SEARCH · Search NASA

Results for “Synthetic data generation”

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 109 records · Page 6

All sources NASA NTRS DOE OSTI

DOE OSTI2024

A Visual Analytic Platform for Interactive Validation of Human Mobility Simulations

Human mobility insights guide domain experts in an array of decisions, including critical infrastructure design, disaster response, epidemic modeling, national security, and policy making. Due to the inherent noise and privacy concerns in real-world individual-level mobility data, it is often preferred to leverage simulators that generate synthetic mobility data instead. However, it is critical to inspect and validate the output of such simulators to ensure the synthetic data is aligned with the characteristics of the population and the area of interest known to domain experts. While there exist many quantitative approaches for validating synthetic data, we argue it is also important to also validate such data qualitatively to capture aspects that are known to domain experts but difficult to quantify. In this work, we demonstrate a visual analytic platform that empowers domain experts to interact with their simulation outputs along spatial and temporal dimensions. By augmenting automated techniques and human skills, our visual analytic platform is a step towards interactive capabilities for model steering and quality control of mobility simulators.

Monadjemi, Shayan↗

NASA NTRSDate not supplied

Evaluating the Efficacy of Conditional Variational Autoencoders in Generating Synthetic Single Nuclei RNA-Seq Data for Space Biology Research

Astronauts are subject to unique stressors during spaceflight, leading to changes in their cellular function. However, neither astronauts nor model organisms respond the same to spaceflight, and research implicates a contribution of omics components in differential responses. Understanding how gene expression affects astronaut health is critical for the success of long-term space missions, prompting interest in developing personalized predictive models leveraging artificial intelligence (AI) and machine learning (ML) techniques. Developing such models requires extensive data, which is challenging to obtain and share. This study explores the use of conditional variational autoencoders (CVAEs) to synthetically generate single-nuclei RNA-seq (snRNA-seq) data. CVAEs build on standard variational autoencoders (VAEs) by conditioning data generation on covariates like sample identity and mission parameters, enhancing the relevance of generated data for specific contexts. For our work, we built two CVAEs with varying degrees of sparsity to optimize both interpretability and generative power. We train and validate models on existing snRNA-seq data collected from the brain tissue of mice subjected to spaceflight conditions and their ground control counterparts. We evaluate model performance using statistical tests and visualizations to compare synthetic data to real data. We aim to demonstrate that these prototype CVAE architectures could be used in future space biology work and that this is a method worth further exploring.

Sarah Golts↗

DOE OSTI2022

A computational pipeline to generate a synthetic dataset of metal ion sorption to oxides for AI/ML exploration

The charged mineral/electrolyte interfaces are ubiquitous in the surface and subsurface–including the surroundings of the geological disposal sites for radioactive waste. Therefore, understanding how ions interact with charged surfaces is critically important for predicting radionuclide mobility in the case of waste leakage. At present, the Surface Complexation Models (SCMs) are the most successful thermodynamic frameworks to describe ion retention by mineral surfaces. SCMs are interfacial speciation models that account for the effect of the electric field generated by charged surfaces on sorption equilibria. These models have been successfully used to analyze and interpret a broad range of experimental observations including potentiometric and electrokinetic titrations or spectroscopy. Unfortunately, many of the current procedures to solve and fit SCM to experimental data are not optimal, which leads to a non-transferable or non-unique description of interfacial electrostatics and consequently of the strength and extent of ion retention by mineral surfaces. Recent developments in Artificial Intelligence (AI) offer a new avenue to replace SCM solvers and fitting algorithms with trained AI surrogates. Unfortunately, there is a lack of a standardized dataset covering a wide range of SCM parameter values available for AI exploration and training–a gap filled by this study. Here, we described the computational pipeline to generate synthetic SCM data and discussed approaches to transform this dataset into AI-learnable input. First, we used this pipeline to generate a synthetic dataset of electrostatic properties for a broad range of the prototypical oxide/electrolyte interfaces. The next step is to extend this dataset to include complex radionuclide sorption and complexation, and finally, to provide trained AI architectures able to infer SCMs parameter values rapidly from experimental data. Here, we illustrated the AI-surrogate development using the ensemble learning algorithms, such as Random Forest and Gradient Boosting. These surrogate models allow a rapid prediction of the SCM model parameters, do not rely on an initial guess, and guarantee convergence in all cases.

Li, Chunhui↗

DOE OSTI2025

Synthetic Scientific Image Generation with VAE, GAN, and Diffusion Model Architectures

Generative AI (genAI) has emerged as a powerful tool for synthesizing diverse and complex image data, offering new possibilities for scientific imaging applications. This review presents a comprehensive comparative analysis of leading generative architectures, ranging from Variational Autoencoders (VAEs) to Generative Adversarial Networks (GANs) on through to Diffusion Models, in the context of scientific image synthesis. We examine each model's foundational principles, recent architectural advancements, and practical trade-offs. Our evaluation, conducted on domain-specific datasets including microCT scans of rocks and composite fibers, as well as high-resolution images of plant roots, integrates both quantitative metrics (SSIM, LPIPS, FID, CLIPScore) and expert-driven qualitative assessments. Results show that GANs, particularly StyleGAN, produce images with high perceptual quality and structural coherence. Diffusion-based models for inpainting and image variation, such as DALL-E 2, delivered high realism and semantic alignment but generally struggled in balancing visual fidelity with scientific accuracy. Importantly, our findings reveal limitations of standard quantitative metrics in capturing scientific relevance, underscoring the need for domain-expert validation. We conclude by discussing key challenges such as model interpretability, computational cost, and verification protocols, and discuss future directions where generative AI can drive innovation in data augmentation, simulation, and hypothesis generation in scientific research.

Generative Adversarial Networks↗

NASA NTRS2000

Estimates of Ground Temperature and Atmospheric Moisture from CERES Observations

A method is developed to retrieve surface ground temperature (T(sub g)) and atmospheric moisture using clear sky fluxes (CSF) from CERES-TRMM observations. In general, the clear sky outgoing longwave radiation (CLR) is sensitive to upper level moisture (q(sub l)) over wet regions and (T(sub g)) over dry regions The clear sky window flux from 800 to 1200/cm (RadWn) is sensitive to low level moisture (q(sub t)) and T(sub g). Combining these two measurements (CLR and RadWn), Tg and q(sub h) can be estimated over land, while q(sub h) and q(sub l) can be estimated over the oceans. The approach capitalizes on the availability of satellite estimates of CLR and RadWn and other auxiliary satellite data. The basic methodology employs off-line forward radiative transfer calculations to generate synthetic CSF data from two different global 4-dimensional data assimilation products. Simple linear regression is used to relate discrepancies in CSF to discrepancies in T(sub g), q(sub h) and q(sub l). The slopes of the regression lines define sensitivity parameters that can be exploited to help interpret mismatches between satellite observations and model-based estimates of CSF. For illustration, we analyze the discrepancies in the CSF between an early implementation of the Goddard Earth Observing System Data Assimilation System (GEOS-DAS) and a recent operational version of the European Center for Medium-Range Weather Prediction data assimilation system. In particular, our analysis of synthetic total and window region SCF differences (computed from two different assimilated data sets) shows that simple linear regression employing Delta(T(sub g)) and broad layer Delta(q(sub l) from .500 hPa to surface and Delta(q(sub h)) from 200 to .300 hPa provides a good approximation to the full radiative transfer calculations. typically explaining more than 90% of the 6-hourly variance in the flux differences. These simple regression relations can be inverted to "retrieve" the errors in the geophysical parameters. Uncertainties (normalized by standard deviation) in the monthly mean retrieved parameters range from 7% for Delta(T(sub g)) to about 20% for Delta(q(sub l)). Our initial application of the methodology employed an early CERES-TRMM data set (CLR and Radwn) to assess the quality of the GEOS2 data. The results showed that over the tropical and subtropical oceans GEOS2 is, in general, too wet in the upper troposphere (mean bias of 0.99 mm) and too dry in the lower troposphere (mean bias of -4.7 min). We note that these errors, as well as a cold bias in the T(sub g). have largely been corrected in the current version of GEOS-2 with the introduction of a land surface model, a moist turbulence scheme and the assimilation of SSM/I total precipitable water.

Wu, Man Li C.↗

NASA NTRS2000

Estimates of Ground Temperature and Atmospheric Moisture from CERES Observations

A method is developed to retrieve surface ground temperature (Tg) and atmospheric moisture using clear sky fluxes (CSF) from CERES-TRMM observations. In general, the clear sky outgoing long-wave radiation (CLR) is sensitive to upper level moisture (q(sub h)) over wet regions and Tg over dry regions The clear sky window flux from 800 to 1200 /cm (RadWn) is sensitive to low level moisture (q(sub j)) and Tg. Combining these two measurements (CLR and RadWn), Tg and q(sub h) can be estimated over land, while q(sub h) and q(sub t) can be estimated over the oceans. The approach capitalizes on the availability of satellite estimates of CLR and RadWn and other auxiliary satellite data. The basic methodology employs off-line forward radiative transfer calculations to generate synthetic CSF data from two different global 4-dimensional data assimilation products. Simple linear regression is used to relate discrepancies in CSF to discrepancies in Tg, q(sub h) and q(sub t). The slopes of the regression lines define sensitivity parameters that can be exploited to help interpret mismatches between satellite observations and model-based estimates of CSF. For illustration, we analyze the discrepancies in the CSF between an early implementation of the Goddard Earth Observing System Data Assimilation System (GEOS-DAS) and a recent operational version of the European Center for Medium-Range Weather Prediction data assimilation system. In particular, our analysis of synthetic total and window region SCF differences (computed from two different assimilated data sets) shows that simple linear regression employing (Delta)Tg and broad layer (Delta)q(sub l) from 500 hPa to surface and (Delta)q(sub h) from 200 to 500 hPa provides a good approximation to the full radiative transfer calculations, typically explaining more than 90% of the 6-hourly variance in the flux differences. These simple regression relations can be inverted to "retrieve" the errors in the geophysical parameters. Uncertainties (normalized by standard deviation) in the monthly mean retrieved parameters range from 7% for (Delta)T to about 20% for (Delta)q(sub t). Our initial application of the methodology employed an early CERES-TRMM data set (CLR and Radwn) to assess the quality of the GEOS2 data. The results showed that over the tropical and subtropical oceans GEOS2 is, in general, too wet in the upper troposphere (mean bias of 0.99 mm) and too dry in the lower troposphere (mean bias of -4.7 mm). We note that these errors, as well as a cold bias in the Tg, have largely been corrected in the current version of GEOS-2 with the introduction of a land surface model, a moist turbulence scheme and the assimilation of SSTM/I total precipitable water.

Wu, Man Li C.↗

DOE OSTI2020

Generation and representation of synthetic smart meter data

Advanced energy algorithms running at big-data scale will be necessary to identify, realize, and verify energy savings to meet government and utility goals of building energy efficiency. Any algorithm must be well characterized and validated before it is trusted to run at these scales. Smart meter data from real buildings will ultimately be required for the development, testing, and validation of these energy algorithms and processes. However, for initial development and testing, smart meter data are difficult to work with due to privacy restrictions, noise from unknown sources, data accessibility, and other concerns which can complicate algorithm development and validation. This paper describes a new methodology to generate synthetic smart meter data of electricity use in buildings using detailed building energy modeling, which aims to capture the variability and stochastics of real energy use in buildings. The methodology can create datasets tailored to represent specific scenarios with known truth and controllable amounts of synthetic noise. Knowledge of ground truth also allows the development and validation of enhanced processes which leverage building metadata, such as building type or size (floor area), in addition to smart meter data. The methodology described in this paper includes the key influencing factors of real-world building energy use including weather data, occupant-driven loads, building operation and maintenance practices, and special events. Data formats to support workflows leveraging both synthetic meter data and associated metadata are proposed and discussed. Finally, example use cases of the synthetic meter data are described to illustrate potential applications.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2023

MultiLoad-GAN: A GAN-Based Synthetic Load Group Generation Method Considering Spatial-Temporal Correlations

This paper presents a deep-learning framework, Multi-load Generative Adversarial Network (MultiLoad-GAN), for generating a group of synthetic load profiles (SLPs) simultaneously. The main contribution of MultiLoad-GAN is the capture of spatial-temporal correlations among a group of loads that are served by the same distribution transformer. This enables the generation of a large amount of correlated SLPs required for microgrid and distribution system studies. Here, the novelty and uniqueness of the MultiLoad-GAN framework are three-fold. First, to the best of our knowledge, this is the first method for generating a group of load profiles bearing realistic spatial- temporal correlations simultaneously. Second, two complementary realisticness metrics for evaluating generated load profiles are developed: computing statistics based on domain knowledge and comparing high-level features via a deep-learning classifier. Third, to tackle data scarcity, a novel iterative data augmentation mechanism is developed to generate training samples for enhancing the training of both the classifier and the MultiLoad-GAN model. Simulation results show that MultiLoad- GAN can generate more realistic load profiles than existing approaches, especially in group level characteristics. With little finetuning, MultiLoad-GAN can be readily extended to generate a group of load or PV profiles for a feeder or a service area.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2021

Synthetic galaxy clusters and observations based on Dark Energy Survey Year 3 Data

ABSTRACT We develop a novel data-driven method for generating synthetic optical observations of galaxy clusters. In cluster weak lensing, the interplay between analysis choices and systematic effects related to source galaxy selection, shape measurement, and photometric redshift estimation can be best characterized in end-to-end tests going from mock observations to recovered cluster masses. To create such test scenarios, we measure and model the photometric properties of galaxy clusters and their sky environments from the Dark Energy Survey Year 3 (DES Y3) data in two bins of cluster richness $\lambda \in [30; 45)$, $\lambda \in [45; 60)$ and three bins in cluster redshift ($z\in [0.3; 0.35)$, $z\in [0.45; 0.5)$ and $z\in [0.6; 0.65)$. Using deep-field imaging data, we extrapolate galaxy populations beyond the limiting magnitude of DES Y3 and calculate the properties of cluster member galaxies via statistical background subtraction. We construct mock galaxy clusters as random draws from a distribution function, and render mock clusters and line-of-sight catalogues into synthetic images in the same format as actual survey observations. Synthetic galaxy clusters are generated from real observational data, and thus are independent from the assumptions inherent to cosmological simulations. The recipe can be straightforwardly modified to incorporate extra information, and correct for survey incompleteness. New realizations of synthetic clusters can be created at minimal cost, which will allow future analyses to generate the large number of images needed to characterize systematic uncertainties in cluster mass measurements.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2024

Denoising diffusion probabilistic models for generative alloy design

Inverse material design is an extremely challenging optimization task made difficult by, in part, the highly nonlinear relationship linking performance with composition. Quantitative approaches have improved significantly owing to advances in high throughput experimentation and computational thermodynamics. However, existing physics-based tools are mostly forward models; input a chemistry and obtain a prediction. More recently the materials community has leveraged advances in the machine learning community to establish novel inverse design frameworks. Very recently denoising diffusion probabilistic models have been shown to be extremely powerful generators producing synthetic data of various modalities e.g. images, text, audio, tables, etc.. In this work a novel framework for alloy design and optimization is proposed leveraging these class of models. Five key generative tasks are demonstrated (1) unconditional generation (2) composition conditioned generation (3) property conditioned generation (4) multi-feedstock conditioned generation and (5) generative optimization. These methods were tested on three case studies: high entropy alloy design, superalloy binder jet additive manufacturing, and in-situ dual-feedstock wire-arc additive manufacturing. Results indicate that the established models are extremely flexible, expressive, and robust. The architecture’s flexibility and training procedure empower the model to learn complex intra-compositional and composition-property relationships. Furthermore, the probabilistic nature of these models makes them well suited for addressing solution non-uniqueness and tackling uncertainty quantification tasks. While the fidelity and quantity of the underlying training data is paramount, we envision that future alloy design frameworks will make extensive use of these kinds of machine learning models as “search” tools bolstering the utility of experimental and computational approaches.

36 MATERIALS SCIENCE↗

DOE OSTI2021

SeismoGen: Seismic Waveform Synthesis Using GAN With Application to Seismic Data Augmentation

Abstract Detecting earthquake arrivals within seismic time series can be a challenging task. Visual, human detection has long been considered the gold standard but requires intensive manual labor that scales poorly to large data sets. In recent years, automatic detection methods based on machine learning have been developed to improve the accuracy and efficiency. However, the accuracy of those methods relies on access to a sufficient amount of high‐quality labeled training data, often tens of thousands of records or more. We aim to resolve this dilemma by answering two questions: (1) provided with a limited amount of reliable labeled data, can we use them to generate additional, realistic synthetic waveform data? and (2) can we use those synthetic data to further enrich the training set through data augmentation, thereby enhancing detection algorithms? To address these questions, we use a generative adversarial network (GAN), a type of machine learning model which has shown supreme capability in generating high‐quality synthetic samples in multiple domains. Once trained, our GAN model is capable of producing realistic seismic waveforms of multiple labels (noise and event classes). Applied to real Earth seismic data sets in Oklahoma, we show that data augmentation from our GAN‐generated synthetic waveforms can be used to improve earthquake detection algorithms in instances when only small amounts of labeled training data are available.

Wang, Tiantong↗

DOE OSTI2021

Modelling dust transport in DIII-D with DTOKS-Upgrade

Comprehensive upgrades to the dust transport code Dust in TOKamaks (DTOKS) that extend the plasma-dust interaction model are presented and compared with recent measurements of dust transport in DIII-D. Simulations incorporating variation in physical properties of graphite dust with temperature and size in a stationary plasma background suggest a substantial decrease in lifetimes due principally to thermal expansion. The trajectories of 53 dust grains identified from analysis of visible camera data taken across two similar shots were used to measure the dust particle velocity distributions. Dust tracks terminated mostly at the outer divertor strike point having a mean observation time of 2.1 ± 0.4 ms. Stochastic modelling of 200 graphite dust particles in the DIII-D tokamak performed with DTOKS-U using plasma simulations generated by OEDGE found similar behaviour, with particles ablating rapidly after acquiring a positive charge in the region close to the outer strike point, creating an acute source of neutral carbon atoms. The simulated mean lifetime, 11 ± 2 ms, showed approximate agreement with experimental observation when corrected by accounting for dust visibility and ignoring the longest trajectories 1.5 ± 0.2 ms. Synthetic diagnostic data generated from coupling the results of DTOKS-U with the visualisation software Calcam offers a powerful new tool for validation of simulations and predictive calculations of dust dynamics.

Physics↗

NASA NTRSDate not supplied

Data Science and Urban Air Mobility: Challenges and Opportunities

Aviation is broadly a combination of aircraft, airspace and airports. The data science life cycle comprises of five steps - capture, maintain, process, analyze and communicate. The presentation introduces the legacy of conventional aviation research in the context of the data science life cycle to motivate the challenges with Urban Air Mobility, a field that is quite nascent. A summary of recent research will be presented to highlight the innovative ways to address the challenges. Examples provided will include the generation of synthetic data, encounter models from simulations, and leveraging novel and diverse data sets from traditional transportation and non-aviation sources, to analyze problems of operation in urban airspace. Finally, opportunities will be identified for further exploration, niche development and filling the gaps in the field of data science for UAM.

Urban Air Mobility↗

NASA NTRS2020

Data Science and Urban Air Mobility: Challenges and Opportunities

Urban Air Mobility↗

NASA NTRSDate not supplied

Data Science Challenges for Urban Air Mobility

Aviation is a combination of aircraft, airspace and airports. The data science life cycle comprises of five steps - capture, maintain, process, analyze and communicate. The presentation introduces the legacy of conventional aviation research in the context of the data science life cycle to motivate the challenges with Urban Air Mobility, a field that is quite nascent. A summary of recent research will be presented to highlight the innovative ways to address the challenges. Examples provided will include the generation of synthetic data, encounter models from simulations, and leveraging novel and diverse data sets from traditional transportation and non-aviation sources, to analyze problems of operation in urban airspace. Finally, opportunities will be identified for further exploration, niche development and filling the gaps in the field of data science for UAM.

Data Science↗

NASA NTRS2020

Data Science Challenges for Urban Air Mobility

Data Science↗

NASA NTRSDate not supplied

Bond Line Thickness Estimation in Composite Structures Using Multiple Inspection Techniques

Imaging and other nondestructive evaluation techniques are commonly used for material characterization and defect recognition in safety critical aerospace applications, with data fusion providing the framework for uncertainty quantification in these contexts. Most commonly, forward physics-based modeling predicts the response conditioned on material properties and defect assumptions, and probabilistic methods are used to infer the hidden state of subject of the inspection from a combination of prior information, likelihoods, and inspection data. In this paper Bayesian methods are used to estimate bond thickness in lap joints comprised of aluminum adherends using a combination of infrared thermography, digital radiography, and ultrasound. The accuracy of the fused estimates are validated against data generated from synthetic specimens, and by comparison against high resolution X-ray computed tomography inspections of built specimens.

thermal nondestructive evaluation↗

DOE OSTI2026

Generalizing synthetic data-trained acoustic predictive models to real-world measurements

Acoustic Resonance Spectroscopy (ARS) is highly sensitive to structural properties such as material, geometry, and environmental conditions; as a consequence, it can noninvasively measure internal properties that are unobservable by most other methods. Because of its sensing capabilities and low implementation cost and complexity, ARS has potential as a paradigm shift in noninvasive sensing, characterization, and monitoring applications. However, extracting specific properties from ARS measurements, comprising the vibration spectrum of a test object, is challenging due to the sensitivity of the spectra to other structural changes not being measured, e.g. manufacturing tolerances, component coupling, environmental variation, etc. Neural Networks are promising tools for identifying trends in ARS measurements, but their training typically requires large datasets, which are often impractical to obtain for real-world systems. Synthetic data can be simulated efficiently, but discrepancies between synthetic and real-world data frequently lead to poor generalization when testing on the real-world data. We propose a novel ARS model training framework that enables networks trained exclusively on synthetic ARS data to generalize effectively to real-world measurements. Our approach leverages the Correlation Alignment (CORAL) technique to enforce the extraction of features common to both synthetic and real-world domains. As a case study, we demonstrate noninvasive ARS-based pressure measurements in sealed systems. Finite element method (FEM) simulations were used to generate synthetic training data across diverse vessel configurations and pressure conditions, and model performance was then tested on real-world measurements. We demonstrate that robust machine learning models for ARS can be developed without large real-world datasets, significantly broadening the applicability of ARS for noninvasive sensing. Moreover, the approach is extensible to other sensing modalities where synthetic data are abundant but real-world data are limited.

36 MATERIALS SCIENCE↗