Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “constrained learning”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 289 records · Page 16
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

A cross-dimensional analysis of data-driven short-term load forecasting methods with large-scale smart meter data

Electricity load forecasting is essential to utility operation and power grid stability. A wide spectrum of data-driven methods, ranging from linear regression models to more recent deep learning models have been adopted to forecast electric load over the years. However, there still lacks a holistic evaluation of the applicability of conventional statistical and machine learning based algorithms with respect to different temporal and spatial scopes, computational requirements, and sensitivity of model-tuning. Enabled by a large-scale electricity load profile dataset of over 40,000 residential customers in a utility region, we conducted a cross-dimensional analysis of data-driven load forecasting methods. Three regression-based and seven deep learning algorithms with different model configurations were evaluated in terms of their overall and peak load prediction accuracy, and training burdens, across spatial aggregation levels ranging from the transformer, feeder, substation, to neighborhood. We found, first, the load forecasting accuracy is constrained by a predictability boundary, influenced by the forecasting horizon and spatial aggregation level. Specifically, RandomForest, XGBoost, TFT, TSMixer, and TiDE models achieved less than 10 % prediction error for up to 96-h ahead forecasting for district, substation, and feeder levels, while other models struggle at long-horizon predictions; Second, for winter and summer peak load dates, most models were able to predict the peak demand timing within ± 1 h, but the prediction percentage error varied by models, with TFT and TiDE models being the top performers; Third, models with similar prediction accuracy can differ in training burden by an order of magnitude. Therefore, choosing model configurations that balance prediction performance and computational resource is an important practical consideration for large-scale deployment of the machine learning based load forecasting. The outcome of this study can guide researchers and practitioners to choose the proper load forecasting algorithms based on their problem scope, required accuracy, and available resources. The predictability boundary can serve as a benchmark for electricity load forecasting problems with new algorithms and datasets.

Li, Han↗
DOE OSTI2022

Designing Monte Carlo Simulation and an Optimal Machine Learning to Optimize and Model Space Missions

This paper investigates applying artificial intelligence (AI) algorithms to attitude control system of satellites to optimally tune the controller using high performance computing. This methodology is applied to the Virtual Telescope for X-ray Observation mission, which is a precise formation of two separate spacecraft observing multiple objects in the space in the X-ray domain. The mission is divided into phases based on the instrumentation and the mission goal. To reach an stable precise formation robust to stochastic slew and slew rate (i.e., Euler angles and angular velocities) in a minimal constrained time T , consumed energy of the attitude control system, denoted as E, and root-mean-square state error of attitude control system, denoted as e, are minimized. Monte-Carlo simulation is used for the sensitivity analysis of optimization and designing a controller. Deep neural networks (DNN), Gaussian processes (GP), and support vector regression (SVR) learn this optimization as a surrogate model, while their hyperparameters are optimized in a novel approach. THETA supercomputer at Argonne Leadership Computing Facility (ALCF) is used for optimizing the hyperparameters of DNN. The surrogate model meets the requirements of the mission, and it shows a better performance over the optimization and Monte-Carlo. The optimal DNN can satisfy the mission requirements e and T while reducing E for 90% compared to the other given methods.

42 ENGINEERING↗
DOE OSTI2021

Development of Integrated Mechanical Pods

This presentation highlights early wins, updated progress, and upcoming developments on ‘national-scale shared development platform’ for rapid prototyping, testing and validation of various integrated Mechanical Pod solutions and form factors. Such pod solutions consist of a set of all-electric heat pump mechanical equipment that have integrated functionalities through built-in controls, with heating, cooling, hot water, ventilation (including energy recovery), electrical management, and battery storage within a single package. The presentation draws inspiration from the success of bathroom pods in the US modular construction industry, UK’s efforts with unitizing mechanical systems as ‘utility cupboards’, and VEIC’s early wins in design-build of all-electric Mechanical Pod solutions in Vermont. The presentation includes researchers and partners involved with NREL in Design for Manufacturing and Assembly (DfMA), Virtual Design and Construction (VDC), and digital twin based process optimization modeling of integrated Mechanical Pod solutions. The presentation aims to highlight early wins from such a platform and how various physical and virtual tools are currently being employed as part of NREL’s ongoing multi-year project funded by US DOE. Streamlined procurement, coordination, installation, and O&M of Mechanical Pods such that the majority of work is delegated to the off-site modular factory implies monetary savings. Such a seemingly basic shift in location of the construction process leads to great reduction in complexity, first cost, lead time, and waste, and greater opportunities for innovative compartmentalization and integration of mechanical systems appropriately sized for each apartment or hotel guest room. However, past studies on unitized combination systems show that high installation costs, maintenance issues, challenges with system integration, limitations in existing electrical infrastructure, and lack of architecturally appealing solutions are key barriers. NREL and partners aim to address key barriers through DfMA approach, rapid prototyping and testing, and digital twin process optimization modeling. The presentation is also a call for interested entities to partner with NREL as part of the national-scale development platform, help drive both product and process innovation, and encourage open source sharing of learnings. Learning objectives include (1) learn about the vision of national-scale shared development platform for process-product innovation on integrated mechanical pod solutions and how to get involved, (2) gain an understanding of the components of an all-electric, high performance home, design characteristics and equipment included in an all-electric mechanical pod, integration of mechanical systems within a modular factories’ assembly line, and the system’s commissioning, operation and maintenance. The pre-planning and coordination with the factory and sub-contractors are also highlighted, (3) gain an understanding of using process modeling tools to quantify resource-constrained performance of operations (such as integration of energy efficiency strategies) to manufacture modules of varying design, (4) gain insights on virtual design, rapid prototyping, and emulated testing of various form factors across different climatic conditions. The need for such preliminary testing with open source sharing of learnings will also be highlighted.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2025

Early Career: Constraining the Electromagnetic Shower Energy Scale at LArTPC Neutrino Detectors Near and Far (Final Scientific/Technical Report)

This is the final scientific/technical report associated with the award "Early Career: Constraining the Electromagnetic Shower Energy Scale at LArTPC Neutrino Detectors Near and Far"; the primary research topics associated with the award are neutrino oscillation physics, reconstruction of neutrino interaction events using machine learning techniques, and detector calibration.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

Future marine biofuels in the port of Seattle region

Marine transportation, a vital global sector, emits 3% of global annual greenhouse gas emissions, which are predicted to increase in the future. Marine biofuels derived from biomass or waste sources like wood residue, waste oil and municipal solid waste can be used for decarbonization. However, limited studies have explored if sufficient marine biofuels could be produced and supplied to major regional ports given feedstock, supply chain and technological constraints. We fill this gap by evaluating the feasibility of supplying marine biofuels to the Port of Seattle. The Regional Bio-Economy Model (RBEM) and the Freight and Fuel Transportation Optimization Tool (FTOT) are used to build scenarios for simulating marine biofuel production in the Port region. We harmonized technoeconomic assumptions for RBEM and FTOT, input FTOT feedstock utilization and routing outputs into RBEM, and modelled conversion, feedstock, and policy scenario variations in RBEM. In RBEM, overall biofuel production was constrained primarily by the biofuel cost, and then by feedstock availability. Providing policy incentives and reducing permitting time frames alleviated these constraints and spurred the buildout of a robust industry through industrial learning dynamics in the initial years. With these measures in place, the RBEM results show that 100% of fuel demand at the Port can be supplied by biofuels with policy incentives and suitable technoeconomic conditions, but the addition of transportation cost considerations using FTOT led to 27.8% of demand being able to be met by biofuels at reasonable fuel delivery cost.

09 BIOMASS FUELS↗
DOE OSTI2026

Creating a Training Dataset for Semantic Segmentation of Canal Networks for Irrigation Modernization

Canal infrastructure has provided critical irrigation water to the western United States for over a century. To continue providing vital water resources to the semi-arid West, irrigation systems must undergo maintenance and modernization. Many canal companies are resource-constrained, and because funding opportunities often require detailed knowledge of existing infrastructure, they can struggle to secure financial capital. We address this problem by creating training data for a semantic segmentation deep learning model to map canal networks throughout the western United States. To create a diverse and robust training dataset, we labelled 1-m NAIP imagery with the locations of no canals, wet canals, and dry/vegetated canals. Since creating these datasets is time consuming, we first developed a preprocessing methodology to identify canals within our four study areas. We used NAIP imagery and provided canal centerline data to buffer, standardize, and cluster the imagery, automating the labeling process as much as possible. However, this still required manual cleaning and manual classification of canal type. Challenges arose when canals were interrupted (e.g., road culverts or piped sections) or when nearby features shared similar characteristics (e.g., irrigated fields, trees, and shadows). Combining automated preprocessing with manual refinement produced four detailed canal masks to be used in the semantic segmentation model developed by Richard Tapia.

13 - HYDRO ENERGY↗
DOE OSTI2025

Physics-Informed Neural Networks for PDE-Constrained Optimization and Control

The goal of optimal control is to determine a sequence of inputs for maximizing or minimizing a given performance criterion subject to the dynamics and constraints of the system under observation. This work introduces Control Physics-Informed Neural Networks (PINNs), which simultaneously learn both the system states and the optimal control signal in a single-stage framework that leverages the system’s underlying physical laws. While prior approaches often follow a two-stage process-modeling, the system first and then devising its control—the presented novel framework embeds the necessary optimality conditions directly into the network architecture and loss function. We demonstrate the effectiveness of the novel methodology by solving various open-loop optimal control problems governed by analytical, one-dimensional, and two-dimensional partial differential equations (PDEs).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Uncovering dark matter density profiles in dwarf galaxies with graph neural networks

Dwarf galaxies are small, dark-matter-dominated galaxies, some of which are embedded within the Milky Way. Their lack of baryonic matter (e.g., stars and gas) makes them perfect test beds for probing the properties of dark matter—understanding the spatial dark matter distribution in these systems can be used to constrain microphysical dark matter interactions that influence the formation and evolution of structures in our Universe. Here, we introduce a new method that leverages simulation-based inference and graph-based machine learning in order to infer the dark matter density profiles of dwarf galaxies from observable kinematics of stars gravitationally bound to these systems. Our approach aims to address some of the limitations of established methods based on dynamical Jeans modeling. We show that this novel method can place stronger constraints on dark matter profiles and, consequently, has the potential to weigh in on some of the ongoing puzzles associated with the small-scale structure of dark matter halos, such as the core-cusp discrepancy.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Deciphering Degradation: Machine Learning on Real-World Performance Data

The solar investment community is in need of up-to-date and data-derived system degradation rates to use in financial models that calculate the risk and expectation of energy revenue. To date, financial models choose degradation rates from past studies that are limited and may not be representative of solar projects in development today. As the solar industry expands in all sectors (residential, commercial, and utility), there is a growing amount of energy generation data available to analyze. With the help of the open source degradation analysis code (RdTools) it is now possible to derive degradation on an ongoing basis and continuously provide up-to-date durability statistics to the investment community. Ongoing and accurate statistics of fielded photovoltaic systems allows financial stakeholders to constrain energy revenue projections, lower financial risk, and increase the bankability of solar. This presentation will provide a mid-term progress update on a 2 year SETO-funded degradation project that involves applying RdTools analysis on kWh Analytics’ industry database. With results from the RdTools analysis in hand, machine learning methods are employed to quantify how much variance of system performance degradation can be explained by predictor variables such as environmental factors, equipment bill of materials characteristics, and system design information. The goal of this presentation is to promote the usage of RdTools on an ongoing to data owners and to solicit feedback from researchers and stakeholders on how to maximize impact of the SETO-funded kWh Analytics degradation project.

14 SOLAR ENERGY↗
DOE OSTI2021

Learning viscoelasticity models from indirect data using deep neural networks

In this study, we propose a novel approach to model viscoelasticity materials, where rate-dependent and non-linear constitutive relationships are approximated with deep neural networks. We assume that inputs and outputs of the neural networks are not directly observable, and therefore common training techniques with input–output pairs for the neural networks are inapplicable. To that end, we develop a novel computational approach to both calibrate parametric and learn neural-network-based constitutive relations of viscoelasticity materials from indirect displacement data in the context of multiple-physics systems. We show that limited displacement data holds sufficient information to quantify the viscoelasticity behavior. We formulate the inverse computation – modeling viscoelasticity properties from observed displacement data – as a PDE-constrained optimization problem and minimize the error functional using a gradient-based optimization method. The gradients are computed by a combination of automatic differentiation and implicit function differentiation rules. The effectiveness of our method is demonstrated through numerous benchmark problems in geomechanics and porous media transport.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

A Combined Machine Learning and High-Energy X-ray Diffraction Approach to Understanding Liquid and Amorphous Metal Oxides

Determining the structure-property relations of liquid and amorphous metal oxides is challenging, due to their variable short-range order and polyhedral connectivity. To predict chemically realistic structures, we have developed a Machine Learned, Gaussian Approximation Potential (GAP) for HfO 2 , with a focus on enhanced sampling of the training database and accurate density functional theory calculations. By using training datasets for the GAP model at the level of Density Functional Theory-Strongly Constrained and Appropriately Normed (DFT-SCAN) level of theory, our results show that the topology of both the low viscosity liquid and the amorphous form are dominated by edge-shared chains and small corner-shared rings of polyhedra. This topology is shown to be consistent with the structure of other liquid and amorphous transition metal oxides of variable ion size, such as TiO 2 and ZrO 2 . Current limitations of the ML-GAP modeling method for obtaining glass structures and future perspectives are also discussed.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Safe and Robust Binary Classification and Fault Detection Using Reinforcement Learning

In this paper, we propose a learning-based method utilizing the Soft Actor-Critic (SAC) algorithm to train a binary Support Vector Machine (SVM) classifier. This classifier is designed to identify valid input spaces in high-dimensional, highly constrained systems while minimizing the total runtime of offline simulations. The simulations adapt their runtime based on the likelihood that a given training input will be informative to the classifier. Furthermore, we introduce a method for using the trained SAC model to predict whether a desired system input is likely to violate constraints, along with a technique to adjust the input as necessary. Additionally, we explore the potential of this model to detect faults or adversarial attacks within the system. The effectiveness of our approach is demonstrated through various simulations of challenging classification problems and a constrained quadrotor model.

Netter, Josh [Georgia Institute of Technology, Atl↗
DOE OSTI2020

EGS Collab Experiment 1: 3D Seismic Velocity Model and Updated Microseismic Catalog Using Transfer-Learning Aided Double-Difference Tomography

This package contains a 3D Seismic velocity model and an updated microseismic catalog associated with a proceedings paper (Chai et al., 2020) published in the 45th Workshop on Geothermal Reservoir Engineering. The 3D_seismic_velocity_model text file contains x (m), y(m), z(m), P-wave velocity (km/s), P-wave velocity quality indicator (1 for well-constrained; 0 for poorly constrained), S-wave velocity (km/s), and S-wave velocity quality indicator (1 for well-constrained; 0 for poorly constrained). The Updated_MEQ_catalog text file contains event origin time, x(m), y(m), z(m), error in x (m), error in y (m), error in z (m), and RMS misfit (millisecond). The 3D_seismic_P-wave_velocity_model animation file shows slices of the 3D P-wave velocity model. The 3D_seismic_S-wave_velocity_model animation file shows slices of the 3D S-wave velocity model. The Interactive_MEQ_locations API file is an interactive visualization of the updated microseismic event locations. The visualization allows users to view the event locations by dragging, rotating, and zooming in. References: Chai, C., Maceira, M., Santos-Villalobos, H. J., Venkatakrishnan, S. V., Schoenball, M., and EGS Collab Team, 2020, Automatic Seismic Phase Picking Using Deep Learning for the EGS Collab Project, in PROCEEDINGS, 45th Workshop on Geothermal Reservoir Engineering, edited, Stanford University, Stanford, California, 45, 1266-1276.

15 GEOTHERMAL ENERGY↗
DOE OSTI2024

Diversity of visual inputs to Kenyon cells of the Drosophila mushroom body

The arthropod mushroom body is well-studied as an expansion layer representing olfactory stimuli and linking them to contingent events. However, 8% of mushroom body Kenyon cells in Drosophila melanogaster receive predominantly visual input, and their function remains unclear. Here, we identify inputs to visual Kenyon cells using the FlyWire adult whole-brain connectome. Input repertoires are similar across hemispheres and connectomes with certain inputs highly overrepresented. Many visual neurons presynaptic to Kenyon cells have large receptive fields, while interneuron inputs receive spatially restricted signals that may be tuned to specific visual features. Individual visual Kenyon cells randomly sample sparse inputs from combinations of visual channels, including multiple optic lobe neuropils. These connectivity patterns suggest that visual coding in the mushroom body, like olfactory coding, is sparse, distributed, and combinatorial. However, the specific input repertoire to the smaller population of visual Kenyon cells suggests a constrained encoding of visual stimuli.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Energy Efficient Streaming Time Series Classification with Attentive Power Iteration

Efficiently processing time series data streams in real-time on resource-constrained devices offers significant advantages in terms of enhanced computational energy efficiency and reduced time-related risks. We introduce an innovative streaming time series classification network that utilizes attentive power iteration, enabling real-time processing on resource-constrained devices. Our model continuously updates a compact representation of the entire time series, enhancing classification accuracy while conserving energy and processing time. Notably, it excels in streaming scenarios without requiring complete time series access, enabling swift decisions. Experimental results show that our approach excels in classification accuracy and energy efficiency, with over 70% less consumption and threefold faster task completion than benchmarks. This work advances real-time responsiveness, energy conservation, and operational effectiveness for constrained devices, contributing to optimizing various applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

How to Safely Build 100-plus Kilograms of Weapons-Grade Plutonium

The goal of the EUCLID (Experiments Underpinned by Computational Learning for Improvements in Nuclear Data) project was to reduce compensating errors by utilizing machine learning to both help determine which reactions contain compensating errors as well as optimizing an experiment which can be used to maximally reduce these errors. Compensating errors can adversely impact the predictive power of application simulations, and therefore it’s useful to further constrain nuclear data and reduce these errors. The EUCLID project included building two configurations at the National Criticality Experiments Research Center (NCERC). These two configurations had very different geometries (one was cube-like and one was slab-like). Previous works focus on selection of the target experiment(s), radiation transport capabilities developed in the project, the experiment optimization, and the performance of the experiments. This work will focus only on the safety aspects of performing this experiment, which utilized over 100 kg of weapons-grade plutonium.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2020

Distributed non-negative matrix factorization with determination of the number of latent features

The holistic analysis and understanding of the latent (that is, not directly observable) variables and patterns buried in large datasets is crucial for data-driven science, decision making and emergency response. Such exploratory analyses require devising unsupervised learning methods for data mining and extraction of the latent features, and non-negative matrix factorization (NMF) is one of the prominent such methods. NMF is based on compute-intense non-convex constrained minimization, which, for large datasets requires fast and distributed algorithms. However, current parallel implementations of NMF fail to estimate the number of latent features. In practice, identifying these features is both difficult and significant for pattern recognition and latent feature analysis, especially for large dense matrices. Here, we introduce a distributed NMF algorithm coupled with distributed custom clustering followed by a stability analysis on dense data, which we call DnMFk, to determine the number of latent variables. The results on synthetic data and the classical Swimmer data set demonstrate the accuracy of model determination while scaling nearly linearly across multiple processors for large data. Further, we employ DnMFk to determine the number of hidden features from a terabyte matrix.

97 MATHEMATICS AND COMPUTING↗