SEARCH · Engineering Papers

Results for “input estimation”

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 1,405 records · Page 78

All sources NASA NTRS DOE OSTI

DOE OSTI2024

Photometric redshifts probability density estimation from recurrent neural networks in the DECam local volume exploration survey data release 2

Photometric wide-field surveys are imaging the sky in unprecedented detail. These surveys face a significant challenge in efficiently estimating galactic photometric redshifts while accurately quantifying associated uncertainties. In this work, we address this challenge by exploring the estimation of Probability Density Functions (PDFs) for the photometric redshifts of galaxies across a vast area of 17,000 square degrees, encompassing objects with a median 5 σ point-source depth of g = 24.3, r = 23 . 9 , i = 23.5, and z = 22.8 mag. Our approach uses deep learning, specifically integrating a Recurrent Neural Network architecture with a Mixture Density Network, to leverage magnitudes and colors as input features for constructing photometric redshift PDFs across the whole DECam Local Volume Exploration (DELVE) survey sky footprint. Subsequently, we rigorously evaluate the reliability and robustness of our estimation methodology, gauging its performance against other well-established machine learning methods to ensure the quality of our redshift estimations. Our best results constrain photometric redshifts with the bias of − 0 . 0013 , a scatter of 0.0293, and an outlier fraction of 5.1%. These point estimates are accompanied by well-calibrated PDFs evaluated using diagnostic tools such as Probability Integral Transform and Odds distribution. We also address the problem of the accessibility of PDFs in terms of disk space storage and the time demand required to generate their corresponding parameters.We present a novel Autoencoder model that reduces the size of PDF parameter arrays to one-sixth of their original length, significantly decreasing the time required for PDF generation to one-eighth of the time needed when generating PDFs directly from the magnitudes.

79 ASTRONOMY AND ASTROPHYSICS↗

NASA NTRS1990

Global surface radiation flux results for all seasons using ISCCP-C1 data

Algorithms have been developed at the Langley Research Center to estimate fluxes for all of the surface radiation components utilizing the new International Satellite Cloud Climatology Project (ISSCP) C1 data and grid system. These surface radiation budget data are the first all-component flux set encompassing the entire globe, based on satellite-derived C1 meteorological data. A description of the input data and the algorithms utilized, examples of the global flux results, discussion of how the current fluxes compare with ground-truth fluxes, and remarks on errors in the current flux estimates are presented.

Darnell, Wayne L.↗

NASA NTRS1972

Hill-climbing controller for plants with any dynamics and rapid drifts.

Description of extremal control systems which utilize extrapolation procedures to optimize plant operations or processes involving any kind of dynamics before and after the static extremal characteristic. The optimization sequence starts with an 'action of drift detection' - i.e., a measurement of the drift rate of the output. Then a 'search action' is initiated which consists of stepped shifts of the input variable; the output response due to these changes in the input variable is evaluated in the light of the predicted shift and is used to provide information on the shape of the extremal characteristic at the actual operating point. The position of the extremum is estimated by a parabolic extrapolation of the shape of the characteristic from the operating point to the extremum. During the 'control action' the operating point is adjusted to the position corresponding to the predicted optimum of the process by first applying a forcing step to the input variable to obtain a quicker response at the beginning of the transient state, and then applying a definitive step to bring the input variable back to the position which corresponds to the predicted extremum. A 'rest time' may be introduced before starting the next optimization sequence.

Jacob, H. G.↗

NASA NTRS2007

Simulation Studies of the Effect of Forest Spatial Structure on InSAR Signature

The height of scattering phase retrieved from InSAR data is considered being correlated with the tree height and the spatial structure of the forest stand. Though some researchers have used simple backscattering models to estimate tree height from the height of scattering center, the effect of forest spatial structure on InSAR data is not well understood yet. A three-dimensional coherent radar backscattering model for forest canopies based on realistic three-dimensional scene was used to investigate the effect in this paper. The realistic spatial structure of forest canopies was established either by field measurements (stem map) or through use of forest growth model. Field measurements or a forest growth model parameterized using local environmental parameters provides information of forest species composition and tree sizes in certain growth phases. A fractal tree model (L-system) was used to simulate individual 3- D tree structure of different ages or heights. Trees were positioned in a stand in certain patterns resulting in a 3-D medium of discrete scatterers. The radar coherent backscatter model took the 3-D forest scene as input and simulates the coherent radar backscattering signature. Interferometric SAR images of 3D scenes were simulated and heights of scattering phase centers were estimated from the simulated InSAR data. The effects of tree height, crown cover, crown depth, and the spatial distribution patterns of trees on the scattering phase center were analyzed. The results will be presented in the paper.

Sun, Guoqing↗

DOE OSTI2025

wa-hls4ml and lui-gnn: A benchmark and GNN-based surrogate model for hls4ml resource and latency estimation

As machine learning (ML) increasingly serves as a tool for addressing real-time challenges in scientific applications, the development of advanced tooling has significantly reduced the time required to iterate on various designs. These advancements have solved major obstacles, but also exposed new challenges. For example, processes that were not previously considered bottlenecks, such as model synthesis, are now becoming limiting factors in the rapid iteration of designs. To reduce these emerging constraints, multiple efforts are being launched toward designing an ML-based surrogate model that estimates resource usage of synthesized accelerator architectures. This model would reduce the design iteration time, especially when designing within a set of given hardware constraints. This approach shows considerable potential, but as it stands, the effort is early and would benefit from coordination and standardization to assist future work as it emerges. We introduce wa-hls4ml, a benchmark for ML accelerator resource and latency estimation, and its corresponding initial dataset of more than 100,000 fully connected neural networks, all synthesized using hls4ml and targeting Xilinx FPGAs. In addition to the resource utilization and latency data provided, the dataset includes generated artifacts and log files for many of the synthesized neural networks, in order to support future research in ML-based code generation. The benchmark evaluates the performance of resource and latency predictors against several common ML model architectures, primarily originating from scientific domains, as exemplar models, as well as the average performance across a subset of the dataset. We measure the performance of a given predictor model through multiple metrics, including $R^2$ score and SMAPE on regression tasks, as well as inference time to further characterize the estimator under test. Additionally, we introduce the latency/utilization inference graph neural network (lui-gnn), a surrogate model that uses a graph neural network to represent input architectures in the form of a directed graph. This graph representation allows for a diverse set of model architectures to all be effectively handled by a surrogate model. We present the architecture and performance of the model, as evaluated by the new proposed benchmark, including SMAPE, $R^2$ score, and inference times, and find that lui-gnn generally predicts latency and utilization for the 75\% quantile within several percent of the synthesized resources on the synthetic test dataset, indicating that this approach of estimating resource and latency via a surrogate models has promise and warrants further research.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

NASA NTRS2018

Estimation of Terrestrial Global Gross Primary Production (GPP) with Satellite Data-Driven Models and Eddy Covariance Flux Data

We estimate global terrestrial gross primary production (GPP) based on models that use satellite data within a simplified light-use efficiency framework that does not rely upon other meteorological inputs. Satellite-based geometry-adjusted reflectances are from the MODerate-resolution Imaging Spectroradiometer (MODIS) and provide information about vegetation structure and chlorophyll content at both high temporal (daily to monthly) and spatial (1 km) resolution. We use satellite-derived solar-induced fluorescence (SIF) to identify regions of high productivity crops and also evaluate the use of downscaled SIF to estimate GPP. We calibrate a set of our satellite-based models with GPP estimates from a subset of distributed eddy covariance flux towers (FLUXNET 2015). The results of the trained models are evaluated using an independent subset of FLUXNET 2015 GPP data. We show that variations in light-use efficiency (LUE) with incident PAR are important and can be easily incorporated into the models. Unlike many LUE-based models, our satellite-based GPP estimates do not use an explicit parameterization of LUE that reduces its value from the potential maximum under limiting conditions such as temperature and water stress. Even without the parameterized downward regulation, our simplified models are shown to perform as well as or better than state-of-the-art satellite data-driven products that incorporate such parameterizations. A significant fraction of both spatial and temporal variability in GPP across plant functional types can be accounted for using our satellite-based models. Our results provide an annual GPP value of 140 Pg C year 1 for 2007 that is within the range of a compilation of observation-based, model, and hybrid results, but is higher than some previous satellite observation-based estimates

CO2↗

NASA NTRS2006

Thermal Band Atmospheric Correction Using Atmospheric Profiles Derived from Global Positioning System Radio Occultation and the Atmospheric Infrared Sounder

This Rapid Prototyping Capability study explores the potential to use atmospheric profiles derived from GPS (Global Positioning System) radio occultation measurements and by AIRS (Atmospheric Infrared Sounder) onboard the Aqua satellite to improve surface temperature retrieval from remotely sensed thermal imagery. This study demonstrates an example of a cross-cutting decision support technology whereby NASA data or models are shown to improve a wide number of observation systems or models. The ability to use one data source to improve others will be critical to the GEOSS (Global Earth Observation System of Systems) where a large number of potentially useful systems will require auxiliary datasets as input for decision support. Atmospheric correction of thermal imagery decouples TOA radiance and separates surface emission from atmospheric emission and absorption. Surface temperature can then be estimated from the surface emission with knowledge of its emissivity. Traditionally, radiosonde sounders or atmospheric models based on radiosonde sounders, such as the NOAA (National Oceanic & Atmospheric Administration) ARL (Air Resources Laboratory) READY (Real-time Environmental Application and Display sYstem), provide the atmospheric profiles required to perform atmospheric correction. Unfortunately, these types of data are too spatially sparse and too infrequently taken. The advent of high accuracy, global coverage, atmospheric data using GPS radio occultation and AIRS may provide a new avenue for filling data input gaps. In this study, AIRS and GPS radio occultation derived atmospheric profiles from the German Aerospace Center CHAMP (CHAllenging Minisatellite Payload), the Argentinean Commission on Space Activities SAC-C (Satellite de Aplicaciones Cientificas-C), and the pair of NASA GRACE (Gravity Recovery and Climate Experiment) satellites are used as input data in atmospheric radiative transport modeling based on the MODTRAN (MODerate resolution atmospheric TRANsmittance) radiative transport software to separate out the atmospheric component of measured top of atmosphere radiance. Simulated water bodies across a variety of MODTRAN model atmospheres including desert, mid-latitude, tropical and sub-artic conditions provide test bed conditions. Atmospherically corrected radiance and surface temperature results were compared to those obtained using traditional radiosonde balloon data and models. In general, differences between the different techniques were less than 2 percent indicating the potential value satellite derived atmospheric profiles have to atmospherically correct thermal imagery.

Pagnutti, Mary↗

DOE OSTI2022

Hydrologic investigations of radar-rainfall error propagation to rainfall-runoff model hydrographs

Rainfall is arguably the most important yet most variable input for rainfall-runoff hydrologic models. In this study, the authors search for the characteristics of radar-rainfall estimates that are most important for skillful streamflow predictions. They perform comprehensive hydrologic investigations of radar-rainfall characteristics, including spatiotemporal resolution, radar range visibility, statistical characterization of rainfall variability, all vis-a-vis basin characteristics such as size and river network topology. Since the true rainfall fields are unknown, the authors exploit a paradigm of using two independently constructed radar-rainfall products i.e., Multi-Radar Multi-Sensor and IFC-ZR used operationally by the Iowa Flood Center (IFC). Using the distributed hydrologic model called the Hillslope-Link Model for the domain of the state of Iowa, they evaluate streamflow prediction at 140 USGS gauge stations that monitor rivers in Iowa. Through spatial and temporal rainfall aggregation experiments, the authors show that the impact of spatial and temporal resolution of rainfall is significant typically for smaller basins while starts reducing significantly for basins larger than 1,000 km 2 . Other rainfall characteristics they explored do not reveal a strong signature in the relationship of rainfall differences between the two products and hydrograph errors. However, exploring the product similarities rather than differences reveals that the basin-wide rainfall volume has the most significant effect on streamflow prediction. The results from this study are generalizable for all rainfall observing systems.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2020

An encoder–decoder LSTM-based EMPC framework applied to a building HVAC system

Numerous studies have demonstrated the benefit of economic model predictive control (EMPC) applied to building heating, ventilation, and air conditioning (HVAC) systems. However, the construction and training of predictive models for building HVAC systems are widely recognized as a key technological barrier preventing large-scale adoption of EMPC for buildings. In this work, an encoder–decoder long short-term memory-based EMPC framework is developed. The key advantage of the approach is that a model may be automatically generated from a list of inputs and outputs. From the definition of inputs and outputs, the constructed model may be trained and automatically embedded into the EMPC framework for real-time estimation and control. The overall end-to-end EMPC framework from model training to on-line estimation and control are described. To this end, the encoder–decoder model provides a natural framework for state estimation (encoder), which is required to provide an initial condition for the predictive model of EMPC (decoder). Closed-loop simulations using EnergyPlus are performed to demonstrate the approach. The simulated closed-loop system consists of a building zone from a multi-zone building, which is served by an air handling unit-variable air volume HVAC system. For the HVAC example considered, the trained encoder–decoder model can predict the indoor air temperature and HVAC sensible cooling rate of a building zone over a two-day horizon with high accuracy. Overall, we find that considering a time-of-use electric rate structure, the EMPC, which manipulates the zone temperature setpoint, can reduce the HVAC power consumption cost relative to keeping the zone temperature setpoint at its maximum value (i.e., minimum energy approach).