SEARCH · Engineering Papers
Results for “Imputation”
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.
Improving Groundwater Imputation through Iterative Refinement Using Spatial and Temporal Correlations from In Situ Data with Machine Learning
Explore the source record for details and available documents.
Salvaging Data Records with Missing Data: Data Imputation using the Multivariate t Distribution
No abstract provided
NASA Instrument Cost Model: Mission Class’s Impact, Imputation, Multiple Builds and More: New Features in the Latest Version of NICM
No abstract provided
Complete Global Total Electron Content Map Dataset based on a Video Imputation Algorithm VISTA
Explore the source record for details and available documents.
Leveraging large language models to address data scarcity in machine learning for graphene synthesis
Machine learning in experimental materials science faces significant challenges due to the scarcity of data, which are costly and time-consuming to generate, particularly when relying on in-house experiments. Literature data mining offers a potential solution but introduces issues like mixed data quality, inconsistent formats, and non-uniform reporting of synthesis parameters, resulting in partially missing and heterogeneous features across the dataset. Here, we propose data imputation and feature engineering methods that employ pre-trained large language models (LLMs) to enhance machine learning performance on scarce, heterogeneous datasets, demonstrated on graphene CVD synthesis data and the ML-HydPARK hydrogen storage dataset. GPT models perform data imputation via tailored prompting and semantic normalization of inconsistently reported features through embeddings, for example, to harmonize the complex nomenclature of CVD substrates. Beyond yielding more diverse and richer feature representations than traditional methods such as K-nearest neighbors (KNN) and Multivariate Imputation by Chained Equations (MICE), LLM-based data imputation is evaluated against dataset characteristics and prompting strategies. We vary the level of autonomy granted to the LLM, from generic prompting that leverages pre-trained knowledge for autonomous data generation to data-informed prompting that constrains outputs using target-specific information, and demonstrate which level of autonomy yields superior imputation performance across datasets and feature types. The proposed data engineering methods markedly improve downstream performance; for example, in graphene layer number classification using a support vector machine (SVM), binary accuracy increases from 39% to 65% and ternary accuracy from 52% to 72%. Fine-tuning experiments on both datasets show that combining our proposed LLM-based data imputation and feature encoding methods with numerical machine learning predictors outperforms standalone fine-tuned LLM predictors in data-scarce settings. The proposed strategies emphasize data enhancement techniques rather than refining learning architectures or regularizing loss functions, offering a broadly applicable framework for improving machine learning performance on scarce, inhomogeneous datasets.
MultiPEM Toolbox: User Manual [Rev. 2]
This document explains use of the Multi-Phenomenology Explosion Monitoring (Multi PEM) Toolbox, a collection of R scripts for estimating the unknown device parameters of a new event with uncertainty quantification. The methodology and application used for illustration in this user manual are fully documented in a Los Alamos National Laboratory technical report hereafter designated “WPA” for reference. Additional details on the application are found in a recent journal article. Two assessment types are available: rapid and complete. Rapid assessments are conducted in two stages, as described in Section 2. In the first stage, calibration data are used to estimate forward and error model parameters (WPA, §5.1) and (if relevant) errors-in-variables yield values for calibration sources (WPA, §3, Equation (3)). In the second stage, new event data are used to estimate the unknown new event device parameters (WPA, §5.2) with uncertainty quantification. Two options for treating the inferred first stage parameters in second stage Bayesian analysis are available: fixing them at their maximum likelihood estimate (default), or multiple imputation. Multiple imputation involves utilizing several posterior samples (imputations) of the first stage parameters as fixed values in the second stage posterior sampling of the new event device parameters. Second stage sampling is conducted across imputations in parallel to improve computational efficiency. This method produces improved uncertainty quantification of the new event device parameters compared with the default treatment of the first stage parameters, at the expense of additional computation. Complete assessments are conducted in a single stage, as described in Section 3. Calibration and (if relevant) new event data are used simultaneously to estimate all forward model, error model, and (if relevant) new event device parameters with uncertainty quantification on the latter. As the name suggests, rapid assessments generally run substantially faster than complete assessments (even with multiple imputation), because the results of first stage analysis can be stored and incorporated into estimating a relatively low-dimensional space of new event device parameters whenever relevant new event data becomes available. On the other hand, complete assessments must be run on the full set of model and device parameters with calibration and new event data every time the latter becomes available.
Evaluation of Time-Series Gap-Filling Methods for Solar Irradiance Applications
A complete solar resource data set is essential for any stage of a solar energy project - from feasibility studies to daily operations. But measured or modeled solar resource data are prone to data gaps and data quality issues. To mitigate these issues, a data imputation process should be implemented to obtain a complete and reliable temporal and spatial data series. This study focused on imputing temporal scales by applying random and artificial data gaps and then implementing eight imputation methods, including the Kalman filtering and smoothing and stine interpolations. These methods were implemented on 1-minute to half hourly irradiance data for 1 year using a few locations from the National Solar Radiation Database (NSRDB) and ground measurement data set. The results demonstrated that some of the simpler methods, such as the stine and linear interpolation methods, were the relatively best models based on the statistical metrics for imputing NSRDB and ground measurement data, respectively.
Astronaut Preflight Cardiovascular Variables Associated with Vascular Compliance are Highly Correlated with Post-Flight Eye Outcome Measures in the Visual Impairment Intracranial Pressure (VIIP) Syndrome Following Long Duration Spaceflight
The detection of the first VIIP case occurred in 2005, and adequate eye outcome measures were available for 31 (67.4%) of the 46 long duration US crewmembers who had flown on the ISS since its first crewed mission in 2000. Therefore, this analysis is limited to a subgroup (22 males and 9 females). A "cardiovascular profile" for each astronaut was compiled by examining twelve individual parameters; eleven of these were preflight variables: systolic blood pressure, pulse pressure, body mass index, percentage body fat, LDL, HDL, triglycerides, use of anti‐lipid medication, fasting serum glucose, and maximal oxygen uptake in ml/kg. Each of these variables was averaged across three preflight annual physical exams. Astronaut age prior to the long duration mission, and inflight salt intake was also included in the analysis. The group of cardiovascular variables for each crew member was compared with seven VIIP eye outcome variables collected during the immediate post‐flight period: anterior-posterior axial length of the globe measured by ultrasound and optical biometry; optic nerve sheath diameter, optic nerve diameter, and optic nerve to sheath ratio‐ each measured by ultrasound and magnetic resonance imaging (MRI), intraocular pressure (IOP), change in manifest refraction, mean retinal nerve fiber layer (RNFL) on optical coherence tomography (OCT), and RNFL of the inferior and superior retinal quadrants. Since most of the VIIP eye outcome measures were added sequentially beginning in 2005, as knowledge of the syndrome improved, data were unavailable for 22.0% of the outcome measurements. To address the missing data, we employed multivariate multiple imputation techniques with predictive mean matching methods to accumulate 200 separate imputed datasets for analysis. We were able to impute data for the 22.0% of missing VIIP eye outcomes. We then applied Rubin's rules for collapsing the statistical results across our 200 multiply imputed data sets to assess the canonical correlation between the eye outcomes and the twelve astronaut cardiovascular variables available for all 31 subjects. Results: A highly significant canonical correlation was observed among the canonical solutions (p<.00001), with an average best canonical correlation of.97. The results suggest a strong association between astronauts' measures of cardiovascular health and the seven eye outcomes of the VIIP syndrome used in this analysis. Furthermore, the "joint test" revealed a significant difference in cardiovascular profile between male and female astronauts (Prob > F = 0.00001). Overall, female astronauts demonstrated a significantly healthier cardiovascular status. Individually, the female astronauts had significantly healthier profiles on seven of twelve cardiovascular variables than the men (p values ranging from <0.0001 to <0.05). Male astronauts did not demonstrate significantly healthier values on any of the twelve cardiovascular variables measured
Multi-agent voltage control in distribution systems using GAN-DRL-based approach
Active distribution grids can experience voltage fluctuations and violations due to the high penetration of variable distributed energy resources (DERs). These problems might occur because of the uncertain and variable generation natures of these resources, especially solar photovoltaic resources, during panel shadowing scenarios. Volt-VAR control (VVC) is an efficient method that controls the reactive power set-points of the inverters to regulate the voltage of distribution grids. Although several VVC approaches have been proposed recently, the performance of these approaches degrades significantly if behind-the-meter solar generation data are unobservable/missing. Therefore, it is necessary to impute missing/unobservable PV data accurately to be utilized in VVC approaches. Further, this paper proposes a model-free, data-driven, centrally trained, and decentrally executed multi-agent deep reinforcement learning-based VVC architecture to regulate the voltage of distribution networks. A generative adversarial network (GAN) is incorporated to impute the unobservable PV data accurately, which improves the performance of the proposed control architecture. The proposed multi-agent-soft-actor–critic algorithm (MASAC)-based VVC technique utilizes the actual PV dataset as well as the imputed dataset from the GAN framework to learn the optimal coordinated control policy for controlling the optimal reactive power set-points of PV inverters. The effectiveness of the proposed approach is analyzed on a modified IEEE 34-bus test case with added PV inverters. The results are compared and analyzed with a base case model with no VVC and VVC with a local droop control approach, genetic algorithm optimization, and a centralized soft actor–critic-based approach. Moreover, the performance of the proposed approach is compared with that of a multi-agent VVC framework without using the PV generation data and load information as the system state. The results illustrate that the proposed method with more state input improves the voltage profile and reduces the power loss of the network across various loading and PV generation scenarios.
Revised monthly energy generation estimates for 1,500 hydroelectric power plants in the United States
Abstract The U.S. Energy Information Administration (EIA) conducts a regular survey (form EIA-923) to collect annual and monthly net generation for more than ten thousand U.S. power plants. Approximately 90% of the ~1,500 hydroelectric plants included in this data release are surveyed at annual resolution only and thus lack actual observations of monthly generation. For each of these plants, EIA imputes monthly generation values using the combined monthly generating pattern of other hydropower plants within the corresponding census division. The imputation method neglects local hydrology and reservoir operations, rendering the monthly data unsuitable for various research applications. Here we present an alternative approach to disaggregate each unobserved plant’s reported annual generation using proxies of monthly generation—namely historical monthly reservoir releases and average river discharge rates recorded downstream of each dam. Evaluation of the new dataset demonstrates substantial and robust improvement over the current imputation method, particularly if reservoir release data are available. The new dataset—named RectifHyd—provides an alternative to EIA-923 for U.S. scale, plant-level, monthly hydropower net generation (2001–2020). RectifHyd may be used to support power system studies or analyze within-year hydropower generation behavior at various spatial scales.
Enhancing approximate modular Bayesian inference by emulating the conditional posterior
In modular Bayesian analyses, complex models are composed of distinct modules, each representing different aspects of the data or prior information. In this context, fully Bayesian approaches can sometimes lead to undesirable feedback between modules, compromising the integrity of the inference. The “cut-distribution” prevents unwanted influence between modules by “cutting” feedback. The direct sampling (DS) algorithm is standard practice for approximating the cut-distribution, but it can be computationally intensive, especially when the number of imputations required is large. An enhanced method is proposed, the Emulating the Conditional Posterior (ECP) algorithm, which leverages emulation to increase the number of imputations. Through numerical experiment it is demonstrated that the ECP algorithm outperforms the traditional DS approach in terms of accuracy and computational efficiency, particularly when resources are constrained. Here, it is also shown how the DS algorithm can be improved using ideas from design of experiments. Some practical recommendations are given for algorithm choice in modular Bayesian analyses.
Latent Neural ODE for Integrating Multi-Timescale Measurements in Smart Distribution Grids
Under a smart grid paradigm, there has been an increase in sensor installations to enhance situational awareness. The measurements from these sensors can be leveraged for real-time monitoring, control, and protection. However, these measurements are typically irregularly sampled. These measure-ments may also be intermittent due to communication bandwidth limitations. To tackle this problem, this paper proposes a novel latent neural ordinary differential equations (LODE) approach to aggregate the unevenly sampled multivariate time-series measurements. The proposed approach is flexible in performing both imputations and predictions while being computationally efficient. Simulation results on IEEE 37 bus test systems illustrate the efficiency of the proposed approach.
LandScan Mosaic
The LandScan program at Oak Ridge National Laboratory (ORNL), in collaboration with the National Geospatial-Intelligence Agency (NGA), continues to deliver the most accurate and up to date global, high resolution gridded population data. Additionally, the latest advancements in the LandScan HD methodology led to reduced latency in development of rapid updates for geopolitical events. With momentum towards reporting more up to date population estimates, feedback from the user community expressed interest in reporting population estimates in ranges - whether to express a level of uncertainty or confirm to leadership and stakeholders the modeled data are estimates. Building upon the need to understand uncertainty or confidence in the modeled data and report ranges at the global scale, LandScan Mosaic was developed. LandScan Mosaic represents the next generation of high-resolution population modeling, building upon the established success of previous LandScan HD iterations. While LandScan HD employed a deterministic big data fusion approach, LandScan Mosaic enhances this methodology by integrating advanced machine learning techniques to impute missing, yet crucial, population model parameters. This advancement allows for probabilistic modeling of building occupancy and population distribution, incorporating uncertainty quantification through Monte Carlo sampling methods. By combining big data fusion with machine learning-driven imputation and stochastic modeling, LandScan Mosaic provides a more comprehensive and robust representation of population dynamics. LandScan Mosaic will be following the in the footsteps of its longstanding counterpart LandScan Global and releasing a global gridded population raster, at the 3-arcsecond resolution. This technical report documents the current stage of development of LandScan Mosaic, detailing the methodologies and data sources behind the modeling. Stakeholders are encouraged to use this document as an authoritative reference for insight into Mosaic’s data development processes. However, readers should note that LandScan Mosaic remains in a late-stage research and development phase, and methodologies and data presented here are subject to refinements ahead of the anticipated global release in Summer 2025. Feedback and inquiries from users and stakeholders are welcomed as we continue to refine and enhance this important population resource.
Cleaning Images with Gaussian Process Regression
Many approaches to astronomical data reduction and analysis cannot tolerate missing data: corrupted pixels must first have their values imputed. This paper presents astrofix, a robust and flexible image imputation algorithm based on Gaussian process regression. Through an optimization process, astrofix chooses and applies a different interpolation kernel to each image, using a training set extracted automatically from that image. It naturally handles clusters of bad pixels and image edges and adapts to various instruments and image types. For bright pixels, the mean absolute error of astrofix is several times smaller than that of median replacement and interpolation by a Gaussian kernel. We demonstrate good performance on both imaging and spectroscopic data, including the SBIG 6303 0.4 m telescope and the FLOYDS spectrograph of Las Cumbres Observatory and the CHARIS integral-field spectrograph on the Subaru Telescope.
A consistent dataset for the net income distribution for 190 countries and aggregated to 32 geographical regions from 1958 to 2015
Abstract. Data on income distributions within and across countries are becoming increasingly important for informing analysis of income inequality and understanding the distributional consequences of climate change. While datasets on income distribution collected from household surveys are available for multiple countries, these datasets often do not represent the same concept of inequality (or income concept) and therefore make comparisons across countries, over time and across datasets difficult. Here, we present a consistent dataset of income distributions across 190 countries from 1958 to 2015 measured in terms of net income. We complement the observed values in this dataset with values imputed from a summary measure of the income distribution, specifically the Gini coefficient. For the imputation, we use a recently developed nonparametric principal-component-based approach that shows an excellent fit to data on income distributions compared to other approaches. We also present another version of this dataset aggregated from the country level to 32 geographical regions. Our dataset is developed for the purpose of calibrating models such as integrated human–Earth system models with detailed data on income distributions. This dataset will enable more robust analysis of income distribution at multiple scales. The latest version of our data are available on Zenodo: https://doi.org/10.5281/zenodo.7093997 (Narayan et al., 2022b).
Full-Coverage High-Resolution Daily PM(sub 2.5) Estimation using MAIAC AOD in the Yangtze River Delta of China
Satellite aerosol optical depth (AOD) has been used to assess population exposure to fine particulate matter (PM (sub 2.5)). The emerging high-resolution satellite aerosol product, Multi-Angle Implementation of Atmospheric Correction(MAIAC), provides a valuable opportunity to characterize local-scale PM(sub 2.5) at 1-km resolution. However, non-random missing AOD due to cloud snow cover or high surface reflectance makes this task challenging. Previous studies filled the data gap by spatially interpolating neighboring PM(sub 2.5) measurements or predictions. This strategy ignored the effect of cloud cover on aerosol loadings and has been shown to exhibit poor performance when monitoring stations are sparse or when there is seasonal large-scale missngness. Using the Yangtze River Delta of China as an example, we present a Multiple Imputation (MI) method that combines the MAIAC high-resolution satellite retrievals with chemical transport model (CTM) simulations to fill missing AOD. A two-stage statistical model driven by gap-filled AOD, meteorology and land use information was then fitted to estimate daily ground PM(sub 2.5) concentrations in 2013 and 2014 at 1 km resolution with complete coverage in space and time. The daily MI models have an average R(exp 2) of 0.77, with an inter-quartile range of 0.71 to 0.82 across days. The overall Ml model 10-fold cross-validation R(exp 2) (root mean square error) were 0.81 (25 gm(exp 3)) and 0.73 (18 gm(exp 3)) for year 2013 and 2014, respectively. Predictions with only observational AOD or only imputed AOD showed similar accuracy.Comparing with previous gap-filling methods, our MI method presented in this study performed bette rwith higher coverage, higher accuracy, and the ability to fill missing PM(sub 2.5) predictions without ground PM(sub 2.5) measurements. This method can provide reliable PM(sub 2.5)predictions with complete coverage that can reduce biasin exposure assessment in air pollution and health studies.
Linkage map construction using limited parental genotypic information
Abstract Genetic linkage maps based on single nucleotide polymorphisms (SNPs) represent an essential tool for a variety of genomic analyses. Today, next-generation sequencing (NGS) enables rapid genotyping of different mapping populations based on thousands of SNPs and the construction of highly saturated linkage maps. Nevertheless, missing data in the genotyping of the parental lines creates a bottleneck that determines the number of SNPs that can be used for the linkage map. As a proof of concept, a highly saturated genetic linkage map was constructed using the imputed genotypic data of a recombinant inbred line (RIL) population and the limited genotypic information of its parental lines. Two ABH genotype files were created from a pseudo-parental genotypic data set that includes all the SNPs present in the RIL population. In the first ABH file pseudo-parental 1 was considered parental A, while in the second pseudo-parental 1 was considered parental B. These two duplicate ABH genotype files were merged by chromosome and subjected to linkage map analysis. Since the ABH data were duplicated, two mirrored linkage groups were generated per chromosome. The correct linkage map was identified and selected based on the partial genotypic data of the parental lines. This strategy was effective for constructing a highly saturated linkage map of 33,421 SNPs based on the genotyping of 205 RILs and a limited number of 100 SNPs present in the parental lines. This strategy enables the use of all the NGS SNP data obtained from a low-coverage sequencing experiment in the mapping population.