recent years, high-capacity resistive memory (RRAM, aka, memristor) has demonstrated great potential in mimicking synaptic behaviors. Several RRAM-based spiking neuromorphic designs exist, most of which focus on rate coding schemes. These designs simplify circuit implementations of neuron models and explore challenges such as unsatisfactory speed, resolution, and performance. As an alternative, we will explore temporal coding spiking neuromorphic systems that encode information as the relative timing of neuron activations (spikes), which have been proven to be more adaptive and energy-efficient. Developing a neuromorphic system for spiking neural network (SNN) inference and online training, however, faces some major technical challenges: (1) It lacks circuit implementation support for temporal-coding SNN to achieve satisfying power efficiency and accuracy; (2) Although existing research works have investigated memristive synapse and neuron designs for spike-timing-dependent plasticity, the non-ideal conditions in implementation, such as device variations and signal degradation, degrade online learning accuracy of large scale systems; and (3) Non-optimized, inter-layer data traffic in SNNs, leads to unnecessary data communication costs. In this project, we plan to address these challenges by a hardware and software co-design framework that incorporates solutions at the circuit, architecture, and algorithm levels. At the circuit-level, we will elaborate on the in-situ SNN processing element designs for supporting both inference and online training modes. Variation-aware schemes will be studied to improve reliability. At the architecture level, we propose a pipelined, asynchronous architecture to retain the timing resolution of spikes. At the algorithm level, we will investigate an innovative SNN training algorithm for enabling activation sparsification and reducing unnecessary data communication costs. This neuromorphic system will provide an effective solution to real-life energy-constrained applications and significantly contribute to the exploration of next-generation high-performance computing systems under the DOE context.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Replication Data for: Neural network assisted analysis of bimetallic nanocatalysts using X-ray absorption near edge structure spectroscopy

The data underlying this published work have been made publicly available in this repository as part of the IMASC Data Management Plan. This work was supported as part of the Integrated Mesoscale Architectures for Sustainable Catalysis (IMASC), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award # DE-SC0012573.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Cutting the Deployment Costs of Physics-Based MPC in Buildings by Simulation-Based Imitation Learning

It has been shown that model predictive control (MPC) is a promising solution for energy-efficient building operations. However, the deployment of MPC in a large portion of the building stock has not been possible partially because of high installation costs. Every building is unique and requires a tailored MPC solution. The best performing solutions are often based on physics-based modeling, which is, however, computationally expensive and requires dedicated software. A promising direction that tackles this problem is to train a neural network-based optimal control policy to imitate the behavior of physics-based MPC from the simulation data generated offline. The neural networks give control actions that closely approximate those produced by physics-based MPC, but with a fraction of the computational and memory requirements and without the need for licensed software. The main advantage of the proposed approach stems from simple evaluation at execution time, leading to low computational footprints and easy deployment on embedded HW platforms. In the case study, we present the energy savings potential of physics-based MPC applied to an office building in Belgium. We demonstrate how neural network approximators can be used to cut the implementation and maintenance costs of MPC deployment without compromising performance. We also critically assess the presented approach by pointing out the remaining challenges and open research questions.

Drgona, Jan↗
DOE OSTI2023

Few-Shot Learning Enables Population-Scale Analysis of Leaf Traits in Populus trichocarpa

Plant phenotyping is typically a time-consuming and expensive endeavor, requiring large groups of researchers to meticulously measure biologically relevant plant traits, and is the main bottleneck in understanding plant adaptation and the genetic architecture underlying complex traits at population scale. In this work, we address these challenges by leveraging few-shot learning with convolutional neural networks to segment the leaf body and visible venation of 2,906 Populus trichocarpa leaf images obtained in the field. In contrast to previous methods, our approach (a) does not require experimental or image preprocessing, (b) uses the raw RGB images at full resolution, and (c) requires very few samples for training (e.g., just 8 images for vein segmentation). Traits relating to leaf morphology and vein topology are extracted from the resulting segmentations using traditional open-source image-processing tools, validated using real-world physical measurements, and used to conduct a genome-wide association study to identify genes controlling the traits. In this way, the current work is designed to provide the plant phenotyping community with (a) methods for fast and accurate image-based feature extraction that require minimal training data and (b) a new population-scale dataset, including 68 different leaf phenotypes, for domain scientists and machine learning researchers. All of the few-shot learning code, data, and results are made publicly available.

59 BASIC BIOLOGICAL SCIENCES↗