Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Training Analysis”

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 1,387 records · Page 77
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Do Machine Learning Approaches Offer Skill Improvement for Short-Term Forecasting of Wind Gust Occurrence and Magnitude?

Abstract Wind gusts, and in particular intense gusts, are societally relevant but extremely challenging to forecast. This study systematically assesses the skill enhancement that can be achieved using artificial neural networks (ANNs) for forecasting of wind gust occurrence and magnitude. Geophysical predictors from the ERA5 reanalysis are used in conjunction with an autoregressive term in regression and ANN models with different predictors, and varying model complexity. Models are derived and assessed for the warm (April–September) and cold (October–March) seasons for three high passenger volume airports in the United States. Model uncertainty is assessed by deriving models for 1000 different randomly selected training (70%) and testing (30%) subsets. Gust prediction fidelity in independent test samples is critically dependent on inclusion of an autoregressive term. Gust occurrence probabilities derived using five-layer ANNs exhibit consistently higher fidelity than those from regression models and shallower ANNs. Inclusion of the autoregressive term and increasing the number of hidden layers in ANNs from 1 to 5 also improve the model performance for gust magnitudes (lower RMSE, increased correlation, and model standard deviations that more closely approximate observed values). Deeper ANNs (e.g., 20 hidden layers) exhibit higher skill in forecasting strong (17–25.7 m s −1 ) and damaging (≥25.7 m s −1 ) wind gusts. However, such deep networks exhibit evidence of overfitting and still substantially underestimate (by 50%) the frequency of strong and damaging wind gusts at the three airports considered herein. Significance Statement Improved short-term forecasting of wind gusts will enhance aviation safety and logistics and may offer other societal benefits. Here we present a rigorous investigation of the relative skill of models of wind gust occurrence and magnitude that employ different statistical methods. It is shown that artificial neural networks (ANNs) offer considerable skill enhancement over regression methods, particularly for strong and damaging wind gusts. For wind gust magnitudes in particular, application of deeper learning networks (e.g., five or more hidden layers) offers tangible improvements in forecast accuracy. However, deeper networks are vulnerable to overfitting and exhibit substantial variability with the specific training and testing data subset used. Also, even deep ANNs reproduce only half of strong and damaging wind gusts. These results indicate the need for future work to elucidate the dynamical mechanisms of intense wind gusts and advance solutions to their prediction.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Fluid-Kinetic Coupling: Advanced Discretizations for Simulations on Emerging Heterogeneous Architectures (LDRD FY20-0643)

Plasma physics simulations are vital for a host of Sandia mission concerns, for fundamental science, and for clean energy in the form of fusion power. Sandia's most mature plasma physics simulation capabilities come in the form of particle-in-cell (PIC) models and magnetohydrodynamics (MHD) models. MHD models for a plasma work well in denser plasma regimes when there is enough material that the plasma approximates a fluid. PIC models, on the other hand, work well in lower-density regimes, in which there is not too much to simulate; error in PIC scales as the square root of the number of particles, making high-accuracy simulations expensive. Real-world applications, however, almost always involve a transition region between the high-density regimes where MHD is appropriate, and the low-density regimes for PIC. In such a transition region, a direct discretization of Vlasov is appropriate. Such discretizations come with their own computational costs, however; the phase-space mesh for Vlasov can involve up to six dimensions (seven if time is included), and to apply appropriate homogeneous boundary conditions in velocity space requires meshing a substantial padding region to ensure that the distribution remains sufficiently close to zero at the velocity boundaries. Moreover, for collisional plasmas, the right-hand side of the Vlasov equation is a collision operator, which is non-local in velocity space, and which may dominate the cost of the Vlasov solver. The present LDRD project endeavors to develop modern, foundational tools for the development of continuum-kinetic Vlasov solvers, using the discontinuous Petrov-Galerkin (DPG) methodology, for discretization of Vlasov, and machine-learning (ML) models to enable efficient evaluation of collision operators. DPG affords several key advantages. First, it has a built-in, robust error indicator, allowing us to adapt the mesh in a very natural way, enabling a coarse velocity-space mesh near the homogeneous boundaries, and a fine mesh where the solution has fine features. Second, it is an inherently high-order, high-intensity method, requiring extra local computations to determine so-called optimal test functions, which makes it particularly suited to modern hardware in which floating-point throughput is increasing at a faster rate than memory bandwidth. Finally, DPG is a residual-minimizing method, which enables high-accuracy computation: in typical cases, the method delivers something very close to the $L^2$ projection of the exact solution. Meanwhile, the ML-based collision model we adopt affords a cost structure that scales as the square root of a standard direct evaluation. Moreover, we design our model to conserve mass, momentum, and energy by construction, and our approach to training is highly flexible, in that it can incorporate not only synthetic data from direct-simulation Monte Carlo (DSMC) codes, but also experimental data. We have developed two DPG formulations for Vlasov-Poisson: a time-marching, backward-Euler discretization and a space-time discretization. We have conducted a number of numerical experiments to verify the approach in a 1D1V setting. In this report, we detail these formulations and experiments. We also summarize some new theoretical results developed as part of this project (published as papers previously): some new analysis of DPG for the convection-reaction problem (of which the Vlasov equation is an instance), a new exponential integrator for DPG, and some numerical exploration of various DPG-based time-marching approaches to the heat equation. As part of this work, we have contributed extensively to the Camellia open-source library; we also describe the new capabilities and their usage. We have also developed a well-documented methodology for single-species collision operators, which we applied to argon and demonstrated with numerical experiments. We summarize those results here, as well as describing at a high level a design extending the methodology to multi-species operators. We have released a new open-source library, MLC, under a BSD license; we include a summary of its capabilities as well.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

Application of Principal Component Analysis to Electrochemical Reprocessing PM and NMAC

In this report, data from an electrorefiner (ER) for nuclear fuel reprocessing is evaluated for process monitoring (PM) conclusions. This data comes from tests performed at the Idaho National Laboratory in 2022. Multivariate approaches utilizing methods of Principal Component Analysis (PCA) is applied. This is based off established work in process monitoring for fault detection in industrial facilities. This report will discuss the background, methods, and results of the application and some of the conclusions and applications that can be drawn from them. PCA is applied to two different electrorefiner (ER) operations that occurred at Idaho National Laboratory between August and October 2022. The first operation occurred with little incident while the second had several noted faults in the equipment in operational logs. The data from the first run was used to train the data for “normal” operations and applied to both sets of data to determine when operations were in an “off-normal” condition and identify where the fault occurs through PCA. PCA was able to identify off-normal events and identify the cause for off-normal operations. These identified off-normal events matched with the events and their causes in the operational logs. However, small amounts of variance in the data led to false detection of “off-normal” events. Thus, careful selection of training data and a-posteriori conclusions based off operator assessments will both be required for application of PCA to PM applications. This work demonstrated that multivariate approaches and latent variables are applicable to pyroprocessing PM applications and can be further expanded in future work as quality variables such as salt concentration from sensors and sampling become available.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2022

Latent-Space Dynamics for Prediction and Fault Detection in Geothermal Power Plant Operations

This paper presents a latent-space dynamic neural network (LSDNN) model for the multi-step-ahead prediction and fault detection of a geothermal power plant’s operation. The model was trained to learn the dynamics of the power generation process from multivariate time-series data and the effects of exogenous variables, such as control adjustment and ambient temperature. In the LSDNN model, an encoder–decoder architecture was designed to capture cross-correlation among different measured variables. In addition, a latent space dynamic structure was proposed to propagate the dynamics in the latent space to enable prediction. The prediction power of the LSDNN was utilized for monitoring a geothermal power plant and detecting abnormal events. The model was integrated with principal component analysis (PCA)-based process monitoring techniques to develop a fault-detection procedure. The performance of the proposed LSDNN model and fault detection approach was demonstrated using field data collected from a geothermal power plant.

15 GEOTHERMAL ENERGY↗
DOE OSTI2022

Constructing A New CHF Look-Up Table Based on the Domain Knowledge Informed Machine Learning Methodology

Accurate prediction of CHF under various fluid flow conditions continues to be required for design, operation and safety analysis of light water reactor rod bundles. Due to the lack of in-depth physical understanding as well as limited high-resolution data in the micro-scale flow and heat transfer, the existing models feature a sub-optimal uncertainty band. In this study, driven by the prior domain knowledge information obtained, an improved CHF look-up table is developed through unified machine learning algorithms for the vertical flow conditions within tube and annulus geometry. The Groeneveld 2006 look-up table is used as the domain knowledge to train machine learning process against tube and annulus CHF data for both DNB and DO type. The new look-up table shows improved accuracy for conditions relevant to PWRs and BWRs. In addition, its domain knowledge informed nature ensures that a rationale prediction can be made, thus accounting for previous valuable information in the machine learning model training process.

Jin, Yue↗
DOE OSTI2026

Taming nuclear mass models with Gaussian processes

We propose a new set of nuclear mass predictions based on multiple theoretical mass models. By employing Gaussian process regression with the Matérn kernel, we achieved root-mean-square (rms) deviations below 100 keV for the training dataset. The best-performing mass models achieved rms deviations below 150 keV for the new precise mass data from AME2020, whereas the ensemble average showed robust performance across the nuclear chart. Our approach uniquely combines: (1) systematic refinement of eight mass models through their residuals, (2) physics-informed features, including magic numbers, nucleon parity numbers, neutron excess, and nuclear collectivity, and (3) theory-to-theory validation demonstrating robust extrapolation capability. We find that the Matérn kernel provides superior uncertainty quantification compared to the RBF kernel, with a length-scale analysis revealing enhanced inter-nuclei correlations. We provide complete mass predictions for all unknown nuclides in AME2020, offering valuable constraints for nuclear structure studies and astrophysical modeling when used with proper uncertainty propagation.

Gaussian processes↗
DOE OSTI2022

Continuous-time probabilistic models for longitudinal electronic health records

Analysis of longitudinal Electronic Health Record (EHR) data is an important goal for precision medicine. Difficulty in applying Machine Learning (ML) methods, either predictive or unsupervised, stems in part from the heterogeneity and irregular sampling of EHR data. Here, we present an unsupervised probabilistic model that captures nonlinear relationships between variables over continuous-time. This method works with arbitrary sampling patterns and captures the joint probability distribution between variable measurements and the time intervals between them. Inference algorithms are derived that can be used to evaluate the likelihood of future using under a trained model. As an example, we consider data from the United States Veterans Health Administration (VHA) in the areas of diabetes and depression. Likelihood ratio maps are produced showing the likelihood of risk for moderate-severe vs minimal depression as measured by the Patient Health Questionnaire-9 (PHQ-9).

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Explaining machine-learning models for gamma-ray detection and identification

As more complex predictive models are used for gamma-ray spectral analysis, methods are needed to probe and understand their predictions and behavior. Recent work has begun to bring the latest techniques from the field of Explainable Artificial Intelligence (XAI) into the applications of gamma-ray spectroscopy, including the introduction of gradient-based methods like saliency mapping and Gradient-weighted Class Activation Mapping (Grad-CAM), and black box methods like Local Interpretable Model-agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP). In addition, new sources of synthetic radiological data are becoming available, and these new data sets present opportunities to train models using more data than ever before. In this work, we use a neural network model trained on synthetic NaI(Tl) urban search data to compare some of these explanation methods and identify modifications that need to be applied to adapt the methods to gamma-ray spectral data. We find that the black box methods LIME and SHAP are especially accurate in their results, and recommend SHAP since it requires little hyperparameter tuning. We also propose and demonstrate a technique for generating counterfactual explanations using orthogonal projections of LIME and SHAP explanations.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2021

Machine-Learned Manifold-Based Models for Large Eddy Simulation of Turbulent Combustion

Reduced-order manifold combustion models are commonly used to lower the cost of reacting Large Eddy Simulations (LES) and may be derived either from physical principles as in flamelet models or from data-driven methods like Principal Component Analysis (PCA). In either case, neural networks are increasingly used as part of these models to provide a nonlinear mapping between a small set of pre-defined variables that parameterize the manifold and outputs of interest, such as reaction rates. In this work, we propose a new manifold-based modeling approach that combines the definition of the manifold-parameterizing variables (linear combinations of species), the nonlinear mapping to the outputs, and closure of filtered quantities for LES into the structure of a single neural network. This allows the process used to train the neural network to simultaneously optimize both the functional form of the model and the identities of the inputs to the model. The new approach can flexibly incorporate thermochemical data from any combustion system; if trained on data from 1D flames it can be interpreted as an optimized flamelet model, but it can also be used to learn models from data from more complex configurations. This work presents a priori evaluations of the new approach in both contexts. Evaluation using data from 1D premixed flames demonstrates the physical interpretability of the manifold variables generated by the new approach. Evaluation using data from direct numerical simulations of turbulent flames shows improved predictions relative to either flamelet or PCA-based models in a more complex configuration.

47 OTHER INSTRUMENTATION↗
DOE OSTI2022

Recurrent neural networks for short-term and long-term prediction of geothermal reservoirs

Accurate prediction of geothermal reservoir responses to alternative energy production scenarios is critical for optimizing the development of the underlying resources. While the conventional physics-based models offer a comprehensive prediction tool, data-driven models provide an efficient alternative to build fit-for-purpose predictive models by extracting and using the statistical patterns in the collected data to make predictions. The recurrent neural network (RNN) is a data-driven model that is commonly applied to predict time series sequences. This paper presents a variant of RNN that also utilizes the efficiency of convolutional neural networks (CNN) for the prediction of energy production from geothermal reservoirs. Specifically, a CNN–RNN architecture is developed that takes historical well controls as input (features) and their corresponding production response data as output (labels) to learn an input-output mapping that can predict the future well production responses/performance for any given future well control inputs. The model is paired with a labeling scheme to handle real field disturbances that create data gaps. In addition to the model structure, we introduce a thorough workflow for applying the model, which includes data pre-processing, feature selection, as well as different training strategies for short-term and long-term prediction. Finally, the performance and accuracy of the model are evaluated by applying it to multiple datasets, including a field reservoir model.

15 GEOTHERMAL ENERGY↗
DOE OSTI2022

Estimating cluster masses from SDSS multiband images with transfer learning

ABSTRACT The total masses of galaxy clusters characterize many aspects of astrophysics and the underlying cosmology. It is crucial to obtain reliable and accurate mass estimates for numerous galaxy clusters over a wide range of redshifts and mass scales. We present a transfer-learning approach to estimate cluster masses using the ugriz-band images in the SDSS Data Release 12. The target masses are derived from X-ray or SZ measurements that are only available for a small subset of the clusters. We designed a semisupervised deep learning model consisting of two convolutional neural networks. In the first network, a feature extractor is trained to classify the SDSS photometric bands. The second network takes the previously trained features as inputs to estimate their total masses. The training and testing processes in this work depend purely on real observational data. Our algorithm reaches a mean absolute error (MAE) of 0.232 dex on average and 0.214 dex for the best fold. The performance is comparable to that given by redMaPPer, 0.192 dex. We have further applied a joint integrated gradient and class activation mapping method to interpret such a two-step neural network. The performance of our algorithm is likely to improve as the size of training data set increases. This proof-of-concept experiment demonstrates the potential of deep learning in maximizing the scientific return of the current and future large cluster surveys.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2020

Technoeconomic Analysis of the Electrochemically Mediated Amine Regeneration CO 2 Capture Process

The electrochemically mediated amine regeneration (EMAR) process presents an alternative route to the conventional thermal amine regeneration for carbon capture from a flue gas source. In this study, we conducted an economic analysis on the EMAR system for postcombustion CO2 capture from a 550 MWe power plant capturing 3.1 MtCO2 annually and from a mini steel mill with annual capture close to 170 ktCO2. We followed the recommendation of the National Energy Technology Laboratory (NETL) 2010 report to estimate the cost of CO2 avoided ([$/tCO2]). Detailed cost modeling of the electrochemical separation stage was conducted. This is integrated with the entire process flowsheet (e.g., including absorber, compressor, pumps, and other auxiliary equipment). Here, we identified the membrane cost as the dominant capital cost for the electrochemical separation train. At a membrane unit price of less than 10/m2, the CO2 capture cost can be reduced to below 50/tCO2 with optimized process conditions (e.g., desorption pressure and utilization of waste heat). Improvements in process design, cell construction, and solvent formulation may lead to additional reductions in the CO2 capture cost.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Versatile recognition of graphene layers from optical images under controlled illumination through green channel correlation method

In this study, a simple yet versatile method is proposed for identifying the number of exfoliated graphene layers transferred on an oxide substrate from optical images, utilizing a limited number of input images for training, paired with a more traditional number of a few thousand well-published Github images for testing and predicting. Two thresholding approaches, namely the standard deviation-based approach and the linear regression-based approach, were employed in this study. The method specifically leverages the red, green, and blue color channels of image pixels and creates a correlation between the green channel of the background and the green channel of the various layers of graphene. This method proves to be a feasible alternative to deep learning-based graphene recognition and traditional microscopic analysis. The proposed methodology performs well under conditions where the effect of surrounding light on the graphene-on-oxide sample is minimum and allows rapid identification of the various graphene layers. Here, the study additionally addresses the functionality of the proposed methodology with nonhomogeneous lighting conditions, showcasing successful prediction of graphene layers from images that are lower in quality compared to typically published in literature. In all, the proposed methodology opens up the possibility for the non-destructive identification of graphene layers from optical images by utilizing a new and versatile method that is quick, inexpensive, and works well with fewer images that are not necessarily of high quality.

36 MATERIALS SCIENCE↗
DOE OSTI2024

The Pixel Anomaly Detection Tool : a user-friendly GUI for classifying detector frames using machine-learning approaches

Data collection at X-ray free electron lasers has particular experimental challenges, such as continuous sample delivery or the use of novel ultrafast high-dynamic-range gain-switching X-ray detectors. This can result in a multitude of data artefacts, which can be detrimental to accurately determining structure-factor amplitudes for serial crystallography or single-particle imaging experiments. Here, a new data-classification tool is reported that offers a variety of machine-learning algorithms to sort data trained either on manual data sorting by the user or by profile fitting the intensity distribution on the detector based on the experiment. This is integrated into an easy-to-use graphical user interface, specifically designed to support the detectors, file formats and software available at most X-ray free electron laser facilities. The highly modular design makes the tool easily expandable to comply with other X-ray sources and detectors, and the supervised learning approach enables even the novice user to sort data containing unwanted artefacts or perform routine data-analysis tasks such as hit finding during an experiment, without needing to write code.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

Pretraining Billion-Scale Geospatial Foundational Models on Frontier

As AI workloads increase in scope, generalization capability becomes challenging for small task-specific models and their demand for large amounts of labeled training samples increases. On the contrary, Foundation Models (FMs) are trained with internet-scale unlabeled data via self-supervised learning and have been shown to adapt to various tasks with minimal fine-tuning. Although large FMs have demonstrated significant impact in natural language processing and computer vision, efforts toward FMs for geospatial applications have been restricted to smaller size models, as pretraining larger models requires very large computing resources equipped with state-of-the-art hardware accelerators. Current satellite constellations collect 100+TBs of data a day, resulting in images that are billions of pixels and multimodal in nature. Such geospatial data poses unique challenges opening up new opportunities to develop FMs. We investigate billion scale FMs and HPC training profiles for geospatial applications by pretraining on publicly available data. We studied from end-to-end the performance and impact in the solution by scaling the model size. Our larger 3B parameter size model achieves up to 30% improvement in top1 scene classification accuracy when comparing a 100M parameter model. Moreover, we detail performance experiments on the Frontier supercomputer, America's first exascale system, where we study different model and data parallel approaches using PyTorch's Fully Sharded Data Parallel library. Specifically, we study variants of the Vision Transformer architecture (ViT), conducting performance analysis for ViT models with size up to 15B parameters. By discussing throughput and performance bottlenecks under different parallelism configurations, we offer insights on how to leverage such leadership-class HPC resources when developing large models for geospatial imagery applications.

Tsaris, Aristeidis (aris)↗
DOE OSTI2025

Robust Dark Energy Constraints with the Dark Energy Spectroscopic Survey (Final Technical Report)

This project developed and applied advanced theoretical, computational, and data-analysis methodologies to extract robust and precise cosmological constraints from the Dark Energy Spectroscopic Instrument (DESI). The work focused on maximizing the scientific return of DESI through optimized survey strategy, novel higher-order clustering statistics, improved modeling of small-scale structure, and rigorous mitigation of observational systematics. Over the award period, the project made substantial contributions to DESI science planning, produced new methods for bispectrum and three-point correlation function analyses, advanced constraints on primordial non-Gaussianity, and delivered widely used software tools. The project also played a major role in training graduate students and a postdoctoral researcher who contributed directly to DESI key projects. The results have significantly enhanced the cosmological reach of DESI and provide a strong foundation for future surveys such as DESI-II and Stage-V experiments.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

Early Alarm: Robust Event Analysis for Power Systems using 1-D Fully Convolutional Network

This work presents a novel deep learning model for early, accurate, and robust detection, recognition, and temporal localization of multi-type events in large-scale power systems. The proposed method develops a unified 1-D fully convolutional network (FCN) model that takes time series of raw frequency signals measured from a power system as input, extracts distinguishing features, and predicts at every temporal point in the time series if an event is happening and what the type of the event is. Compared to existing methods, the proposed model eliminates the necessity for hand-crafted feature extraction or complicated data pre-processing, can flexibly handle input signals of arbitrary length, and precisely infer the event occurrence time. Most importantly, the model is intentionally trained with incomplete patterns, such that it is more robust to partial features of an event which is common in real-world online recognition, resulting in early alarm for power system failures. Extensive experimental results demonstrate that the proposed method achieves superior performance to the state-of-the-art, and also shows strong robustness to noise and system oscillations.

Li, Chengcheng↗
DOE OSTI2026

Superstructure Optimization of Waste Plastic Pyrolysis, Integrating Thermal, Catalytic, and Plasma Technologies with Machine Learning

Global plastic waste generation exceeds 430 million tonnes per year, yet fewer than 9% are recycled in the United States. Pyrolysis offers a chemical recycling route at scale, but existing techno-economic and life cycle assessments fix product yields to single pure polymers, producing economic and environmental outputs that break down when the feed composition changes. Here, we present a superstructure optimization framework that addresses this by embedding a composition-aware random forest yield predictor, trained on 566 pyrolysis experiments, within a full-scale process simulation. Product distributions update automatically as feed allocation shifts across four reactor chemistries: conventional thermal, catalytic (HZSM-5), thermal oxo-degradation, and nonequilibrium CO2 plasma. The optimal superstructure achieves minimum selling prices of −0.56 to −0.76/kg feed and global warming potentials of −0.276 to −0.322 kg CO2-eq/kg feed across four commodity price scenarios, confirming profitable, carbon-negative operation without tipping fees. Carbon abatement costs of $\$$0.46 to $\$$1.25/kg CO2-eq are competitive with direct air capture. Sensitivity analysis shows that the catalytic-plasma split fraction is the single largest driver of both economic and climate performance, while hydrocracking allocation in the wax upgrading stage is emission-neutral across the full variable range. Mixed plastic waste streams, evaluated as composition-variable feedstocks rather than pure resins, are profitable and carbon-negative across realistic market conditions. These results give a quantitative basis for reactor selection, circular economy investment, and policy design targeting chemical recycling on a large scale.

Life cycle assessment↗