SEARCH · Engineering Papers

Results for “generalizability”

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.

SourceResearch connectionFrom yearThrough year

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

At least 469 records · Page 26

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Orthogonal chemical genomics approaches reveal genomic targets for increasing anaerobic chemical tolerance in Zymomonas mobilis

Genetically engineered microbes have the potential to increase efficiency in the bioeconomy by overcoming growth-limiting production stress. Screens of gene perturbation libraries against production stressors can identify high-value engineering targets, but follow-up experiments needed to guard against false positives are slow and resource-intensive. In principle, the use of orthogonal gene perturbation approaches could increase recovery of true positives over false positives because the strengths of one technique compensate for the weaknesses of the other, but, in practice, two parallel screens are rarely performed at the genome scale. Here, we screen genome-scale CRISPRi (CRISPR interference) knockdown and transposon insertion libraries of the bioenergy-relevant Alphaproteobacterium, Zymomonas mobilis, against growth inhibitors commonly found in deconstructed plant material. Integrating data from the two gene perturbation techniques, we established an approach for defining engineering targets with high specificity. This allowed us to identify all known genes in the cytochrome bc1 and cytochrome c synthesis pathway as potential targets for engineering resistance to phenolic acids under anaerobic conditions, a subset of which we validated using precise gene deletions. Strikingly, this finding is specific to the cytochrome bc1 and cytochrome c pathway and does not extend to other branches of the electron transport chain. We further show that exposure of Z. mobilis to ferulic acid causes substantial remodeling of the cell envelope proteome, as well as the downregulation of TonB-dependent transporters. Our work provides a generalizable strategy for identifying high-value engineering targets from gene perturbation screens that is broadly applicable.

CRISPRi↗

DOE OSTI2022

When Do Extended Physics-Informed Neural Networks (XPINNs) Improve Generalization?

Physics-informed neural networks (PINNs) have become a popular choice for solving high-dimensional partial differential equations (PDEs) due to their excellent approximation power and generalization ability. Recently, extended PINNs (XPINNs) based on domain decomposition methods have attracted considerable attention due to their effectiveness in modeling multiscale and multiphysics problems and their parallelization. However, theoretical understanding of their convergence and generalization properties remains unexplored. In this study, we take an initial step towards understanding how and when XPINNs outperform PINNs. Specifically, for general multilayer PINNs and XPINNs, we first provide a prior generalization bound via the complexity of the target functions in the PDE problem and a posterior generalization bound via the posterior matrix norms of the networks after optimization. Moreover, based on our bounds, we analyze the conditions under which XPINNs improve generalization. Concretely, our theory shows that the key building block of XPINN, namely, the domain decomposition, introduces a tradeoff for generalization. On the one hand, XPINNs decompose the complex PDE solution into several simple parts, which decreases the complexity needed to learn each part and boosts generalization. On the other hand, decomposition leads to less training data being available in each subdomain, and hence such a model is typically prone to overfitting and may become less generalizable. Empirically, we choose five PDEs to show when XPINNs perform better than, similar to, or worse than PINNs, hence demonstrating and justifying our new theory.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Projective Integral Updates for High-Dimensional Variational Inference

Variational inference is an approximation framework for Bayesian inference that seeks to improve quantified uncertainty in predictions by optimizing a simplified distribution over parameters to stand in for the full posterior. Capturing model variations that remain consistent with training data enables more robust predictions by reducing parameter sensitivity. This work introduces a fixed-point optimization for variational inference that is applicable when every feasible log density can be expressed as a linear combination of functions from a given basis. In such cases, the optimizer becomes a fixed-point of projective integral updates. When the basis spans univariate quadratics in each parameter, the feasible distributions are Gaussian mean-fields and the projective integral updates yield quasi-Newton variational Bayes (QNVB). Other bases and updates are also possible. Since these updates require high-dimensional integration, this work begins by proposing an efficient quasirandom sequence of quadratures for mean-field distributions. Each iterate of the sequence contains two evaluation points that combine to correctly integrate all univariate quadratic functions and, if the mean-field factors are symmetric, all univariate cubics. More importantly, averaging results over short subsequences achieves periodic exactness on a much larger space of multivariate polynomials of quadratic total degree. The corresponding variational updates require four loss evaluations with standard (not second-order) backpropagation to eliminate error terms from over half of all multivariate quadratic basis functions. Furthermore, this integration technique is motivated by first proposing stochastic blocked mean-field quadratures, which may be useful in other contexts. A PyTorch implementation of QNVB allows for better control over model uncertainty during training than competing methods. Experiments demonstrate superior generalizability for multiple learning problems and architectures.

Gaussian mean-field↗

DOE OSTI2023

Hardware-Conscious Optimization of the Quantum Toffoli Gate

While quantum computing holds great potential in combinatorial optimization, electronic structure calculation, and number theory, the current era of quantum computing is limited by noisy hardware. Many quantum compilation approaches can mitigate the effects of imperfect hardware by optimizing quantum circuits for objectives such as critical path length. Few approaches consider quantum circuits in terms of the set of vendor-calibrated operations (i.e., native gates) available on target hardware. This manuscript expands the analytical and numerical approaches for optimizing quantum circuits at this abstraction level. We present a procedure for combining the strengths of analytical native gate-level optimization with numerical optimization. Although we focus on optimizing Toffoli gates on the IBMQ native gate set, the methods presented are generalizable to any gate and superconducting qubit architecture. Our optimized Toffoli gate implementation demonstrates an 18% reduction in infidelity compared with the canonical implementation as benchmarked on IBM Jakarta with quantum process tomography. Assuming the inclusion of multi-qubit cross-resonance (MCR) gates in the IBMQ native gate set, we produce Toffoli implementations with only six multi-qubit gates, a 25% reduction from the canonical eight multi-qubit implementations for linearly connected qubits.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2025

Poster Abstract: Leveraging Large Language Models to Reveal Interpretable Cooling Behaviors from Smart Thermostat Data

Frequent heatwaves and hot summers increasingly challenge occupant comfort, health, and energy grid stability. Addressing these challenges requires a detailed understanding of household cooling behaviors, such as thermostat adjustments and adaptive responses to extreme conditions. Traditional analyses often rely on aggregated numerical metrics that overlook subtle but important household-specific variations. In this study, we introduce a generalizable methodology that integrates large language models (LLMs) with vision capabilities to enable scalable and detailed analysis of residential thermostat data. Using Ecobee's Donate Your Data (DYD) dataset—which provides five-minute records of indoor temperatures, thermostat setpoints, and HVAC runtimes—we focus on two U.S. cities with contrasting summer climates : Austin (TX) and Phoenix (AZ). Because raw time-series data are not well suited for direct LLM analysis, we transform them into visual representations, such as daily indoor temperature trajectories and weekly runtime histograms, to better capture behavioral variations. Leveraging LLMs' visual interpretation, we extract descriptive behavioral features, including temperature preferences, time-of-day cooling orientation, anticipatory versus reactive heatwave responses, and behavioral consistency. These semantic features support unsupervised clustering to identify distinct occupant archetypes at scale, revealing differences—such as morning-centric anticipatory coolers versus households that shift toward warmer setpoints during heatwaves—that can inform demand response, resilience planning, and health-aware interventions. By converting raw numerical data into interpretable behavioral patterns, this methodology enables scalable and practical analysis of occupant behavior, supporting actionable insights for comfort, resilience, and energy management.

Nihar, Kopal↗

DOE OSTI2026

TCAD-Machine Learning Enabled TID Compact Model Development for Commercial SiC MOSFET

We propose a TCAD (Technology Computer Aided Design)-machine learning coupled approach that combines a TCAD tool (Charon), optimization/uncertainty quantification tool (Dakota), surrogate models, and Bayesian learning capabilities. The coupling approach is used for accurate modeling and calibration of total ionizing dose (TID) induced threshold voltage (V th ) shifts in Commercial-Off-The-Shelf (COTS) semiconductor devices and to develop physics-informed TID compact models. This versatile approach is applied to model the TID effect in an exemplar COTS 3.3 kV SiC power MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor). With the Charon-Dakota coupling, we can determine key device geometry and doping values based on device physics, which are difficult to obtain or not available for COTS devices but important for TCAD simulation; additionally, we can efficiently generate thousands of simulation results in a large parameter space, which makes it possible to develop data-driven surrogate models and perform Bayesian calibration. Utilizing the full tool-coupling approach, we achieve calibrated TCAD simulation models that accurately capture the average TID-induced V th shifts behavior with total doses and V th shifts saturation at high doses as observed in experimental data. More importantly, the calibrated TCAD simulations are obtained with determined TID model parameters (e.g., hole trap density and capture cross section) values that contain well quantified uncertainties. Furthermore, we can isolate and quantify the noises that are not captured by the TCAD models but exist in the measured data due to measurements and devices variabilities. Lastly, the calibrated surrogate models are used to develop physics-informed TID compact models. The method is generalizable to other devices and/or radiation conditions with few modifications and can provide well-determined uncertainties.

COTS↗

DOE OSTI2026

Identifying Critical Electrode Metrics for Efficient, Selective CO 2 Electrochemical Conversion

Low-temperature electrochemical CO 2 reduction (CO 2 R) in zero-gap membrane electrode assembly (MEA) reactors presents a scalable route to fuels and carbon utilization. However, performance at industrially relevant current densities hinges on mesoscale catalyst layer integration, particularly at the ionomer|catalyst interface. Here, we demonstrate a generalizable in situ electrochemical impedance spectroscopy (EIS) method. We utilize this technique to decouple electrode-level parameters that are correlated to the overall MEA performance. By performing this ex situ EIS method on CO 2 -to-CO catalyst-coated membranes with systematically varied ionomer-to-catalyst (I:C) ratios, we reveal a pronounced dependence of performance, ion transport resistance, and catalyst utilization on the I:C ratio as well as the electrode conditioning. We demonstrate that an optimal I:C ratio exists at which ion transport resistance is minimized and Faradaic efficiency for CO production is maximized. Beyond the electrodes examined, here we compare ion transport resistance to MEA selectivity/Faradaic efficiency obtained in prior studies, revealing a clear correlation between the two. These results suggest that ion transport resistance within the catalyst layer may be a quantitative predictor of MEA performance which underscores the importance of mesoscale integration in achieving scalable CO 2 R technologies.

08 HYDROGEN↗

DOE OSTI2021

SSRS (Stochastic Soaring Raptor Simulator)

SSRS (Stochastic Soaring Raptor Simulator) is a generalizable, probabilistic, and predictive tool for wind energy developers, ecologists, wildlife managers and industry consultants to estimate the potential for soaring raptors to interact with operating wind turbines, without the need for site-specific data collection. Rapid expansion of wind energy development across the world has exposed the risk of turbine collisions for birds and bats. The risk to obligate soaring raptors such as golden eagles is of particular concern due to their small population and influence on ecological balance. Golden eagles rely heavily on updrafts to subsidize their flight, putting them in direct conflict with operational wind turbines that utilize the same wind resource. Understanding the behavior of soaring raptors with varying atmospheric conditions is crucial for predicting and mitigating the risk of turbine collision. This software contains a predictive movement model that simulates individual flight paths of golden eagles during updraft-subsidized long-distance flight, including migration. For a given set of atmospheric conditions, the model simulates thousands of eagles at turbine-scale spatial resolution (50m) to produce a relative presence density map. The simulated eagles rely on updrafts to pursue uninterrupted directional flight with minimal energy expenditure, following fluid-flow principles. The simulator includes a stochastic model of eagle behavior and a systematic method of accounting for spatiotemporal variations in atmospheric conditions. This framework only requires publicly available atmospheric data to estimate orographic and thermal updrafts, ensuring general usability.

Sandhu, Rimple↗

DOE OSTI2021

Multiscale Machine-Learned Modeling Infrastructure Core

MuMMI Core is the central component of the Multiscale Machine-Learned Modeling Infrastructure (MuMMI), a scalable and generalizable framework that couples pairs of models using machine learning and in situ feedback across three scales of resolution.

Bhatia, Harsh↗

DOE OSTI2022

GridDS: Data Science Toolkit for Energy Grid Data

According to the U.S. Energy Information Administration (EIA), the demand for energy is expected to increase 50% by the year 20501. While energy standards, such as the Institute of Electrical and Electronics Engineers (IEEE) Standard 1547, (Basso 2015) and monitoring with wide area management systems (WAMS) (Liu 2017, Zhou 2016) have enabled large scale data collection and storage, the application of this data in mitigating costs associated with increased consumer demand is an ongoing focus for energy research. This ubiquitous data collection presents a promising opportunity for machine learning and data science to improve efficiency of distributed energy resources (DERs). The GridDS software toolkit is designed to leverage advanced metering infrastructure (AMI), outage management systems data (OMS), Supervisory control Data Acquisition (SCADA), and geographic information systems (GIS) to forecast future energy demands and detect incipient grid failures. GridDS is a python software library designed to be modular and generalizable to data recorded by DERs. In adapting to disparate datasets recorded by various WAMS, GridDS provides a range of unique functionality not presently implemented in current WAMS which have highly specific software infrastructure by design. GridDS functionality ranges from data specification and preparation, to training and validation for state of the art machine learning, to interactive data visualization. For data intake, GridDS combines: Pandera: a library for creating data specifications. TimeScaleDB: a postgresSQL database infrastructure for efficient storage of timeseries data. Dataset class: A custom dataset class / interface that ensures modularity between a range of synthetic and live recorded datasets. Is

Ladd, Alexander↗

DOE OSTI2024

DDStore

DDStore is designed to addresses random, read-oriented, global shuffle operations for data-intensive deep learning applications. By preserving global shuffling, DDStore ensures that deep learning model maintains high accuracy and generalizability when it is trained with large volume of data using HPC facilities.

Choi, Jong Youl [Oak Ridge National Laboratory (OR↗

DOE OSTI2023

A python package for analyzing Resilience of Complex Systems (pyRoCS) v.0.0

SAND2024-01040O PyRoCS software synthesizes mathematical equations from several domains—including information theory, ecology, and engineering sciences—to support resilience analysis for complex systems. Resilience is the ability of the complex system being analyzed to withstand, operate through, and recover from a disruption. The complex system can be a physical system such as an electric grid, an organization such as a company, or even a subfunction of an organization. Existing mathematical equations for resilience analysis are found within multiple domains including information theory, biological sciences, and complex systems. This package synthesizes and refactors equations from these various domains to make them more generalizable for application across different types of complex systems relevant for resilience analysis. Users will be able to apply these equations to characterize different components of complex systems based on available data. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Verzi, Stephen↗

DOE OSTI2024

FTL: Transfer Learning Nonlinear Plasma Dynamic Transitions in Low Dimensional Embeddings (FTL) v1.0

Fusion Transfer Learning (FTL) model provides a new paradigm to study high-dimensional dynamical behaviors, such as those in fusion plasma systems. The knowledge transfer process leverages a pre-trained neural encoder-decoder network, initially trained on linear simulations, to effectively capture nonlinear dynamics. The low-dimensional embeddings extract the coherent structures of interest, while preserving the inherent dynamics of the complex system. Experimental results highlight FTL's capacity to capture transitional behaviors and dynamical features in plasma dynamics -- a task often challenging for conventional methods. The model developed in this study is generalizable and can be extended broadly through transfer learning to address various magnetohydrodynamics (MHD) modes.

Bai, Zhe↗

DOE OSTI2026

CVEVOLVE

CVEvolve is an agentic AI system for autonomous algorithm discovery for scientific data processing. It creates workflows where large language model agents freely set up and configure development environments and evaluation harnesses, develop and improve data processing algorithms with designed exploration-exploitation balancing mechanisms, log history and findings in a structured database, and run holdout testing to ensure algorithm generalizability. CVEvolve offers a zero-code interface and does not require users to provide structured data and evaluation scripts.

Cherukara, MatthewJoseph [Argonne National Laborat↗

DOE OSTI2024

Statistical Treatment of Convolutional Neural Network Superresolution of Inland Surface Wind for Subgrid-Scale Variability Quantification

Abstract Machine learning models have been employed to perform either physics-free data-driven or hybrid dynamical downscaling of climate data. Most of these implementations operate over relatively small downscaling factors because of the challenge of recovering fine-scale information from coarse data. This limits their compatibility with many global climate model outputs, often available between ∼50- and 100-km resolution, to scales of interest such as cloud resolving or urban scales. This study systematically examines the capability of a type of superresolving convolutional neural network (SR-CNNs) to downscale surface wind speed data over land from different coarse resolutions (25-, 48-, and 100-km resolution) to 3 km. For each downscaling factor, we consider three convolutional neural network (CNN) configurations that generate superresolved predictions of fine-scale wind speed, which take between one and three input fields: coarse wind speed, fine-scale topography, and diurnal cycle. In addition to fine-scale wind speeds, probability density function parameters are generated through which sample wind speeds can be generated, accounting for the intrinsic stochasticity of wind speed. For assessing generalization to new data, CNN models are tested on regions with different topography and climate that are unseen during training. The evaluation of superresolved predictions focuses on subgrid-scale variability and the recovery of extremes. Models with coarse wind and fine topography as inputs exhibit the best performance when compared with other model configurations, operating across the same downscaling factor. Our diurnal cycle encoding results in lower out-of-sample generalizability when compared with other input configurations.

17 WIND ENERGY↗

DOE OSTI2024

Machine Learning Approach for Spatiotemporal Multivariate Optimization of Environmental Monitoring Sensor Locations

Abstract Long-term environmental monitoring is critical for managing the soil and groundwater at contaminated sites. Recent improvements in state-of-the-art sensor technology, communication networks, and artificial intelligence have created opportunities to modernize this monitoring activity for automated, fast, robust, and predictive monitoring. In such modernization, it is required that sensor locations be optimized to capture the spatiotemporal dynamics of all monitoring variables as well as to make it cost-effective. The legacy monitoring datasets of the target area are important to perform this optimization. In this study, we have developed a machine-learning approach to optimize sensor locations for soil and groundwater monitoring based on ensemble supervised learning and majority voting. For spatial optimization, Gaussian process regression (GPR) is used for spatial interpolation, while the majority voting is applied to accommodate the multivariate temporal dimension. Results show that the algorithms significantly outperform the random selection of the sensor locations for predictive spatiotemporal interpolation. While the method has been applied to a four-dimensional dataset (with two-dimensional space, time, and multiple contaminants), we anticipate that it can be generalizable to higher-dimensional datasets for environmental monitoring sensor location optimization.

Siddiquee, Masudur R.↗

DOE OSTI2024

A Case Study Investigating the Low Summertime CAPE Behavior in the Global Forecast System

Convective available potential energy (CAPE) is an important index for storm forecasting. Recent versions (v15.2 and v16) of the Global Forecast System (GFS) predict lower values of CAPE during summertime in the continental United States than analysis and observation. We conducted an evaluation of the GFS in simulating summertime CAPE using an example from the Unified Forecast System Case Study collection to investigate the factors that lead to the low CAPE bias in GFS. Specifically, we investigated the surface energy budget, soil properties, and near-surface and upper-level meteorological fields. Results show that the GFS simulates smaller surface latent heat flux and larger surface sensible heat flux than the observations. This can be attributed to the slightly drier-than-observed soil moisture in the GFS that comes from an offline global land data assimilation system. The lower simulated CAPE in GFS v16 is related to the early drop of surface net radiation with excessive boundary layer cloud after midday when compared with GFS v15.2. A moisture-budget analysis indicates that errors in the large-scale advection of water vapor does not contribute to the dry bias in the GFS at low levels. Common Community Physics Package single-column model (SCM) experiments suggest that with realistic initial vertical profiles, SCM simulations generate a larger CAPE than runs with GFS IC. SCM runs with an active LSM tend to produce smaller CAPE than that with prescribed surface fluxes. Note that the findings are only applicable to this case study. Including more warm-season cases would enhance the generalizability of our findings.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2026

Substitution or Shared Utilization? Intrahousehold Vehicle Use in Mixed-Powertrain Households

While previous research has focused heavily on understanding the factors deriving alternative fuel vehicle adoption rates, there remains a significant gap in understanding how households distribute mileage across different powertrains. This study utilizes data from the 2022 Next Generation National Household Travel Survey to investigate vehicle miles traveled within a sample of 150 plug-in electric vehicle (PEV)-owning households (in which at least one battery electric vehicle is present), characterizing how different powertrains are integrated into daily mobility. Leveraging a Seemingly Unrelated Regression (SUR) framework the study jointly models the utilization of PEVs, hybrid electric vehicles (HEV), and internal combustion engine vehicles (ICEVs) while accounting for household-level substitution effects. The results provide evidence of an asymmetric substitution effect. In households with mixed-powertrain configurations, the ICEV captures a substantially higher share of household miles (compared with the PEV), acting as a utility sponge. Conversely, the model identifies specific socioeconomic and geographic cohorts that prioritize PEV as the primary household workhorse, indicating a systematic sorting effect. Although the sample size limits broader generalizability, these findings suggest that PEVs are used for frequent, specific routine-intensive roles, whereas the ICEV remains a specialized utility vehicle. These insights highlight distinct intrahousehold vehicle use behaviors that are often obscured by aggregate fleetwide statistics.

25 ENERGY STORAGE↗