SEARCH · Engineering Papers

Results for “validation dataset”

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 523 records · Page 29

All sources NASA NTRS DOE OSTI

DOE OSTI2021

A large dataset of detection and submeter-accurate 3-D trajectories of juvenile Chinook salmon

Abstract Acoustic telemetry has been used extensively to study the behavior of aquatic animals. The Juvenile Salmon Acoustic Telemetry System (JSATS) is one such system; it was developed for studying juvenile salmonids but has been used to study numerous species. A recent innovation of the JSATS system is an acoustic transmitter that is small enough to be implanted through injection or small incision that doesn’t require sutures. Use of the JSATS system involves deploying cabled acoustic receivers at hydroelectric dams, or other structures, and autonomous acoustic receivers in free-flowing sections of a river. The raw detections from acoustic-tagged fish are processed to remove potential false positives. The clean detections (5,147,996 total) are used to generate detection events and to compute 3-D trajectories (403,900 total), which are used to assign fish to a passage route through a dam. Controlled field testing involving a high-accuracy Global Positioning System receiver is done to validate the submeter accuracy of the trajectories. The JSATS dataset could be reused for expanding the understanding of near-dam fish behavior.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2023

Discovery of multi-functional polyimides through high-throughput screening using explainable machine learning

Polyimides have been widely used in modern industries because of their excellent mechanical and thermal properties, e.g., high-temperature fuel cells, displays, and aerospace composites. However, it usually takes decades of experimental efforts to develop a successful product. Aiming to expedite the discovery of high-performance polyimides, we utilize computational methods of machine learning (ML) and molecular dynamics (MD) simulations. Our study provides compelling evidence for the effectiveness of a data-driven approach in discovering novel polyimides. We first build a comprehensive library of more than 8 million hypothetical polyimides based on the polycondensation of existing dianhydride and diamine/diisocyanate molecules. Then we establish multiple ML models for the thermal and mechanical properties of polyimides based on their experimentally reported values, including glass transition temperature, Young’s modulus, and tensile yield strength. The obtained ML models demonstrate excellent predictive performance in identifying the key chemical substructures influencing the thermal and mechanical properties of polyimides. The use of explainable machine learning describes the effect of chemical substructures on individual properties, from which human experts can understand the cause of the ML model decision. Applying the well-trained ML models, we obtain property predictions of the 8 million hypothetical polyimides. Then, we screen the whole hypothetical dataset and identify three (3) best-performing novel polyimides that have better-combined properties than existing ones through Pareto frontier analysis. For an easy query of the discovered high-performing polyimides, we also create an online platform https://polyimide-explorer.herokuapp.com/ that embeds the developed ML model with interactive visualization. Furthermore, we validate the ML predictions through all-atom MD simulations and examine their synthesizability. The MD simulations are in good agreement with the ML predictions and the three novel polyimides are predicted to be easy to synthesize via Schuffenhauer’s synthetic accessibility score. Following the proposed ML guidance, we successfully synthesized a novel polyimide and the experimentally obtained high glass transition/thermal decomposition temperature demonstrated its excellent thermal stability. Here our study demonstrates an efficient way to expedite the discovery of novel polymers using ML prediction and MD validation. The high-throughput screening of a large computational dataset can serve as a general approach for new material discovery in other polymeric material exploration problems, such as organic photovoltaics, polymer membranes, and dielectrics.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

NASA NTRS2020

Spatial validation reveals poor predictive performance of large-scale ecological mapping models

Mapping aboveground forest biomass is central for assessing the global carbon balance. However, current large-scale maps show strong disparities, despite good validation statistics of their underlying models. Here, we attribute this contradiction to a flaw in the validation methods, which ignore spatial autocorrelation (SAC) in data, leading to overoptimistic assessment of model predictive power. To illustrate this issue, we reproduce the approach of large-scale mapping studies using a massive forest inventory dataset of 11.8 million trees in central Africa to train and validate a random forest model based on multispectral and environmental variables. A standard nonspatial validation method suggests that the model predicts more than half of the forest biomass variation, while spatial validation methods accounting for SAC reveal quasi-null predictive power. This study underscores how a common practice in big data mapping studies shows an apparent high predictive power, even when predictors have poor relationships with the ecological variable of interest, thus possibly leading to erroneous maps and interpretations.

biomass↗

DOE OSTI2026

WINDPROF: Merged Best-Estimate Wind Profile Data – Nantucket (WFIP3 Campaign)

WINDPROF provides 10-minute wind and turbulence profiles, integrating Doppler lidars, wind profiling radars, and sonic anemometers across Northeast U.S. coastal/offshore sites during the WFIP3 campaign. Key data include wind speed, direction, vertical velocity, and turbulence parameters, with standardized quality control (e.g., instrument-specific thresholds and inter-instrument validation). Profiles are interpolated to a height grid (20 m spacing below 100 m; 30 m above) and include comprehensive uncertainty estimates. The Nantucket dataset covers February 2024–September 2025, offering reproducible methods for atmospheric research, model validation, and wind energy studies.

17 WIND ENERGY↗

DOE OSTI2025

A dataset for understanding self-reported patterns influencing residential energy decisions

Household occupant behavior and decision-making dynamics substantially impact technology uptake and residential building energy performance. Although significant research underscores the importance of social science in energy studies, few public data with representative samples on household energy decision-making patterns are available. The dataset (UPGRADE-E: Understanding Patterns Guiding Residential Adoption and Decisions about Energy Efficiency) presents 9,919 responses from U.S. residents of single-family and small multifamily homes. Derived from a national-scale internet survey, the dataset contains 391 variables: demographics, building characteristics, home modifications, willingness to adopt new technologies, motivations for making changes, barriers, program participation, trusted information sources, and energy scenarios. Responses were validated via internal consistency checks and comparison with other U.S. national scale datasets. UPGRADE-E advances knowledge of household energy related decision-making, tying demographics, home modifications, and self-reported cognitive drivers together at a scale and breadth that has not been previously achieved. Policymakers and researchers at local, regional, and national levels may leverage this dataset to understand drivers influencing the adoption of key technologies in U.S. homes.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2026

WINDPROF: Merged Best-Estimate Wind Profile Data – Block Island (WFIP3 Campaign)

WINDPROF provides 10-minute wind and turbulence profiles, integrating Doppler lidars, wind profiling radars, and sonic anemometers across Northeast U.S. coastal/offshore sites during the WFIP3 campaign. Key data include wind speed, direction, vertical velocity, and turbulence parameters, with standardized quality control (e.g., instrument-specific thresholds and inter-instrument validation). Profiles are interpolated to a height grid (20 m spacing below 100 m; 30 m above) and include comprehensive uncertainty estimates. The Block Island dataset covers February 2024–September 2025, offering reproducible methods for atmospheric research, model validation, and wind energy studies.

17 WIND ENERGY↗

DOE OSTI2026

Toward an AI-Powered Software Pipeline for Real-Time Tracking and Analysis of Wildfire and Smoke

Real-time tracking of wildfires and smoke is crucial for effective response, minimizing damage, protecting lives, and efficiently managing resources during fire emergencies. We develop a web-based AI-powered pipeline that detects wildfires in aerial video and estimates deployment-relevant behavior metrics, including cumulative burned area, burned-area growth rate, fire spread direction, and smoke dispersion. The system combines a YOLO-based detector with YCbCr-based fire segmentation, HSV-based smoke segmentation, Farneback optical flow, and centroid-based spatiotemporal tracking. Using ground sampling distance (GSD), pixel-level fire masks are converted to physical burned-area measurements by correlating fire pixel counts with camera altitude and tilt angle. We benchmark YOLO variants and non-YOLO baselines (GoogLeNet, CNN, DBN, Autoencoder, U-Net, and AlexNet) on the IEEE FLAME dataset and a newly created aerial frame dataset, Wildfire-DB. Cross-dataset evaluation uses a strict threshold-transfer protocol: decision thresholds are selected on FLAME validation and transferred unchanged to Wildfire-DB to quantify generalization under domain shift. YOLOv6 achieves the strongest cross-dataset frame-level fire detection on Wildfire-DB (ROC-AUC 0.8200, PR-AUC 0.8044, and transferred-threshold F1 0.7596). For tracking-oriented deployment requiring oriented localization, YOLO11-OBB provides the most reliable cross-dataset behavior among OBB-capable models while remaining computationally feasible. To analyze the feasibility of UAV deployment, we further measure inference efficiency using synchronized GPU and CPU power logs on a fixed workload of 1569 frames. YOLO-family models process the video in 5.73–12.47 seconds with net energy of 1247.28–1775.39 J, substantially lower latency and energy than heavier classification and reconstruction baselines. Overall, model optimality depends on operational objectives: YOLOv6 is best for cross-dataset detection robustness, whereas YOL...

Color segmentation↗

NASA NTRS2018

Simulating Forest Dynamics of Lowland Rainforests in Eastern Madagascar

Ecological modeling and forecasting are essential tools for the understanding of complex vegetation dynamics. The parametric demands of some of these models are often lacking or scant for threatened ecosystems, particularly in diverse tropical ecosystems. One such ecosystem and also one of the world's biodiversity hotspots, Madagascar's lowland rainforests, have disappeared at an alarming rate. The processes that drive tree species growth and distribution remain as poorly understood as the species themselves. We investigated the application of the process-based individual-based FORMIND model to successfully simulate a Madagascar lowland rainforest using previously collected multi-year forest inventory plot data. We inspected the model's ability to characterize growth and species abundance distributions over the study site, and then validated the model with an independently collected forest-inventory dataset from another lowland rainforest in eastern Madagascar. Following a comparative analysis using inventory data from the two study sites, we found that FORMIND accurately captures the structure and biomass of the study forest, with r(squared) values of 0.976, 0.895, and 0.995 for 1:1 lines comparing observed and simulated values across all plant functional types for aboveground biomass (tonnes/ha), stem numbers, and basal area (m(squared)/ha), respectively. Further, in validation with a second study forest site, FORMIND also compared well, only slightly over-estimating shade-intermediate species as compared to the study site, and slightly under-representing shade-tolerant species in percentage of total aboveground biomass. As an important application of the FORMIND model, we measured the net ecosystem exchange (NEE, in tons of carbon per hectare per year) for 50 ha of simulated forest over a 1000-year run from bare ground. We found that NEE values ranged between 1 and -1 t Cha(exp -1)year(exp -1), consequently the study forest can be considered as a net neutral or a very slight carbon sink ecosystem, after the initial 130 years of growth. Our study found that FORMIND represents a valuable tool toward simulating forest dynamics in the immensely diverse Madagascar rainforests.

Forest Productivity↗

NASA NTRS2022

Colorado Eastern Plains Agriculture: Rangeland Monitoring to Inform Grazing Management in Eastern Colorado

Adaptive management on cattle ranches requires rangeland managers to decide the location and duration of the cattle grazing activity across different pastures. Biodiversity, forage availability, and cattle health are all affected by rangeland management. Virtual fencing is a tool that rangeland managers can use to potentially increase biodiversity and improve ranching operations. NASA DEVELOP and Colorado State University (CSU) collaborated with the Nature Conservancy (TNC), and Red Top Ranch to demonstrate the efficacy of virtual fencing. We sought to identify annual and monthly biomass patterns on the ranch through the creation of monthly max biomass productivity maps. We utilized a dataset from the Agricultural Research Service (ARS) to calculate biomass on the ranch. To validate our remotely-sensed results, we compared model-predicted biomass values to field-collected biomass clipping data and an additional biomass dataset from the Rangeland Analysis Platform (RAP). We used satellite imagery from Landsat 5 Thematic Mapper (TM), Landsat 7 Enhanced Thematic Mapper Plus (ETM+), Landsat 8 Operational Land Imager (OLI), Landsat 9 –OLI-2, and Sentinel-2 MultiSpectral Instrument (MSI) for 2021 and 2022. We found that there was heterogeneity in biomass across the ranch, with higher biomass on the western side. The highest peak of biomass was in the summer months, with a smaller peak in mid-September. The ARS biomass dataset had a significant relationship with RAP for 2021. ARS biomass did not have a significant relationship with the biomass field data collected in 2022. The results of our study are aimed to support rotation management, ranch production, biodiversity, and conservation management.

cattle↗

NASA NTRS2020

High Spectral Resolution Datasets of in Situ Aquatic Inherent and Apparent Optical Properties

Light emerging from natural water bodies and measured by remote sensing radiometers contains information about the local type and concentrations of phytoplankton, non-algal particles and colored dissolved organic matter in the underlying waters. An increase in spectral resolution in forthcoming satellite and airborne remote sensing missions is expected to lead to new or improved capabilities to characterize aquatic ecosystems. Such upcoming missions include NASA’s Plankton, Aerosol, Cloud, ocean Ecosystem PACE Mission; the NASA Surface Biology and Geology observable mission; and NASA Airborne Visible / Infrared Imaging Spectrometer (AVIRIS) - Next Generation airborne missions. In anticipation of these missions, we present a dataset of geographically diverse, quality-controlled, high spectral resolution inherent and apparent optical property (IOP/AOP) aquatic data. The data are intended to be of use to increase our understanding of aquatic optical properties, to develop aquatic remote sensing data product algorithms, and to perform calibration and validation activities for forthcoming aquatic-focused imaging spectrometry missions. The dataset is comprised of contributions from several investigators and investigating teams collected over a range of geographic areas and water types, including inland water, estuaries and oceans. Specific in situ measurements include particulate absorption, particulate attenuation, non-algal particulate or detritus absorption, colored dissolved organic matter absorption, phytoplankton absorption, total absorption, total attenuation, particulate backscattering, total backscatter, remote sensing reflectance, and irradiance reflectance. The dataset can be downloaded from PANGAEA: (https://www.pangaea.de/TBD ).

K A Casey↗

DOE OSTI2026

A Centralized AI Lakehouse Framework for Brain Tumor MRI Classification and Segmentation, University KPI Forecasting, and Water Potability Prediction

In many university and healthcare projects, models are built for very different data types such as tables, institutional time series, and medical images, but they are deployed as separate applications. In this work, that separation made testing and maintenance difficult because each module had its own pipeline and runtime requirements. This paper presents an integrated AI lakehouse-style implementation that runs three model pipelines inside one containerized backend. For medical imaging, we used MRI datasets from IEEE DataPort: a four-class classification set with 7012 images (5708 train/1304 test) and a segmentation set with 3063 image–mask pairs. The classification model (ResNet50 transfer learning) is evaluated using a proper train–validation–test protocol across multiple splits (80/10/10, 70/10/20, 60/10/30, and 10/30/60), achieving a test accuracy of 99.00% under the standard 80/10/10 split. Additionally, a patient-level evaluation is conducted using an external glioma dataset to provide a more realistic assessment without data leakage. The segmentation model (DeepLabV3-ResNet50) achieved 83.09% validation mIoU and 88.79% Dice score. For university KPI forecasting, we used annual IPEDS and NSF HERD data from 2010 to 2023 for three universities (BSU, EOU, and UAB). To examine the effect of preprocessing on forecasting performance, two case studies are conducted. In the first case, linear interpolation is applied to generate semester-level data. In the second case, the original annual data is used directly without interpolation. Random Forest regression and ARIMA models are evaluated using MAE, RMSE, MAPE, and R 2 . The results showed that interpolation improved apparent forecasting performance due to smoothing, while evaluation on the original annual data provided a more realistic assessment of model behavior. To further validate the framework on a larger dataset, an additional case study is conducted using a student dropout dataset. For water potability, we trained and compared multiple tabular classifiers on a large dataset (1,048,575 samples). A Random Forest model (100 trees, max depth 10) achieved 85.86% test accuracy and high recall for unsafe samples (0.8447). All modules are served via FastAPI and deployed together using Docker, with workflow automation routing requests to the correct endpoint. System-level benchmarking indicates that the backend maintains stable throughput and latency under concurrent requests.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2018

Improved Hydrology over Peatlands in a Global Land Modeling System

Peatlands of the Northern Hemisphere represent an important carbon pool that mainly accumulated since the last ice age under permanently wet conditions in specific geological and climatic settings. The carbon balance of peatlands is closely coupled to water table dynamics. Consequently, the future carbon balance over peatlands is strongly dependent on how hydrology in peatlands will react to changing boundary conditions, e.g. due to climate change or regional water level drawdown of connected aquifers or streams. Global land surface modeling over organic-rich regions can provide valuable global-scale insights on where and how peatlands are in transition due to changing boundary conditions. However, the current global land surface models are not able to reproduce typical hydrological dynamics in peatlands well. We implemented specific structural and parametric changes to account for key hydrological characteristics of peatlands into NASA's GEOS-5 Catchment Land Surface Model (CLSM, Koster et al. 2000). The main modifications pertain to the modeling of partial inundation, and the definition of peatland-specific runoff and evapotranspiration schemes. We ran a set of simulations on a high performance cluster using different CLSM configurations and validated the results with a newly compiled global in-situ dataset of water table depths in peatlands. The results demonstrate that an update of soil hydraulic properties for peat soils alone does not improve the performance of CLSM over peatlands. However, structural model changes for peatlands are able to improve the skill metrics for water table depth. The validation results for the water table depth indicate a reduction of the bias from 2.5 to 0.2 m, and an improvement of the temporal correlation coefficient from 0.5 to 0.65, and from 0.4 to 0.55 for the anomalies. Our validation data set includes both bogs (rain-fed) and fens (ground and/or surface water influence) and reveals that the metrics improved less for fens. In addition, a comparison of evapotranspiration and soil moisture estimates over peatlands will be presented, albeit only with limited ground-based validation data. We will discuss strengths and weaknesses of the new model by focusing on time series of specific validation sites.

Bechtold, M.↗

DOE OSTI2023

Insights of Boundary Layer Turbulence Over the Complex Terrain of Central Himalaya from GVAX Field Campaign

Limited observations hinder understanding of turbulent characteristics in mountainous terrain resulting from heating or cooling of slopes, wind, vertical motions, and heat or moisture advection, which disperse aerosols and other pollutants over the region. In this study, the 1290 MHz radar wind profiler data are utilized to compute the boundary layer height (BLH), the refractive index structure constant (C n 2 ), and the energy dissipation rate (ε) over the central Himalayan site for the period of November 2011 to March 2012, from the intense Ganges Valley Aerosol Experiment (GVAX) field measurements. The radar wind profiler (RWP) based estimation of BLH and ε is validated against the radiosonde, representing the effectiveness of the datasets for further investigation. The strong seasonal variation of log C n 2 and log ε, with average values of ≈ -12 m -2/3 and -2 m 2 s -3 , respectively, is associated with the mountain-induced local circulations and stability in the atmospheric boundary layer. The weak stratification during weak flow is found to be responsible for deep mixing, particularly in the nocturnal boundary layer in spring. Furthermore, the level of cloud cover significantly impacts the strength of turbulence, with the highest cloud cover resulting in a substantial increase in log C n 2 (approximately -11 m -2/3 ) due to intense updraft and downdraft motions compared to clear skies. Additionally, the distribution of aerosol loading across the site, coupled with the behavior of BLH, atmospheric stability, and orographic-induced circulations, implies distinctive seasonal mechanisms for transporting aerosols toward the mountains. This study offers valuable insights into the diurnal and seasonal patterns of turbulent mixing and the mechanisms behind the transport of pollutants through boundary layer processes over the region.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2023

The interactive indoor-outdoor building energy modeling for enhancing the predictions of urban microclimates and building energy demands

There is a lack of an urban building energy modeling framework that considers the influence of surrounding buildings and local urban climate on building thermal performance. This can lead to inaccurate results since the thermal performance of individual buildings is heavily influenced by their surrounding built and climatic environment. This study establishes an interactive indoor-outdoor building energy modeling method to enhance the predictions of urban microclimates and building energy demands by coupling an urban physics model with a physics-based building energy model. Validation of the interactive coupling scheme uses field measurement datasets. Parametric simulation and analysis are conducted to understand the influence of the roof-to-canyon width ratio, canyon orientation, and ground vegetation fraction on canyon temperature, building energy consumption, and energy demand. Furthermore, the impacts of building energy model complexity (e.g., detailed vs. simplified building models) and coupling approaches on canyon temperature and building energy profiles are demonstrated using two case study buildings. In comparison with the one-way coupling approach, cooling energy consumption predicted with the dynamic two-way coupling approach varies by 3.5% and 0.5% for the detailed medium office building model and high-rise building model, respectively, and peak cooling demand varies by 8.4% and 7.0% for the detailed medium office building model and high-rise building model, respectively. Here this study also suggests that adopting a complex two-way coupling approach with environmental data exchange at various elevations is necessary for modeling tall buildings at the urban scale.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2024

Evaluating direct vessel injection accident-event progression of AP1000 and key figures of merit to support the design and development of water-cooled small modular reactors

The passive safety systems (PSSs) within water-cooled reactors are meticulously engineered to function autonomously, requiring no external power source or manual intervention. They depend exclusively on inherent natural forces and the fundamental principles of reactor physics, such as gravity, natural convection, and phase changes, to manage, alleviate, and avert the release of radioactive materials into the environment during accident scenarios like a loss-of-coolant accident (LOCA). PSSs are already integrated into such operating commercial reactors as the Advanced Pressurized Reactor-1000 MWe (AP1000) and the Water-Water Energetic Reactor-1200 MWe (WWER-1200) are adopted in most of the upcoming small modular reactor (SMR) designs. Examples of water-cooled SMR PSSs are the passive emergency core-cooling system (ECCS), passive containment cooling system (PCCS), and passive decay-heat removal system, the designs of which vary based on reactor system-design requirements. However, understanding the accident-event progression and phases of a LOCA is pivotal for adopting a specific PSS for a new SMR design. This study covers the accident-event progression for direct vessel injection (DVI) small-break loss-of-coolant accident (SB-LOCA), associated physics phenomena, knowledge gaps, and important figures of merit (FOMs) that may need to be evaluated and assessed to validate thermal-hydraulics models with an available experimental dataset to support new SMR design and development.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

DOE OSTI2026

Synthetic data-driven deep learning for label-free autonomous atomic force microscopy

Atomic force microscopy (AFM) is a widely used tool for nanoscale characterization across materials science, energy research, and biology. However, its adoption in high-throughput materials discovery and statistically driven studies remains limited by a strong dependence on expert operator input and by the scarcity of annotated experimental AFM datasets needed to enable data-driven automation. Here, we introduce SimuScan, a synthetic-data–driven framework that enables reliable AFM feature identification, segmentation, and targeted imaging without requiring large manually labeled experimental datasets. SimuScan generates tunable, high-fidelity synthetic AFM images of defined morphologies while incorporating realistic experimental artifacts, including tip–sample convolution, noise, flattening distortions, and surface debris. These datasets are shown to support scalable, label-free training of modern deep learning models for AFM analysis. When integrated into data-driven AFM workflows, SimuScan-trained models can locate and analyze nanoscale structures across large datasets and guide targeted follow-up imaging. We validate this approach on nanostructured surfaces, DNA assemblies, and bacterial cells, demonstrating robust generalization across diverse sample types with minimal operator intervention. More broadly, this work establishes a general strategy for generating explicitly conditioned, task-relevant synthetic data to improve the reliability of downstream models in autonomous microscopy.

Millan-Solsona, Ruben [Oak Ridge National Laborato↗

DOE OSTI2025

Counterpart identification and classification for eRASS1 and characterisation of the active galactic nuclei content

Context. Accurately accounting for the Active Galactic Nucleus (AGN) phase in galaxy evolution requires a large, clean AGN sample. This is now possible with SRG/eROSITA, which completed its first all-sky X-ray survey (eRASS1) on June 12, 2020. The public Data Release 1 (DR1, Jan 31, 2024) includes 930,203 sources from the western Galactic hemisphere. Aims. The data enable the selection of a large AGN sample and the discovery of rare sources. However, scientific return depends on accurate characterisation of the X-ray emitters, requiring high-quality multi-wavelength data. This paper presents the identification and classification of optical and infrared counterparts to eRASS1 sources. Methods. Counterparts to eRASS1 X-ray point sources were identified using Gaia DR3, CatWISE2020, and Legacy Survey DR10 (LS10) with the Bayesian NWAY algorithm and trained priors. Sources were classified as Galactic or extragalactic via a machine-learning model combining optical/IR and X-ray properties, trained on a reference sample. For extragalactic LS10 sources, photometric redshifts were computed using CIRCLEZ. Results. Within the LS10 footprint, all 656,614 eROSITA/DR1 sources have at least one possible optical counterpart; ∼570 000 are extragalactic and likely AGN. Half are new detections compared to AllWISE, Gaia, and Quaia AGN catalogues. Gaia and CatWISE2020 counterparts are less reliable, due to the survey’s shallowness and the limited amount of features available to assess the probability of being an X-ray emitter. In the Galactic plane, where the overdensity of stellar sources also increases the chance of associations, using conservative reliability cuts, we identified approximately 18 000 Gaia and 55 000 CatWISE2020 extragalactic sources. Conclusions. We have released three high-quality counterpart catalogues – plus the training and validation sets – as a benchmark for the field. These datasets have many applications, but in particular, they empower researchers to build AGN samples tailored for completeness and purity, accelerating the hunt for the Universe’s most energetic engines.

X-rays: general↗

DOE OSTI2024

Enhancing Solar Power Forecasting with Regularized Constrained Quantile Regression Averaging and Bootstrapping Techniques

Probabilistic solar power forecasting (SPF) plays an essential role in optimizing power-grid operations by quantifying the forecast uncertainty. To improve the accuracy and robustness of probabilistic SPF, this paper introduces the regularized constrained quantile regression averaging (rCQRA) method to combine outputs from multiple PSPF models. In addition, a bootstrapping method was used to quantify model uncertainty, providing insights into the reliability and significance of each ensemble component. To evaluate its efficacy, the proposed rCQRA method is used to integrate four PSPF methods. The resulting SPF models are trained and validated using a real-world six-year dataset from a rooftop solar plant in the USA. The performance of the proposed rCQRA method is evaluated and compared with two benchmark methods under three categories of weather conditions. It is shown that the rCQRA method has superior performance in its forecast reliability, sharpness, and accuracy.

Ensemble learning, probabilistic solar power forec↗