SEARCH · Search NASA

Results for “model accuracy”

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 505 records · Page 28

All sources NASA NTRS DOE OSTI

NASA NTRS2022

Machine learning for Deep Space Network antenna motions detection

Highly stable frequency and timing standards are essential for deep-space missions and radio science. At the NASA Deep Space Network (DSN), these standards are distributed through a network of underground fiber cables to support several Goldstone antennas. Independently developed frequency-measuring instruments generate tremendous quantities of data to monitor and validate the antennas’ stringent frequency requirements. In this paper, we propose a lightweight processing tool capable of detecting disturbances on the frequency signal caused by DSN antenna motions. Our training data is sampled from the movement log of the antenna of interest and the generated data from the fiber optic metrology instrument linked to the antenna. We demonstrate that a convolutional neural network (CNN) model can achieve high accuracies on classifying instances of antenna movements and is an effective predictor when used iteratively on longer, variable stretches of metrology data. The simplicity, low training cost, and high accuracies of our model strongly suggest its efficacy in identifying and troubleshooting frequency disturbances caused by the antenna.

Yi, Lin↗

NASA NTRSDate not supplied

GT2024-139262: Development of a Computational Fluid Dynamics Model for a High-Speed Centrifugal Compressor

Computational fluid dynamics (CFD) has become widely used in the design and analysis of turbomachinery components such as centrifugal compressors. However, CFD is only a limited representation of experimental cases and struggles to model complex flows or lack model details to increase computation speed. The High Efficiency Centrifugal Compressor (HECC) was designed by United Technologies Research Center for NASA to investigate the difficulties behind improving centrifugal compressor technology and to provide an open case to the turbomachinery research community. The CFD technology seen in previous numerical models of the compressor have improved since then. In this study, a high-fidelity computational model for the vaneless configuration of HECC operating at its design point was developed and compared to an equivalent experimental case. This model incorporated GPU versions of mesh generation code WAND and solver code LEO from AeroDynamic Solutions, Inc. Blade sections extracted from the solid model formed the basis of the impeller mesh, while the remainder of the stage was modeled as ducts. Standard day aerodynamic conditions, design rotational speed, design mass flow rate, and inlet flow angles served as additional inputs. A mesh sensitivity study was performed prior to using the model for analysis. The accuracy of the model’s predictions of impeller and stage performance were assessed by comparing the total pressure ratio, total temperature rise ratio, and adiabatic efficiency to the corresponding parameters observed in the experimental case. Results showed a general overprediction of compressor performance by the model. Predictions were more accurate for the impeller performance compared to those for the full compressor stage. Spanwise analyses showed that the model tended to follow flow behavior patterns exhibited in the experiment. Beyond this study, this model was used to examine HECC aerodynamics and performance in research following thereafter.

centrifugal compressor↗

DOE OSTI2025

Contrasting Time-Frequency Representations for Unknown Waveform Detection

In real-world applications like spectrum management and interference detection, dealing with unseen electromagnetic waveforms is critical. Although some methods attempt to simulate open set data using generator models, they face challenges in generating synthetic samples for open set while simultaneously selecting an optimal discriminator for accurate classification. This results in difficulties capturing distinctive features across classes, especially in dynamic scenarios where new classes emerge. To detect unseen waveforms, we propose combining time and frequency domain features with cosine similarity loss to enhance feature distinctiveness and enabling more accurate predictions. This approach efficiently captures more comprehensive information than single-domain representations or approaches without cosine loss. Additionally, our model avoids generic feature vectors by extracting class-specific features during training, resulting in improved class representation. The experiment results show that this combined feature approach with cosine loss outperforms single-domain models and improves accuracy by 10\% over models without cosine loss.

99 - GENERAL AND MISCELLANEOUS↗

DOE OSTI2025

Contrasting Time-Frequency Representations for Unknown Waveform Detection

Identifying unseen electromagnetic waveforms is critical for many applications, like interference management, electronic warfare and spectrum management. Traditionally this is done using statistical methods for anomaly detection, which has evolved to deep learning models for identifying the unseen data, formally termed as open set recognition. Some prior methods use a generative model to emulate open set data, which face challenges in generating synthetic samples for open set while simultaneously selecting an optimal discriminator for accurate classification. To alleviate this issue, we propose a discriminative model that effectively combines time and frequency domain features of communication signals for accurate predictions. We further introduce a cosine similarity loss that makes the domain specific features unique to enhance the prediction rate. Additionally, our model avoids generic feature vectors by extracting class-specific features during training, resulting in improved class representation. The experiment results show that this combined feature approach with cosine loss outperforms single-domain models and improves accuracy by 10% over models without cosine loss.

99 - GENERAL AND MISCELLANEOUS↗

DOE OSTI2021

White paper on Verification and Validation for Cyber Emulation Models

All disciplines that use models to predict the behavior of real-world systems need to determine the accuracy of the models’ results. Techniques for verification, validation, and uncertainty quantification (VVUQ) focus on improving the credibility of computational models and assessing their predictive capability. VVUQ emphasizes rigorous evaluation of models and how they are applied to improve understanding of model limitations and quantify the accuracy of model predictions.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

A systematic method for selecting molecular descriptors as features when training models for predicting physiochemical properties

Machine learning has proven to be a powerful tool for accelerating biofuel development. Although numerous models are available to predict a range of properties using chemical descriptors, there is a trade-off between interpretability and performance. Neural networks provide predictive models with high accuracy at the expense of some interpretability, while simpler models such as linear regression often lack in accuracy. In addition to model architecture, feature selection is also critical for developing interpretable and accurate predictive models. We present a method for systematically selecting molecular descriptor features and developing interpretable machine learning models without sacrificing accuracy. Our method simplifies the process of selecting features by reducing feature multicollinearity and enables discoveries of new relationships between global properties and molecular descriptors. To demonstrate our approach, we developed models for predicting melting point, boiling point, flash point, yield sooting index, and net heat of combustion with the help of the Tree-based Pipeline Optimization Tool (TPOT). For training, we used publicly available experimental data for up to 8351 molecules. Our models accurately predict various molecular properties for organic molecules (mean absolute percent error (MAPE) ranges from 3.3% to 10.5%) and provide a set of features that are well-correlated to the property. This method enables researchers to explore sets of features that significantly contribute to the prediction of the property, offering new scientific insights. To help accelerate early stage biofuel research and development, we also integrated the data and models into a open-source, interactive web tool.

09 BIOMASS FUELS↗

DOE OSTI2025

A Bayesian framework to investigate radiation reaction in strong fields

Recent experiments aiming to measure phenomena predicted by strong-field quantum electrodynamics (SFQED) have done so by colliding relativistic electron beams and high-power lasers. In such experiments, measurements of collision parameters are not always feasible. However, precise knowledge of these parameters is required to accurately test SFQED. Here, we present a novel Bayesian inference procedure that infers collision parameters that could not be measured on-shot. This procedure is applicable to all-optical non-linear Compton scattering experiments investigating radiation reaction. The framework allows multiple diagnostics to be combined self-consistently and facilitates the inclusion of known information pertaining to the collision parameters. Using this Bayesian analysis, the relative validity of the classical, quantum-continuous and quantum-stochastic models of radiation reaction was compared for several test cases, which demonstrates the accuracy and model selection capability of the framework and highlight its robustness if the experimental values of fixed parameters differ from their values in the models.

47 OTHER INSTRUMENTATION↗

DOE OSTI2024

Differentiable modeling and optimization of non-aqueous Li-based battery electrolyte solutions using geometric deep learning

Electrolytes play a critical role in designing next-generation battery systems, by allowing efficient ion transfer, preventing charge transfer, and stabilizing electrode-electrolyte interfaces. In this work, we develop a differentiable geometric deep learning (GDL) model for chemical mixtures, DiffMix, which is applied in guiding robotic experimentation and optimization towards fast charging battery electrolytes. In particular, we extend mixture thermodynamic and transport laws by creating GDL-learnable physical coefficients. We evaluate our model with mixture thermodynamics and ion transport properties, where we show improved prediction accuracy and model robustness of Diff-Mix than its purely data-driven variants. Furthermore, with a robotic experimentation setup, Clio, we improve ionic conductivity of electrolytes by over 18.8% within 10 experimental steps, via differentiable optimization built on DiffMix gradients. By combining GDL, mixture physics laws, and robotic experimentation, DiffMix expands the predictive modeling methods for chemical mixtures and enables efficient optimization in large chemical spaces.

25 - ENERGY STORAGE↗

DOE OSTI2024

Deep learning-enabled probing of irradiation-induced defects in time-series micrographs

Modeling time-series data with convolutional neural networks (CNNs) requires building a model to learn in batches as opposed to training sequentially. Coupling CNNs with in situ or operando techniques opens the possibility of accurately segmenting dynamic reactions and mass transport phenomena to understand how materials behave under the conditions in which they are used. In this article, in situ ion irradiation transmission electron microscopy (TEM) images are used as inputs into the CNN to assess the defect generation rate, defect cluster density, and saturation of defects. We then use the output segmentation maps to correlate with conventional TEM micrographs to assess the model’s ability to detail nanoscale interactions. Next, we discuss the implications of preprocessing and hyperparameters on model variability, accuracy when expanded to other datasets, and the role of regularization when controlling model variance. Ultimately, we eliminate human bias when extrapolating physical metrics, speed up analysis time, decouple reactions that happen at 100 ms intervals, and deploy models that are both accurate and transferable to similar experiments.

Burns, Kory (ORCID:0000000158019909)↗

DOE OSTI2025

Leveraging Artificial Intelligence to Predict Novel Eutectic Alloys

The goal of this project was to train an artificial neural network (ANN) to predict the fractional composition and melting point of eutectic alloys using fundamental atomic properties as inputs. The fundamental properties considered include atomic number, atomic weight, atomic radius, valence electron concentration, electronegativity, and electron affinity. The project involved several phases, starting with data preparation, where phase diagram data was harvested from the ASM International database. Approximately 1300 binary eutectics were collected and cleaned to ensure relevance and accuracy. A regression model was selected for training, utilizing a rectified linear unit as the activation function. Various model configurations were evaluated for predictive accuracy, with validation techniques employed to ensure robustness. The model demonstrated predictive capabilities above random guessing and was able to achieve up to 11% accuracy under certain conditions. An ablative test identified atomic radius and valence electron concentration as critical inputs for model performance. Incorporating the melting point of atomic constituents improved accuracy significantly, although ultimately the model’s predictive capability still fell short of the 80% target. This report details the methodology, results, and implications of the research, contributing to the understanding of employing artificial intelligence to predict the phase transition behavior of eutectic alloys.

36 MATERIALS SCIENCE↗

DOE OSTI2024

Quantifying the Influence of Size, Shape, and Density of Microplastics on Their Transport Modes: A Modeling Approach

Microplastics (MPs) pose significant risks to marine ecosystems and human health, necessitating accurate predictions of their distributions in aquatic environments for effective risk mitigation. However, understanding MP transport dynamics is challenging because of the inadequate representation of MP characteristics such as size, shape, and density in numerical models. Further, the accuracy of the MP vertical profiles in existing models has not been thoroughly validated. Thus, we developed an MP transport model within the Finite Volume Community Ocean Model framework (FVCOM-MP) by integrating MP characteristics. We validated FVCOM-MP against experimental and analytical data, focusing on various MP transport modes and transitions. FVCOM-MP successfully replicates MP profiles in different transport modes, including the bedload, surface-load, suspended-load, and mixed-load modes. Additionally, we introduce phase diagrams for classifying MP transport modes based on particle characteristics, enhancing our understanding of MP dynamics in aquatic systems. The transport modes for a number of real-world MP particles, including fishing line, plastic bag/bottle fragments, synthetic fibers, tire wear particles, polyvinyl chloride and expanded polystyrene pellets, were analyzed with these phase diagrams.

Microplastic transport, Settling velocity, Rising ↗

DOE OSTI2024

A framework for strategic discovery of credible neural network surrogate models under uncertainty

The widespread integration of deep neural networks in developing data-driven surrogate models for high-fidelity simulations of complex physical systems highlights the critical necessity for robust uncertainty quantification techniques and credibility assessment methodologies, ensuring the reliable deployment of surrogate models in consequential decision-making. Here, this study presents the Occam Plausibility Algorithm for surrogate models (OPAL-surrogate), providing a systematic framework to uncover predictive neural network-based surrogate models within the large space of potential models, including various neural network classes and choices of architecture and hyperparameters. The framework is grounded in hierarchical Bayesian inferences and employs model validation tests to evaluate the credibility and prediction reliability of the surrogate models under uncertainty. Leveraging these principles, OPAL-surrogate introduces a systematic and efficient strategy for balancing the trade-off between model complexity, accuracy, and prediction uncertainty. The effectiveness of OPAL-surrogate is demonstrated through two modeling problems, including the deformation of porous materials for building insulation and turbulent combustion flow for ablation of solid fuels within hybrid rocket motors.

42 ENGINEERING↗

DOE OSTI2024

Neural network representations of multiphase Equations of State

Abstract Equations of State model relations between thermodynamic variables and are ubiquitous in scientific modelling, appearing in modern day applications ranging from Astrophysics to Climate Science. The three desired properties of a general Equation of State model are adherence to the Laws of Thermodynamics, incorporation of phase transitions, and multiscale accuracy. Analytic models that adhere to all three are hard to develop and cumbersome to work with, often resulting in sacrificing one of these elements for the sake of efficiency. In this work, two deep-learning methods are proposed that provably satisfy the first and second conditions on a large-enough region of thermodynamic variable space. The first is based on learning the generating function (thermodynamic potential) while the second is based on structure-preserving, symplectic neural networks, respectively allowing modifications near or on phase transition regions. They can be used either “from scratch” to learn a full Equation of State, or in conjunction with a pre-existing consistent model, functioning as a modification that better adheres to experimental data. We formulate the theory and provide several computational examples to justify both approaches, highlighting their advantages and shortcomings.

Science & Technology - Other Topics↗

NASA NTRS2014

Toward Improved Land Surface Initialization in Support of Regional WRF Forecasts at the Kenya Meteorological Service (KMS)

Flooding and drought are two key forecasting challenges for the Kenya Meteorological Service (KMS). Atmospheric processes leading to excessive precipitation and/or prolonged drought can be quite sensitive to the state of the land surface, which interacts with the planetary boundary layer (PBL) of the atmosphere providing a source of heat and moisture. The development and evolution of precipitation systems are affected by heat and moisture fluxes from the land surface, particularly within weakly-sheared environments such as in the tropics and sub-tropics. These heat and moisture fluxes during the day can be strongly influenced by land cover, vegetation, and soil moisture content. Therefore, it is important to represent the land surface state as accurately as possible in land surface and numerical weather prediction (NWP) models. Enhanced regional modeling capabilities have the potential to improve forecast guidance in support of daily operations and high-impact weather over eastern Africa. KMS currently runs a configuration of the Weather Research and Forecasting (WRF) NWP model in real time to support its daily forecasting operations, making use of the NOAA/National Weather Service (NWS) Science and Training Resource Center's Environmental Modeling System (EMS) to manage and produce the KMS-WRF runs on a regional grid over eastern Africa. Two organizations at the NASA Marshall Space Flight Center in Huntsville, AL, SERVIR and the Shortterm Prediction Research and Transition (SPoRT) Center, have established a working partnership with KMS for enhancing its regional modeling capabilities through new datasets and tools. To accomplish this goal, SPoRT and SERVIR is providing enhanced, experimental land surface initialization datasets and model verification capabilities to KMS as part of this collaboration. To produce a land-surface initialization more consistent with the resolution of the KMS-WRF runs, the NASA Land Information System (LIS) is run at a comparable resolution to provide real-time, daily soil initialization data in place of data interpolated from the National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS) model soil moisture and temperature fields. Additionally, realtime green vegetation fraction (GVF) data from the Visible Infrared Imaging Radiometer Suite (VIIRS) on the Suomi National Polar-orbiting Partnership (Suomi- NPP) satellite will be incorporated into the KMS-WRF runs, once it becomes publicly available from the National Environmental Satellite Data and Information Service (NESDIS). Finally, model verification capabilities will be transitioned to KMS using the Model Evaluation Tools (MET; Brown et al. 2009) package in conjunction with a dynamic scripting package developed by SPoRT (Zavodsky et al. 2014), to help quantify possible improvements in simulated temperature, moisture and precipitation resulting from the experimental land surface initialization. Furthermore, the transition of these MET tools will enable KMS to monitor model forecast accuracy in near real time. This paper presents preliminary efforts to improve land surface model initialization over eastern Africa in support of operations at KMS. The remainder of this extended abstract is organized as follows: The collaborating organizations involved in the project are described in Section 2; background information on LIS and the configuration for eastern Africa is presented in Section 3; the WRF configuration used in this modeling experiment is described in Section 4; sample experimental WRF output with and without LIS initialization data are given in Section 5; a summary is given in Section 6 followed by acknowledgements and references.

Case, Johnathan L.↗

NASA NTRSDate not supplied

SMOS Salinity Retrieved from New Seawater Dielectric Constant Models at L-band

The accuracy of the Sea Surface Salinity (SSS) retrieved from L-Band radiometer measurements is strongly dependent on the accuracy of the modelling of the dielectric constant. Two new parametrizations have recently been developed based on one hand on the Soil Moisture and Ocean Salinity (SMOS) satellite multi-angular brightness temperature measurements [1] (BV) and on the other hand on new laboratory measurements [3] (GW2020). These two approaches are fully independent. The brightness temperatures, Tb, simulated with the BV and GW2020 parametrizations are compared with each other and with the ones derived from dielectric constant models previously in use in the SMOS, Soil Moisture Active Passive (SMAP) and Aquarius SSS retrievals. Tb simulated with the BV and GW2020 parametrizations agree particularly well for most SSS and SST commonly observed over the open ocean and are found to be in closer agreement than with earlier parametrizations. Nevertheless, uncertainty remains at low SST where a ∼ 0.1 K relative difference between the two models is observed. A complete reprocessing of SMOS SSS (2010–2020) has been performed using the BV parametrization instead of the Klein and Swift (1977) model previously used in SMOS processing. When compared with Argo derived near surface salinity maps, clear improvements are observed in warm and cold regions. Remaining uncertainties in cold waters will be discussed relatively to the uncertainties in SMOS Tb linked to sea ice contamination and given the constraints given by the GW2020 laboratory measurements.

J. Boutin↗

NASA NTRSDate not supplied

Laser Absorption Spectroscopy Measurements of Post-Shock Non-Equilibrium Species in the NASA Ames Electric Arc Shock Tube

The NASA Ames Electric Arc Shock Tube (EAST) is a unique facility capable of generating high enthalpy, shock-heated impulse gas flows representative of the kinetic and radiative conditions encountered by atmospheric entry vehicles. The facility maintains a successful history with emission spectroscopy, providing measurements of absolute radiance upon which various validation studies are anchored. The central objective of this work is to develop a comprehensive tunable diode laser absorption spectroscopy (TDLAS) sensing capability to provide complementary experimental insights to existing emission techniques. More specifically, a fast-scanning TDLAS-based diagnostic targeting the cyano radical (CN) in the near-infrared(near-IR) near 926.6 nm is employed in mixtures of 2.2% CH4in N2by mole for a sweep of incident shock velocities ranging 3.0-5.5 km/s and fill pressures 0.3-1.15 Torr. Laser scan rates up to 500 kHz probe multiple absorption features to provide quantitative measurements of temperature and species number density profiles behind the incident shock. In all instances, the temperature trend is consistent between emission-inferred and TDLAS-inferred measurements. However, at the lower-velocity conditions, there is an appreciable offset between the Data Parallel Line Relaxation Code (DPLR) simulation and experimental measurements. A similar behavior is observed in number density profiles, with emission measurements providing inferences of the CN(B) and CN(A) state, while TDLAS measurements provide an additional inference of the CN(X) state. In the fast-scanned experiments, measurement inferred number densities suggest a much faster production of CN at early times than captured by the model. This discrepancy motivates ongoing, additional spectroscopic and kinetics CN experiments to resolve the accuracy of modeling assumptions.

Efaine Chang↗

NASA NTRSDate not supplied

Foundation AI Models for Science

Foundation Models (FM) are AI models that are designed to replace a task or an application specific model. These FM can be applied to many different downstream applications. These FM are trained using self supervised techniques and can be built on any type of sequence data. The use of self supervised learning removes the hurdle for developing a large labeled dataset for training. Most FM use transformer architecture utilizes the notion of self attention which allows the network to model the influence of distant data points to each other both in space and time. The FM models exhibit emergent properties that are induced from the data. FM can be an important tool for science. The scale of these models results in better performance for different downstream applications and these applications show better accuracy over models built from scratch. FM drastically reduces the cost of entry to build different downstream applications both in time and effort. FM for selected science datasets such as optical satellite data, can accelerate applications ranging from data quality monitoring, feature detection and prediction. FM can make it easier to infuse AI into scientific research by removing the training data bottleneck and increasing the use of science data.

Manil Maskey↗

NASA NTRSDate not supplied

Revolutionizing Earth Science with Generalized AI Models

Foundation Models (FM) are generalized Artificial Intelligence (AI) models that are designed to replace a task or an application-specific model and can be used for many downstream applications. These FM can be built on any sequence data and are trained utilizing self-supervised approaches. The obstacle of creating a sizable labeled dataset for training is removed by using self-supervised learning. Most FM employ transformer design that takes advantage of the idea of self-attention, allowing the network to represent the impact of distant data points on one another in space and time. The FM models show emergent qualities that are induced from the data. FM can become a valuable tool for Earth science researchers. Due to the size of these models, downstream applications built fine-tuning these FM perform better and exhibit greater accuracy than models created from scratch. FM significantly lowers the entry barrier in terms of both the time and effort required to develop various downstream applications. For some scientific datasets, such as optical remote sensing data, FM can speed up processes like classification, object detection and prediction. By eliminating the training data bottleneck and maximizing the usage of science data, FM can make it simpler to integrate AI into scientific research. Initial results for three different FMs will be presented.

Rahul Ramachandran↗