Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Adaptation 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 901 records · Page 50
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Model Assumptions and Data Characteristics: Impacts on Domain Adaptation in Building Segmentation

Studies on domain adaptation (DA) for remote sensing (RS) imagery analysis lack consistency in selection and description of evaluation scenarios. Without properly characterizing datasets, model assumptions, and evaluation scenarios, it is difficult to objectively compare DA methods and reach conclusions about their suitability across different applications. With this motivation, this work seeks to empirically assess to which extent the interaction between data characteristics and model assumptions influences the effectiveness of DA methods. Using the widely explored task of building footprint segmentation as a case study, we perform a large-scale study across over 200 DA scenarios that include variations across view angles, areas observed, and sensors used for data acquisition. Rather than adopting different model architectures or optimization criteria, we contrast the performances of two DA methods based on adversarial learning that differ only in their assumptions about source and target domains. Informed by metadata and data characteristics unveiled using traditional computer vision (CV) techniques as well as pretrained deep models, we provide a detailed meta-analysis of experiments highlighting the importance of accurately considering data assumptions for DA in RS segmentation tasks. As demonstrated by a “cherry-picking” exercise, different claims regarding which model is best could be made by selecting different subsets of evaluation scenarios. While well-calibrated assumptions can be beneficial, mismatching assumptions can lead to negative biases in DA applications. Furthermore, this study intends to motivate the community toward more consistent evaluation protocols while providing recommendations and insights toward creating novel benchmark datasets, documenting data characteristics, application-specific knowledge, and model assumptions.

42 ENGINEERING↗
DOE OSTI2023

Toward implementing autonomous adaptive data acquisition for scanning hyperspectral imaging of biological systems

Autonomous experimentation is an emerging area of research, primarily related to autonomous vehicles, scientific combinatorial discovery approaches in materials science and drug discovery, and iterative research loops of planning, experimentation, and analysis. However, autonomous approaches developed in these contexts are difficult to apply to high-dimensional mapping technologies, such as scanning hyperspectral imaging of biological systems, due to sample complexity and heterogeneity. We briefly cover the history of adaptive sampling algorithms and surrogate modeling in order to define autonomous adaptive data acquisition as an objective-based, flexible building block for future biological imaging experimentation driven by intelligent infrastructure. We subsequently summarize the recent implementations of autonomous adaptive data acquisition (AADA) for scanning hyperspectral imaging, assess how these address the difficulties of autonomous approaches in hyperspectral imaging, and highlight the AADA design variation from a goal-oriented perspective. Finally, we present a modular AADA architecture that embeds AADA-driven flexible building blocks to address the challenge of time resolution for high-dimensional scanning hyperspectral imaging of nonequilibrium dynamical systems. In our example research-driven experimental design case, we propose an AADA infrastructure for time-resolved, noninvasive, and label-free scanning hyperspectral imaging of living biological systems. This AADA infrastructure can accurately target the correct state of the system for experimental workflows that utilize subsequent expensive, high-information-content analytical techniques.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Strongly Heated Turbulent Flow in a Channel with Pin Fins

Large-eddy simulations (LES) were performed to study the turbulent flow in a channel of height H with a staggered array of pin fins with diameter D = H/2 as a function of heating loads that are relevant to the cooling of turbine blades and vanes. The following three heating loads were investigated—wall-to-coolant temperatures of Tw/Tc = 1.01, 2.0, and 4.0 - where the Reynolds number at the channel inlet was 10,000 and the back pressure at the channel outlet was 1 bar. For the LES, two different subgrid-scale models—the dynamic kinetic energy model (DKEM) and the wall-adapting local eddy-viscosity model (WALE)—were examined and compared. This study was validated by comparing with data from direct numerical simulation and experimental measurements. The results obtained show high heating loads to create wall jets next to all heated surfaces that significantly alter the structure of the turbulent flow. Results generated on effects of heat loads on the mean and fluctuating components of velocity and temperature, turbulent kinetic energy, the anisotropy of the Reynolds stresses, and velocity-temperature correlations can be used to improve existing RANS models.

42 ENGINEERING↗
NASA NTRS1982

Snowmelt runoff modeling in simulation and forecasting modes with the Martinec-Mango model

The Martinec-Rango snowmelt runoff model was applied to two watersheds in the Rio Grande basin, Colorado-the South Fork Rio Grande, a drainage encompassing 216 sq mi without reservoirs or diversions and the Rio Grande above Del Norte, a drainage encompassing 1,320 sq mi without major reservoirs. The model was successfully applied to both watersheds when run in a simulation mode for the period 1973-79. This period included both high and low runoff seasons. Central to the adaptation of the model to run in a forecast mode was the need to develop a technique to forecast the shape of the snow cover depletion curves between satellite data points. Four separate approaches were investigated-simple linear estimation, multiple regression, parabolic exponential, and type curve. Only the parabolic exponential and type curve methods were run on the South Fork and Rio Grande watersheds for the 1980 runoff season using satellite snow cover updates when available. Although reasonable forecasts were obtained in certain situations, neither method seemed ready for truly operational forecasts, possibly due to a large amount of estimated climatic data for one or two primary base stations during the 1980 season.

Shafer, B.↗
NASA NTRS1991

Analysis of field measurements of carbon dioxide and water vapor fluxes

Analysis of the field measurements of carbon dioxide and water vapor fluxes is discussed. These data were examined in conjunction with reflectance obtained from helicopter mounted Modular Multiband Radiometer. These measurements are representative of the canopy scale (10 to 100 m)(exp 2) and provide a good basis for investigating the hypotheses/relationship potentially useful in remote sensing applications. All the micrometeorological data collected during FIFE-89 were processed and fluxes of CO2, water vapor, and sensible heat were calculated. Soil CO2 fluxes were also estimated. Employing these soil CO2 flux values, in conjunction with micrometeorological measurements, canopy photosynthesis is being estimated. A biochemical model of leaf photosynthesis was adapted to the prairie vegetation. The modeled leaf photosynthesis rates were scaled up to the canopy level. This model and a multiplicative stomatal conductance model are also used to calculate canopy conductance.

Verma, Shashi B.↗
NASA NTRS2008

Direct Adaptive Control of Systems with Actuator Failures: State of the Art and Continuing Challenges

In this paper, the problem of controlling systems with failures and faults is introduced, and an overview of recent work on direct adaptive control for compensation of uncertain actuator failures is presented. Actuator failures may be characterized by some unknown system inputs being stuck at some unknown (fixed or varying) values at unknown time instants, that cannot be influenced by the control signals. The key task of adaptive compensation is to design the control signals in such a manner that the remaining actuators can automatically and seamlessly take over for the failed ones, and achieve desired stability and asymptotic tracking. A certain degree of redundancy is necessary to accomplish failure compensation. The objective of adaptive control design is to effectively use the available actuation redundancy to handle failures without the knowledge of the failure patterns, parameters, and time of occurrence. This is a challenging problem because failures introduce large uncertainties in the dynamic structure of the system, in addition to parametric uncertainties and unknown disturbances. The paper addresses some theoretical issues in adaptive actuator failure compensation: actuator failure modeling, redundant actuation requirements, plant-model matching, error system dynamics, adaptation laws, and stability, tracking, and performance analysis. Adaptive control designs can be shown to effectively handle uncertain actuator failures without explicit failure detection. Some open technical challenges and research problems in this important research area are discussed.

Tao, Gang↗
NASA NTRS1994

High Performance Programming Using Explicit Shared Memory Model on the Cray T3D

The Cray T3D is the first-phase system in Cray Research Inc.'s (CRI) three-phase massively parallel processing program. In this report we describe the architecture of the T3D, as well as the CRAFT (Cray Research Adaptive Fortran) programming model, and contrast it with PVM, which is also supported on the T3D We present some performance data based on the NAS Parallel Benchmarks to illustrate both architectural and software features of the T3D.

Saini, Subhash↗
DOE OSTI2025

EvoNet: A phylogenomic and systems biology approach to identify genes underlying plant survival in marginal, low‐N soils

The DOE‐BER “EvoNet” project investigates the genetic and molecular basis of plant resilience in extreme environments. We do this by identifying key genes that enable “extreme survivor” species to thrive in the nitrogen-poor soils of Chile’s hyper-arid Atacama Desert. Our collections focus on 32 Atacama extremophile species, including seven grass species with potential biofuel applications. To identify genes-of-importance to survival we compared genomic and transcriptomic profiles of extremophile species that thrive in the Atacama to those of closely related “sister” species from nitrogen-rich arid and mesic regions of California. Deep RNA sequencing and de novo transcriptome assembly across these triplet species sets supported a phylogenomic framework for identifying positively selected genes associated with adaptive divergence. Our integrative analysis combined ecological and environmental data, metagenomics, evolutionary and systems biology, and metabolomics. This enabled us to create an unprecedented framework for systematically understanding how non-model plants have adapted to survive in extreme conditions. Our resulting database of positively selected ortholog groups in the extremophile plants offers promising targets for engineering crop and biofuel species with enhanced resilience to drought and extreme weather. Additionally, our newest dataset explores and exploits a complementary metabolomic approach. This new aspect provides innovative strategies to manipulate plant cell metabolism, further supporting efforts to improve agricultural productivity in the face of extreme climates. Importantly, our combined evolutionary- and metabolomic-based strategies focused on convergent patterns of adaptation, providing a genetic and metabolomic toolkit for improving crop and biofuel resilience across diverse plant species. Finally, our novel exploration of ecological and evolutionary dynamics delivered to the community a phylogenomic computational pipeline called “PhyloGeneious.” Our continued adaptations of this pipeline are publicly available to expedite evolutionary genomic research for future scientific discoveries. In total, our DOE-BER has provided genomic, metabolomic, and computational strategies to understand how extremophile plants provide evolutionary and physiological targets for improving agricultural and biofuel production.

59 BASIC BIOLOGICAL SCIENCES↗
NASA NTRS1989

Adaptive structures

The fundamental principles of the adaptive structures concept and its applications to current and planned space missions are reviewed and illustrated with diagrams, drawings, graphs, and photographs. An adaptive structure is defined as one which can be modified to meet mission requirements, either by remote commands or automatically in response to external stimuli. Topics addressed include the need for adaptive structures, analytical models, ground testing, sensor/actuator-structure interactions, structural concepts, active damping, wave propagation in large structures, the selection of active member locations, and on-orbit system identification. Particular attention is given to adaptive structures being developed for the NASA Large Deployable Reflector and Optical Interferometer projects.

Wada, Ben K.↗
NASA NTRS1992

SRMAFTE facility checkout model flow field analysis

The Solid Rocket Motor Air Flow Equipment (SRMAFTE) facility was constructed for the purpose of evaluating the internal propellant, insulation, and nozzle configurations of solid propellant rocket motor designs. This makes the characterization of the facility internal flow field very important in assuring that no facility induced flow field features exist which would corrupt the model related measurements. In order to verify the design and operation of the facility, a three-dimensional computational flow field analysis was performed on the facility checkout model setup. The checkout model measurement data, one-dimensional and three-dimensional estimates were compared, and the design and proper operation of the facility was verified. The proper operation of the metering nozzles, adapter chamber transition, model nozzle, and diffuser were verified. The one-dimensional and three-dimensional flow field estimates along with the available measurement data are compared.

Dill, Richard A.↗
DOE OSTI2023

An evaluation of meteorological data prediction over Washington, D.C.: Comparison of DCNet observations and NAM model outputs

This study presents an example of how outputs of operational and readily-available mesoscale numerical models can be adapted to initialize dispersion calculations within the urban surface roughness layer. Three years of urban meteorological observations from central Washington, DC, are compared against forecast outputs of the North American Mesoscale (NAM) model. NAM wind speed predictions underestimate the observations in light winds and overestimate the measurements in high winds. Average wind directions are consistent. However, an adjustment of the predicted direction of the plume by –20° is needed. The uncertainty associated with this adjustment is large in light NAM wind speed with no evident variation by season. The values of the standard deviation of the wind direction, σθ derived from NAM model outputs underestimate the observations by a small amount (about –1.5 to –2.5°). The results presented here indicate that mesoscale numerical model outputs can provide information adequate for dispersion calculations. However, levels of uncertainty associated with implementation of the suggested procedures increase with decreasing wind speed, causing considerable uncertainty in the implementation of adjustments as low wind speed conditions are approached. Results and recommendations reported here should not be extended to other numerical models or other cities without further testing.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Uncertainty Propagation within Chained Models for Machine Learning Reconstruction of Neutrino-LAr Interactions

Sequential or chained models are increasingly prevalent in machine learning for scientific applications, due to their flexibility and ease of development. Chained models are particularly useful when a task is separable into distinct steps with a hierarchy of meaningful intermediate representations. In reliability-critical tasks, it is important to quantify the confidence of model inferences. However, chained models pose an additional challenge for uncertainty quantification, especially when input uncertainties need to be propagated. In such cases, a fully uncertainty-aware chain of models is required, where each step accepts a probability distribution over the input space, and produces a probability distribution over the output space. In this work, we present a case study for adapting a single model within an existing chain, designed for reconstruction within neutrino-Argon interactions, developed for neutrino oscillation experiments such as MicroBooNE, ICARUS, and the future DUNE experiment. We test the performance of an input uncertainty-enabled model against an uncertainty-blinded model using a method for generating synthetic noise. By comparing these two, we assess the increase in inference quality achieved by exposing models to upstream uncertainty estimates.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1998

Psychological and Behavioral Health Issues of Long-Duration Space Missions

It will be the responsibility of the long-duration space flight crew to take the actions necessary to maintain their health and well-being and to cope with medical emergencies without direct assistance from support personnel, including maintaining mental health and managing physiological and psychological changes that may impair decision making and performance. The Behavior and Performance Integrated Product Team at Johnson Space Center, working, within the Space Medicine, Monitoring, and Countermeasures Program, has identified critical questions pertaining to long-duration space crew behavioral health, psychological adaptation, human factors and habitability, and sleep and circadian rhythms. Among the projects addressing these questions are: the development of tools to assess cognitive functions during space missions; the development of a model of psychological adaptation in isolated and confined environments; tools and methods for selecting individuals and teams well-suited for long-duration missions; identification of mission-critical tasks and performance evaluation; and measures of sleep quality and correlation to mission performance.

Eksuzian, Daniel J.↗
DOE OSTI2022

A quantitative model of charge injection by ruthenium chromophores connecting femtosecond to continuous irradiance conditions

A kinetic framework for the ultrafast photophysics of tris(2,2-bipyridine)ruthenium(II) phosphonated and methyl-phosphonated derivatives is used as a basis for modeling charge injection by ruthenium dyes into a semiconductor substrate. By including the effects of light scattering, dye diffusion, and adsorption kinetics during sample preparation and the optical response of oxidized dyes, quantitative agreement with multiple transient absorption datasets is achieved on timescales spanning femtoseconds to nanoseconds. In particular, quantitative agreement with important spectroscopic handles—the decay of an excited state absorption signal component associated with charge injection in the UV region of the spectrum and the dynamical redshift of a ~500 nm isosbestic point—validates our kinetic model. Pseudo-first-order rate coefficients for charge injection are estimated in this work, with an order of magnitude ranging from 10 11 to 10 12 s –1 . The model makes the minimalist assumption that all excited states of a particular dye have the same charge injection coefficient, an assumption that would benefit from additional theoretical and experimental exploration. Here, we have adapted this kinetic model to predict charge injection under continuous solar irradiation and find that as many as 68 electron transfer events per dye per second take place, significantly more than prior estimates in the literature.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

A Nonstationary and Non-Gaussian Moving Average Model for Solar Irradiance

Historically, power has flowed from large power plants to customers. Increasing penetration of distributed energy resources such as solar power from rooftop photovoltaic has made the distribution network a two-way-street with power being generated at the customer level. The incorporation of renewables introduces additional uncertainty and variability into the power grid. Distribution network operation studies are being adapted to include renewables; however, such studies require high quality solar irradiance data that adequately reflect realistic meteorological variability. Data from satellite-based products are spatially complete, but temporally coarse, whereas solar irradiances exhibit high frequency variation at very fine timescales. We propose a new stochastic method for temporally downscaling global horizontal irradiance (GHI) to 1 min resolution, but we do not consider the spatial aspect due to limited availability of the in situ irradiance measurements. Solar irradiance's first and second-order structures vary diurnally and seasonally, and our model adapts to such nonstationarity. Empirical irradiance data exhibits highly non-Gaussian behavior; we develop a nonstationary and non-Gaussian moving average model that is shown to capture realistic solar variability at multiple timescales. We also propose a new estimation scheme based on Cholesky factors of empirical autocovariance matrices, bypassing difficult and inaccessible likelihood-based approaches. The model is demonstrated for a case study of three locations that are located in diverse climates through the United States. The model is compared against competitors from the literature and is shown to provide better uncertainty and variability quantification on testing data.

Cholesky factor↗
DOE OSTI2024

Distributed-Memory Sparse Deep Neural Network Inference Using Global Arrays

Partitioned Global Address Space (PGAS) models exhibit tremendous promise in developing efficient and productive distributed-memory parallel applications. They have been used extensively in scientific computations due to conveniently offering a ``shared-memory''-like model and convenient interfaces that separate communication with synchronization. Traditionally, PGAS communication models have been applied to dense/contiguously distributed data, but most modern applications depict varied levels of sparsity. Existing PGAS models require certain adaptations to support distributed sparse computations, since associated computations often require matrix arithmetic, in addition to data movement. The Global Arrays toolkit from Pacific Northwest National Laboratory (PNNL) is one of the earliest PGAS models to combine one-sided data communication and distributed matrix operations and is still used in the popular NWChem quantum chemistry suite. Recently, we have expanded the Global Arrays toolkit to support common sparse operations, like sparse matrix-dense matrix multiplies (SpMM), sparse matrix-sparse matrix multiplication (SpGEMM) and Sampled Dense-Dense Matrix Multiplication (SDDMM). As it turns out, these operations are the bedrock of sparse Deep Learning (DL); sparse deep neural networks and Graph Neural Networks (GNNs) have gained increasing attention recently in achieving speedups on training and inference with reduced memory footprints. Unlike scientific applications in High Performance Computing (HPC), modern (distributed-memory capable) DL toolkits often rely on non-standardized and closed-source vendor software optimizations, creating challenges in software-hardware co-design at scale. Our goal is to support a variety of distributed-memory sparse matrix operations and helper functions in the newly created Sparse Global Arrays (SGA), such that it is possible to build portable and productive Machine Learning scenarios for algorithm/software and hardware codesign purposes. Contemporary data-parallel schemes for training/inference are undergoing a major overhaul since model replication limits scalability and causes resource inefficiencies. As such, we have adopted tensor parallelism in decomposing the model and inputs, to mitigate memory issues. Current implementation is built on top of MPI and uses CPUs to maximize the portability across the platforms.

Distributed computing, machine learning↗
DOE OSTI2022

Dynamic Decarbonization through Autonomous Physics-Centric Deep Learning and Optimization of Building Operations (Abstract only)

This project directly addresses the primary goal of Area of Interest 2 in the CRADA call: to advance optimization-based integrated energy management systems in commercial and residential buildings. Pacific Northwest National Laboratory (PNNL) and its industry partner PassiveLogic aim to accomplish this by reaching three key objectives. First, to ensure a broad impact in the building controls industry, PNNL will extend its open-source library for predictive control synthesis by augmenting its capabilities with data-driven self-learning of building models and auto-calibration of predictive controllers. The effort will focus on building use cases selected in collaboration with PassiveLogic. The team will specifically address the development of methods for data-driven adaptation of building models, investigation of model architectures that best address specific building types, and automated synthesis of differentiable predictive controllers that optimize diverse objectives. Second, PNNL will collaborate with PassiveLogic to integrate the aforementioned methods with PasiveLogic’s advanced controls platform. The collaborative integration effort will inform the developments under the first objective by providing specific data on the attainable performance of model learning on resource-constrained edge computing platforms. This software integration effort will increase the technical maturity of the developed libraries by exploring the use of software integration tools and methods. Third, PNNL and PassiveLogic will work to improve the technology readiness of the developed predictive controllers by testing their performance in relevant test environments, such as high-fidelity simulation, hardware in the loop, and actual test buildings.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗