Search NASA⌕ Search

SEARCH · Search NASA

Results for “Neural Network Models”

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 OSTI2022

Sparse Deep Neural Network Inference using different Programming Models

Sparse deep neural networks have gained increasing attention recently in achieving speedups on inference with reduced memory footprints. However, the real-world applications are little shown with specialized optimizations, yet a wide variety of DNN tasks remain dense without exploiting the advantages of sparsity in networks. Recent work presented by MIT/IEEE/Amazon GraphChallenge has demonstrated significant speedups and various techniques. Still, we find that there is limited investigation of the impact of various Python and C\slash C++ based programming models to explore new opportunities in general cases. In this work, we provide performance evaluation through different programming models using CuPy, cuSPARSE, and OpenMP to discuss the advantages and disadvantages of our sparse implementations on single-GPU and multiple GPUs of NVIDIA DGX-A100 40GB/80GB platforms.

machine learning, HPC↗
DOE OSTI2021

Large-scale Gravitational Lens Modeling with Bayesian Neural Networks for Accurate and Precise Inference of the Hubble Constant

We investigate the use of approximate Bayesian neural networks (BNNs) in modeling hundreds of time delay gravitational lenses for Hubble constant (H 0 ) determination. Here, our BNN was trained on synthetic Hubble Space Telescope quality images of strongly lensed active galactic nuclei with lens galaxy light included. The BNN can accurately characterize the posterior probability density functions (PDFs) of model parameters governing the elliptical power-law mass profile in an external shear field. We then propagate the BNN-inferred posterior PDFs into an ensemble H 0 inference, using simulated time delay measurements from a plausible dedicated monitoring campaign. Assuming well-measured time delays and a reasonable set of priors on the environment of the lens, we achieve a median precision of 9.3% per lens in the inferred H 0 . A simple combination of a set of 200 test lenses results in a precision of 0.5 km s –1 Mpc –1 (0.7%), with no detectable bias in this H 0 recovery test. The computation time for the entire pipeline—including the generation of the training set, BNN training and H0 inference—translates to 9 minutes per lens on average for 200 lenses and converges to 6 minutes per lens as the sample size is increased. Being fully automated and efficient, our pipeline is a promising tool for exploring ensemble-level systematics in lens modeling for H 0 inference.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRSDate not supplied

A Natural Language Understanding Approach for Digitizing Aircraft Ground Taxi Instructions

Recent Advancements in spoken language processing technologies have enabled the development of reliable automation tools to assist air traffic control (ATC) operations. These advancements present a unique opportunity to strategically implement digital taxi instructions for aircraft movement on the ground. Digital taxi instructions issue taxiing procedures to pilots as textual or graphical instructions. There are several benefits of digital instructions, including a reduction in radio congestion, elimination of communication errors, and improved aircraft monitoring capabilities. Natural language understanding (NLU) models can extract digital taxi instructions from verbal instructions issued by air traffic controllers. This capability enables the implementation of a digital taxi communication framework with minimal changes to the existing air traffic controller operational environment. We explore a novel application for NLU: automatically generating digital taxi instructions from air traffic controller speech. We describe the development of an annotation scheme to represent (aircraft) ground traffic communications in the US National Airspace System (NAS).Our annotation scheme uses intent classification and slot filling to extract taxi instructions, enabling NLU models to leverage syntactic information. Several neural network models were trained to categorize the controller’s intent and label information relevant to his or her intent.Our research demonstrates that it is feasible to use NLU to automatically generate digitaltaxi instructions, suggesting that it is a powerful tool for the implementation of digital taxi communications.

Hillel Steinmetz↗
DOE OSTI2024

Robust errant beam prognostics with conditional modeling for particle accelerators

Abstract Particle accelerators are complex and comprise thousands of components, with many pieces of equipment running at their peak power. Consequently, they can fault and abort operations for numerous reasons, lowering efficiency and science output. To avoid these faults, we apply anomaly detection techniques to predict unusual behavior and perform preemptive actions to improve the total availability. Supervised machine learning (ML) techniques such as siamese neural network models can outperform the often-used unsupervised or semi-supervised approaches for anomaly detection by leveraging the label information. One of the challenges specific to anomaly detection for particle accelerators is the data’s variability due to accelerator configuration changes within a production run of several months. ML models fail at providing accurate predictions when data changes due to changes in the configuration. To address this challenge, we include the configuration settings into our models and training to improve the results. Beam configurations are used as a conditional input for the model to learn any cross-correlation between the data from different conditions and retain its performance. We employ conditional siamese neural network (CSNN) models and conditional variational auto encoder (CVAE) models to predict errant beam pulses at the spallation neutron source under different system configurations and compare their performance. We demonstrate that CSNNs outperform CVAEs in our application.

43 PARTICLE ACCELERATORS↗
DOE OSTI2024

Chemical composition based machine learning model to predict defect formation in additive manufacturing

With a goal of exploiting additive manufacturing to improve the manufacturing of existing reactor materials, we developed a chemical composition-based machine learning model to predict the printability of any given alloy in laser powder bed fusion (L-PBF) using experimental data from peer-reviewed literature. We defined printability as the ability to avoid defects like cracking, balling, porosity, and lack of fusion, that are caused by thermal stresses (during solidification or liquation), molten pool disintegration into disconnected small beads or lack of heat input respectively. Our models predict the tendency of balling defect formation and porosity percentage for a given composition, under a given set of processing conditions. To predict the likelihood of balling defect, three models: a random forest classifier, a gradient boost regressor and a neural network were trained on a dataset containing both traditional alloys and high entropy alloys. The neural network model showed the highest accuracy of 92.3 % in predicting the balling defect formation. A random forest regressor, gradient boost regressor and neural network were trained and tested on a dataset of various alloys to predict porosity. The random forest regressor showed the best predictions with an R 2 score of 0.97. The models also revealed the relative importance of the input descriptors on defect-formation tendency. Of particular significance was the identification of carbon as an important element in determining the occurrence of balling and percent porosity in alloys like steel, as well as being moderately important to the percentage porosity in other alloys as well as steel. Manganese was also identified as a key descriptor for the percentage of porosity in steel and other alloys. Manganese’s low thermal conductivity and consistent presence in the dataset is the likely cause for its contribution. Carbon’s role is attributable to its relatively high specific heat and high melting temperature. In conclusion, our model serves as a swift, chemistry-based tool to design experiments and find modified compositions better suited for additive manufacturing.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Machine learning-based ethylene and carbon monoxide estimation, real-time optimization, and multivariable feedback control of an experimental electrochemical reactor

Electrochemical reduction of CO 2 gas is a novel CO 2 utilization technique that has the potential to mitigate the global climate crisis caused by anthropogenic CO 2 emissions, and enable the large-scale storage of energy generated from renewable sources in the form of carbon-based chemicals and fuels. However, due to the complexity of the electrochemical reactions, the explicit first-principles models for CO2 reduction are not available yet, and there has been a limited effort to develop process modeling, optimization and control of CO 2 electrochemical reactors. To this end, a rotating cylinder electrode (RCE) reactor has been constructed at UCLA to understand the mass transfer and reaction kinetics effects separately on the productivity. In the RCE reactor, the applied potential strongly influences the reaction energetics and the electrode rotation speed affects the hydrodynamic boundary layer and modifies the film mass transfer coefficient, which involves convective and diffusive transport. Further, the present work aims to develop a multi-input multi-output (MIMO) control scheme for the RCE reactor that integrates techniques from artificial and recurrent neural network modeling, nonlinear optimization, and process controller design. Specifically, production rates of two products from the experimental reactor, ethylene and carbon monoxide, are controlled by manipulating two inputs, applied potential and catalyst rotation speed. Process dynamics and controllability are analyzed, a feedback control strategy is designed and the controllers are tuned accordingly. The experimental electrochemical cell is employed to gather data for process modeling and implement the multivariable control system. Finally, the experimental results are presented which demonstrate excellent closed-loop performance by the control system and regulation of the outputs at three different set-points including an economically-optimal set-point.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

DRDMannTurb: A Python package for scalable, data-driven synthetic turbulence

Synthetic turbulence models (STMs) are used in wind engineering to generate realistic flow fields and are employed as inputs to industrial wind simulations. Examples include prescribing inlet conditions in large eddy simulations that model loads on wind turbines and tall buildings. We are interested in STMs capable of generating fluctuations based on prescribed second-moment statistics since such models can simulate environmental conditions that closely resemble on-site observations. To this end, the widely used Mann model (see Mann, 1994, 1998) is the inspiration for DRDMannTurb. The Mann model is described by three physical parameters: a magnitude parameter influencing the global variance of the wind field and corresponding to the Kolmogorov constant multiplied by the rate of viscous dissipation of the turbulent kinetic energy to the two-thirds, αϵ 2/3 , a turbulence length scale parameter L, and a nondimensional parameter Γ related to the lifetime of the eddies. A number of studies, as well as international standards (e.g., those by the International Electrotechnical Commission (IEC)), include recommended values for these three parameters with the goal of standardizing wind simulations according to observed energy spectra. Yet, having only three parameters, the Mann model faces limitations in accurately representing the diversity of observable spectra. This Python package enables users to extend the Mann model and more accurately fit field measurements through flexible neural network models of the eddy lifetime function. Following Keith et al. (2021), we refer to this class of models as Deep Rapid Distortion (DRD) models. DRDMannTurb also includes a general module implementing an efficient method for synthetic turbulence generation based on a domain decomposition technique. This technique is also described in Keith et al. (2021).

17 WIND ENERGY↗
DOE OSTI2021

Embedded training of neural-network subgrid-scale turbulence models

We report that the weights of a deep neural-network model are optimized in conjunction with the governing flow equations to provide a model for subgrid-scale stresses in a temporally developing plane turbulent jet at Reynolds number Re 0 = 6000 . The objective function for training is first based on the instantaneous filtered velocity fields from a corresponding direct numerical simulation, and the training is by a stochastic gradient descent method, which uses the adjoint Navier-Stokes equations to provide the end-to-end sensitivities of the model weights to the velocity fields. In-sample and out-of-sample testing on multiple dual-jet configurations show that its required mesh density in each coordinate direction for prediction of mean flow, Reynolds stresses, and spectra is half that needed by the dynamic Smagorinsky model for comparable accuracy. The same neural-network model trained directly to match filtered subgrid-scale stresses, without the constraint of being embedded within the flow equations during the training, fails to provide a qualitatively correct prediction. The coupled formulation is generalized to train based only on mean-flow and Reynolds stresses, which are more readily available in experiments. The mean-flow training provides a robust model, which is important, though a somewhat less accurate prediction for the same coarse meshes, as might be anticipated due to the reduced information available for training in this case. The anticipated advantage of the formulation is that the inclusion of resolved physics in the training increases its capacity to extrapolate. This is assessed for the case of passive scalar transport, for which it outperforms established models due to improved mixing predictions.

42 ENGINEERING↗
NASA NTRSDate not supplied

Aeroacoustic Characterization of Optimum Hovering Rotors using Artificial Neural Networks

This work illustrates the use of artificial neural network modeling in the aerodynamic and aeroacoustic characterization of optimum hovering rotors over a broad range of design and operating conditions. Design of Experiments was used to create input feature spaces over eight input factors: the number of rotor blades, rotor radius, rotor rotation rate, design thrust condition, collective pitch, airfoil camber, the location of maximum camber, and the airfoil thickness. A low-fidelity tool chain was then used at the discrete data points defined by the designed input feature spaces to analytically design optimum hovering rotors and simulate aerodynamic and aeroacoustic quantities. This allowed for the generation of data sets over which to train and test the artificial neural network prediction models. Prediction models were trained over the data sets for the actual thrust generated by the rotor, power loading, tonal thickness and loading noise at the fundamental blade passage frequency, and broadband self-noise at seventeen one-third octave bands between 1 kHz and 40 kHz. These prediction models were validated by testing over data previously unseen by the models to quantify their capability for generalization to new data within the design feature space. The models were then used to study the effect each input feature had on the aeroacoustics and aerodynamics of optimum hovering rotors and physical insights were concluded to further explain the effect of each input. This characterization study showed that tonal noise and power loading were most sensitive to the number of rotor blades and the rotor rotation rate and that broadband noise was most sensitive to collective pitch and the design thrust condition.

Christopher S Thurman↗
DOE OSTI2025

Inverse model based error detection in beamline optics

Optics tuning in transfer lines and LINACs can be challenging due to the fact that multiple combinations of machine settings can lead to the same diagnostic output. Moreover, the lack of a periodic solution can limit the ability to infer optics in the same way as rings from BPM signals. Model based approaches are often used to assist with the optics tuning in combination with optimization or parameter estimation. Here we have developed a novel approach using machine learning inverse models trained on a known configuration to detect variations in quadrupole settings without explicitly including them in the model. This paper shows a comparison of neural network models and linear models on both a simulation based study and experimental studies conducted at the AGS to RHIC transfer line at Brookhaven National Lab.

43 PARTICLE ACCELERATORS↗
DOE OSTI2023

Machine learning based simultaneous control of air handling unit discharge air and condenser water temperatures set-point for minimized cooling energy in an office building

In this study, an artificial intelligence based real-time prediction and control model to optimize condenser water temperature and discharge air temperature (DAT) set-points in water-cooled air handling unit (AHU) system has been developed. EnergyPlus-MATLAB co-simulation has been conducted to analyze the developed model's effectiveness. Here, to develop artificial neural networks (ANN) model, embedded neural network objects in MATLAB was utilized. The developed model could decide an optimal temperature set-points based on outdoor air wet-bulb temperature to reflect the Korean climate context. As a result, the developed ANN prediction model showed the predictive performance of Cv(RMSE) of approximately 21%. Compared to the conventional fixed temperature algorithm, which fixes AHU DAT at 14°C and condenser water temperature at 32°C, the ANN based optimized control showed a 22% total cooling energy reduction. These results show that significant energy savings can be achieved by simultaneously controlling condenser water temperature and AHU DAT set-points considering Korean climatic characteristics using AI technologies such as ANN models.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

A Data-Driven Framework for Predicting the Sorting and Screening Performance of an Integrated Biomass Feedstock Preprocessing System

The characteristics of mechanically sorted and screened lignocellulosic biomass, such as the mass contents of corn stover anatomical fractions (leaves, husks, stalks, cobs, etc.), can be used to calculate the intermediate feedstock quality attributes “yield” and “purity” that indicate the conversion efficiency of biocrude. No prior study has investigated the correlations from the characteristics of raw biomass and preprocessing unit operation parameters to those intermediate feedstock quality attributes. This work presents a data-driven framework for assessing and predicting the intermediate feedstock quality attributes in an integrated biomass feedstock preprocessing system. Our study used corn stover as a typical type of herbaceous biomass because of its abundance in the U.S. It began with data acquisition of moisture content, particle size distribution, and anatomical fractions of the materials after each unit operation in the system. The objective of this preprocessing system is to minimize husks and leaves and maximizing cobs and stalks by mechanically separating the materials into three streams via disc screen and air separator. Prototype neural network models were then developed to evaluate the feasibility of predicting process outcomes based on measurable parameters. It is found that incorporating physical constraints into these prediction models significantly enhances the accuracy of the predicted yield and purity against the ground truth data. The experimental data and model predictions indicate that decreasing throughput increases purity, while higher throughput results in lower purity. Finally, an optimization problem was introduced to search optimal combinations of feed material properties and preprocessing unit operation parameters, as the intermediate feedstock quality attributes – yield and purity, appeared to be competing factors. The study also suggests the continual need to improve the data-driven framework’s predictability by incorporating more accurate physical models to describe the dynamics in the preprocessing units such as the air separator.

09 - BIOMASS FUELS↗
NASA NTRS2019

Estuarine Dissolved Organic Carbon Flux From Space: With Application to Chesapeake and Delaware Bays

This study uses a neural network model trained with in situ data, combined with satellite data and hydrodynamic model products, to compute the daily estuarine export of dissolved organic carbon (DOC) at the mouths of Chesapeake Bay (CB) and Delaware Bay (DB) from 2007 to 2011. Both bays show large flux variability with highest fluxes in spring and lowest in fall as well as interannual flux variability (0.18 and 0.27 Tg C/year in 2008 and 2010 for CB; 0.04 and 0.09 Tg C/year in 2008 and 2011 for DB). Based on previous estimates of total organic carbon (TOCexp) exported by all Mid-Atlantic Bight estuaries (1.2 Tg C/year), the DOC export (CB + DB) of 0.3 Tg C/year estimated here corresponds to 25% of the TOCexp. Spatial and temporal covariations of velocity and DOC concentration provide contributions to the flux, with larger spatial influence. Differences in the discharge of fresh water into the bays (74 billion cu. m/year for CB and 21 billion cu. m/year for DB) and their geomorphologies are major drivers of the differences in DOC fluxes for these two systems. Terrestrial DOC inputs are similar to the export of DOC at the bay mouths at annual and longer time scales but diverge significantly at shorter time scales (days to months). Future efforts will expand to the Mid-Atlantic Bight and Gulf of Maine, and its major rivers and estuaries, in combination with coupled terrestrial-estuarine-ocean biogeochemical models that include effects of climate change, such as warming and CO2 increase.

Estuarine DOC export↗
DOE OSTI2024

Acoustic-based monitoring and machine learning of component status for microreactor applications

This report provides a description and assessment of recent efforts to couple acoustic-based experimental measurements and characterization with machine learning models in order to enhance structural health monitoring capabilities for nuclear microreactors. With resilient embedded sensors in development by others supported by programs funded by the US Department of Energy’s Office of Nuclear Energy, the work described herein builds upon ongoing efforts to improve non-destructive testing technology that relates measured acoustic signatures to component stresses and/or structural defects, using a combination of new experimental measurements and machine learning architectures. The experimental procedure remained similar to that developed for the previous year’s demonstration of damage detection by the authors, with the same damaged sample tested under similar applied stress conditions. Notably, a new mounting fixture was designed and implemented to improve measurement consistency and a more sophisticated laser Doppler vibrometer was employed to make high-fidelity vibration measurements. Two nominally identical sets of training data were collected for each experimental setup to better understand the repeatability of the experiment and to better test the generality of trained neural network models. Additionally, we obtained new high-quality 3D mode shapes of the damaged test article at various stress and excitation levels, providing greater insights into the physical response of the sample during testing. Previously, we demonstrated that a machine learning model based on a convolutional neural network can predict structural details of an artificially introduced interface (intact, rough cut, smooth cut), and the applied torque level. In this study, we have transitioned to graph-based neural network architectures to better develop and test a flexible framework that is more suitable to being transferred away from controlled benchtop experiments and into more applied settings where less-structured data inputs may be expected. In general, performance testing of a graph neural network on frequency-domain representations of the data indicates strong and consistent identification of test conditions for datasets recorded on damaged components. With goals of predicting damage location and other changing experimental conditions using limited datasets, predictive models using a graph neural network architecture correctly predicted the applied torque level with an accuracy of 85% using only a single measurement point and predicted within one torque level in 95% of test windows. Predictions of damage location had limited success due to the symmetry and minimal number of the damage scenarios presented during model training. Results were ambiguous as to whether the model could detect the location of the artificial damage, or if it was instead learning the location of a given measurement point on the part and subsequently detecting which points were closest to the location of the damage. This finding will be factored into upcoming planned work on damaged graphite components, where new experimental tests with a larger number and variety of damage scenarios are expected to provide improved validation of recent developments in monitoring methodology.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

“Understanding Robustness Lottery”: A Geometric Visual Comparative Analysis of Neural Network Pruning Approaches

Deep learning approaches have provided state-of-the-art performance in many applications by relying on large and overparameterized neural networks. However, such networks are very brittle and are difficult to deploy on resource-limited platforms. Model pruning, i.e., reducing the size of the network, is a widely adopted strategy that can lead to a more robust and compact model. Many heuristics exist for model pruning, but our understanding of the pruning process remains limited due to the black-box nature of a neural network model. Empirical studies show that some heuristics improve performance whereas others can make models more brittle. Here, this work aims to shed light on how different pruning methods alter the network’s internal feature representation and the corresponding impact on model performance. To facilitate a comprehensive comparison and characterization of the high-dimensional model feature space, we introduce a visual geometric analysis of feature representations. We evaluated a set of critical geometric concepts decomposed from the commonly adopted classification loss and used them to design a visualization system to compare and highlight the impact of pruning on model performance and feature representation. The proposed tool provides an environment for an in-depth comparison of pruning methods and a comprehensive understanding of how the model responds to common data corruption. By leveraging the proposed visualization, machine learning researchers can reveal the similarities between pruning methods and redundancy in robustness evaluation benchmarks, obtain geometric insights about the differences between pruned models that achieve superior robustness performance, and identify samples that are robust or fragile to model pruning and common data corruption.

Li, Zhimin [Univ. of Utah, Salt Lake City, UT (Uni↗
NASA NTRS1995

Neurodynamic oscillators

Oscillation of electrical activity has been found in many nervous systems, from invertebrates to vertebrates including man. There exists experimental evidence of very simple circuits with the capability of oscillation. Neurons with intrinsic oscillation have been found and also neural circuits where oscillation is a property of the network. These two types of oscillations coexist in many instances. It is nowadays hypothesized that behind synchronization and oscillation there is a system of coupled oscillators responsible for activities that range from locomotion and feature binding in vision to control of sleep and circadian rhythms. The huge knowledge that has been acquired on oscillators from the times of Lord Rayleigh has made the simulation of neural oscillators a very active endeavor. This has been enhanced with more recent physiological findings about small neural circuits by means of intracellular and extracellular recordings as well as imaging methods. The future of this interdisciplinary field looks very promising; some researchers are going into quantum mechanics with the idea of trying to provide a quantum description of the brain. In this work we describe some simulations using neuron models by means of which we form simple neural networks that have the capability of oscillation. We analyze the oscillatory activity with root locus method, cross-correlation histograms, and phase planes. In the more complicated neural network models there is the possibility of chaotic oscillatory activity and we study that by means of Lyapunov exponents. The companion paper shows an example of that kind.

Espinosa, Ismael↗
DOE OSTI2025

A Parallel Alternative for Energy-Efficient Neural Network Training and Inferencing

Energy efficiency of training and inferencing with large neural network models is a critical challenge facing the future of sustainable large-scale machine learning workloads. This paper introduces an alternative strategy, called phantom parallelism, to minimize the net energy consumption of traditional tensor (model) parallelism, the most energy-inefficient component of large neural network training. The approach is presented in the context of feed-forward network architectures as a preliminary, but comprehensive, proof-of-principle study of the proposed methodology. We derive new forward and backward propagation operators for phantom parallelism, implement them as custom autograd operations within an end-to-end phantom parallel training pipeline and compare its parallel performance and energy-efficiency against those of conventional tensor parallel training pipelines. Formal analyses that predict lower bandwidth and FLOP counts are presented with supporting empirical results on up to 256 GPUs that corroborate these gains. Experiments are shown to deliver ∼50% reduction in the energy consumed to train FFNs using the proposed phantom parallel approach when compared with conventional tensor parallel methods. Additionally, the proposed approach is shown to train smaller phantom models to the same model loss on smaller GPU counts as larger tensor parallel models on larger GPU counts offering the possibility for even greater energy savings.

Seal, Sudip [ORNL] (ORCID:0000000332330656)↗