Search NASA⌕ Search

SEARCH · Search NASA

Results for “ML model”

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.

At least 1,045 records · Page 58

Applying deep learning methods to develop new models of molecular charge transfer, nonadiabatic dynamics, and nonlinear spectroscopy in the condensed phase

Photon- and field-induced charge transfer has central importance in the generation and storage of electricity, the novel properties of materials, photo-induced catalysis, and electro-optic activity (e.g., photovoltaic cells, fuel cells, and organic chromophores for use in optical fibers and light-emission diodes). These non-equilibrium electronic and chemical transformations are probed by ultrafast, nonlinear spectroscopies. Accurate simulations play a crucial role in our ability to understand, optimize, and control these transformations. This project applies modern deep learning and machine learning (ML) methods to dramatically improve models of electronic dynamics, electronic-nuclear dynamics, and spectroscopic measurements for improved simulations of chemistry in complex environments, far from equilibrium phenomena, and processes in extreme environments, such as materials exposed to strong or resonant fields. This project develops accurate neural net models that go beyond predictive capability to also provide new insight into the fundamental physics underlying electron and nuclear dynamics. To achieve its objectives, this project explores and develops customized versions of high-capacity deep learning algorithms/models. These techniques are developed with an emphasis on fundamental chemical insight, not just predictive accuracy, to assist the development of the next generation of quantum simulation methods.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Machine learning-based bias-corrected future projections of ozone concentrations from a chemistry-climate model

Reliable projection of future near-surface ozone is crucial for air quality management and health risk assessment. However, potential biases in spatial distribution, magnitude and trends in ozone concentrations simulated by global chemistry-climate models limit their applicability in regional-scale evaluations. In this study, LightGBM, a machine learning (ML) algorithm is applied to correct biases in CESM2-simulated ozone concentrations over China, the United States and Europe and calibrate future ozone projections under two diverse Shared Socioeconomic Pathways (SSP1-2.6 and SSP5-8.5) scenarios from 2020 to 2060. The ML-based correction significantly improves the spatial distribution and reduces the model bias by 40%–60%. It also reverses the potentially incorrect trend of ozone change under SSP1-2.6 in eastern China. When applying ML-based bias correction to CESM2 future projections, warm season mean ozone concentrations decrease across China, the United States, and Europe by –13.5, –17.9, and –13.7 µg/m³, respectively, between 2020 and 2060 in SSP1-2.6, while they increase by 9.4, 2.0, and 5.2 µg/m³ in SSP5-8.5. Decomposition analysis show that changes in anthropogenic emissions dominate future ozone changes in both scenarios, while strong climate penalty from ozone changes occurs in polluted eastern China and climate benefit is found in western China, the United States and Europe under SSP5-8.5. These findings demonstrate the value of combining ML with chemistry-climate models to produce more accurate air quality projections, thereby informing more effective and region-specific environmental protection strategies.

Chemistry Model↗

Flow Redirection and Induction in Steady State (FLORIS) Wind Plant Power Production Data Sets

This dataset contains turbine- and plant-level power outputs for 252,500 cases of diverse wind plant layouts operating under a wide range of yawing and atmospheric conditions. The power outputs were computed using the Gaussian wake model in NREL's FLOw Redirection and Induction in Steady State (FLORIS) model, version 2.3.0. The 252,500 cases include 500 unique wind plants generated randomly by a specialized Plant Layout Generator (PLayGen) that samples randomized realizations of wind plant layouts from one of four canonical configurations: (i) cluster, (ii) single string, (iii) multiple string, (iv) parallel string. Other wind plant layout parameters were also randomly sampled, including the number of turbines (25-200) and the mean turbine spacing (3D-10D, where D denotes the turbine rotor diameter). For each layout, 500 different sets of atmospheric conditions were randomly sampled. These include wind speed in 0-25 m/s, wind direction in 0 deg.-360 deg., and turbulence intensity chosen from low (6%), medium (8%), and high (10%). For each atmospheric inflow scenario, the individual turbine yaw angles were randomly sampled from a one-sided truncated Gaussian on the interval 0 deg.-30 deg. oriented relative to wind inflow direction. This random data is supplemented with a collection of yaw-optimized samples where FLORIS was used to determine turbine yaw angles that maximize power production for the entire plant. To generate this data, a subset of cases were selected (50 atmospheric conditions from 50 layouts each for a total of additional 2,500 cases) for which FLORIS was re-run with wake steering control optimization. The IEA onshore reference turbine, which has a 130 m rotor diameter, a 110 m hub height, and a rated power capacity of 3.4 MW was used as the turbine for all simulations. The simulations were performed using NREL's Eagle high performance computing system in February 2021 as part of the Spatial Analysis for Wind Technology Development project funded by the U.S. Department of Energy Wind Energy Technologies Office. The data was collected, reformatted, and preprocessed for this OEDI submission in May 2023 under the Foundational AI for Wind Energy project funded by the U.S. Department of Energy Wind Energy Technologies Office. This dataset is intended to serve as a benchmark against which new artificial intelligence (AI) or machine learning (ML) tools may be tested. Baseline AI/ML methods for analyzing this dataset have been implemented, and a link to their repository containing those models has been provided. The .h5 data file structure can be found in the GitHub repository under explore_wind_plant_data_h5.ipynb.

AI↗

Automated Discovery of DOminaNt physics Informed Surrogates (ADDONIS) Framework for Improving Water Cycling Predictability

Focal Areas: The primary alignment of the proposed framework is with focal area 2, because it guides the construction of a hierarchy of predictive models generated from Statistical and AI/ ML techniques. There are connections to focal areas 1 and 3 because surrogates allow for coupled data assimilation techniques where the full model is too expensive (1) and the discovered balance laws can be used as an insightful analysis tool to validate the structure of models (3).

54 ENVIRONMENTAL SCIENCES↗

A Comparison of Machine Learning Methods of Association Tested on Dense Nodal Arrays

The association of phase picks to form events is one of the fundamental components of seismology. Large and dense sensor networks, such as >1000 geophone arrays (and distributed acoustic sensing), offer unique challenges in association due to the vast numbers of observations and high likelihood of errant picks. In addition, the large number of stations can greatly increase the time it takes to perform the association. For this reason, machine learning (ML) methods might provide a more optimal method of association for such networks. In this work, we examine how well ML methods (e.g., Gaussian mixture model association, PhaseLink, and Graph Earthquake Neural Interpretation Engine) can incorporate dense seismic arrays into regional networks and how well they handle the increasing numbers of stations. Here, we test their capabilities on two dense seismic deployments, one within Rock Valley Nevada (52 nodes and a 9-station sparse local network), and the LArge-n Seismic Survey in Oklahoma dense nodal array (>1800 vertical-component geophones). Processing data from these two different styles of dense seismic deployments allows testing of how the ML algorithms can merge array data with a broader regional network, how they deal with poorly picked phases, and how they handle anthropogenic noise. We compare the ML-associated bulletins to those obtained using the Rapid Earthquake Association and Location algorithm, a more traditional method of association. We find that there are very small differences in results between the methods for small networks (<100 stations) with low pick rates. For large networks (>1000), there are enough errant picks that some of the ML methods start to create false events out of noise. We also find that the ML methods vary in computation time significantly but are all faster than the traditional method tested here.

58 GEOSCIENCES↗

Open Science for Plants in Space: Improvements in NASA's Open Science Data Repository

Upcoming deep space missions will rely on plants for crew and ecosystem health. Open access space biology data enables scientists to examine the biological responses of plants to ionizing radiation, altered gravity, elevated CO2, and many other abiotic stressors. NASA has declared 2023 as the ‘Year of Open Science’ and created a 5-year Transform to Open Science (TOPS) initiative designed to rapidly transform the agency toward an inclusive culture of open science. NASA’s Open Science Data Repository (OSDR) combines two databases, GeneLab and Ames Life Sciences Data Archive (ALSDA) to maximize access to standardized ‘omics (e.g., transcriptomics, proteomics) and phenotypic data (e.g., microscopy, biomass), respectively. Current OSDR standards include the ISA (Investigation-Study-Assay) experiment model, assay metadata configurations, and standardized terminology and ontologies. In 2024 OSDR will include a new suite of features for improved FAIR compliance including downloadable plant metadata templates, data submission tools and overall improved AI-readiness of plant datasets. AI/ML methods can be helpful tools to overcome the inherent challenges of space biology research (small sample size, sparse and heterogeneous data etc.). However these methods are built on an assumption of normalized and well-curated data. OSDR’s new curation tools will improve users ability to leverage ML and AI methods to model space biology data and better understand the complex effects of spaceflight on living systems across hierarchical biological levels. We look forward to sharing our advances with the spaceflight community.

FAIR↗

GeoThermalCloud: Cloud Fusion of Big Data and Multi-Physics Models using Machine Learning for Discovery, Exploration, and Development of Hidden Geothermal Resources

The primary goals of this project are exploring hidden geothermal resources in the U.S.A. and designing profitable enhanced geothermal systems (EGS). Many processes and parameters control geothermal exploration and energy production from geothermal fields. Diverse datasets (e.g., geology, geochemistry, geophysics, satellite, airborne geophysics) are available to help characterize subsurface geothermal conditions. Sparse and multi-scale characteristics of these datasets prohibit properly leveraging these datasets for geothermal exploration and profitable EGS design. Recent advancements in machine learning (ML) promise to resolve these issues. The tremendous challenges and risks of geothermal exploration and production bring the demand for novel ML methods and tools that can (1) analyze large field datasets, (2) assimilate model simulations (large inputs and outputs), (3) process sparse datasets, (4) perform transfer learning (between sites with different exploratory levels), (5) extract hidden geothermal signatures in the field and simulation data, (6) label geothermal resources and processes, (7) identify high-value data acquisition targets, and (8) guide geothermal exploration and production by selecting optimal exploration, production, and drilling strategies. To address these necessities, ML-based geothermal resources exploration and enhanced geothermal systems (EGS) design tools have been developed. The exploration tool is called GeoThermalCloud and EGS design tool is called GeoDT-ML. GeoThermalCloud (https://github.com/SmartTensors/GeoThermalCloud.jl) utilizes a LANL unsupervised ML platform called SmartTensors (https://tensors.lanl.gov/) to automate data analyses and interpretations by extracting hidden signatures to identify geothermal prospects. Also, it enables the identification of critical measurements needed to identify geothermal resource signatures. Alternatively, GeoDT-ML (https://github.com/SmartTensors/GeoThermalCloud.jl/tree/master/EGS) is an ML-based alternative to GeoDT (https://github.com/GeoDesignTool/GeoDT.git), a fast, simplified multi-physics solver to evaluate EGS project designs in uncertain geologic systems. GeoDT-ML leverages recent advances in deep learning and high-performance computing. It is a faster and simpler version of GeoDT. To make this project a success, we used capabilities of LANL, PNNL, Google, Stanford, and Julia Computing. We analyzed eight datasets of the U.S.A. using GeothermalCloud and demonstrated potential highly prospective geothermal resources and identified key factors defining highly prospective sites. The first data set includes 44 locations in southwest New Mexico and 18 geological, hydrogeological, geophysical, geothermal, geochemical attributes. We defined low- and medium-temperature hydrothermal systems and discovered a new highly prospective site. The second data set analyzed 18 shallow water chemistry attributes at 14,342 locations in the Great Basin. It demarcated modestly, moderately, and highly prospective sites including key attributes for each type of prospectivity. The third data set analyzed Utah FORGE data including satellite (InSAR), geophysical (gravity, seismic), geochemical, and geothermal attributes. Here, we performed prospectivity analysis to identify future drilling locations using geological, geochemical, and geophysical attributes. Maps of temperature at depth and heat flow are constructed based on the available data. Prospectivity maps were generated, and drilling locations were proposed for future geothermal field exploration. The fourth data set analyzed 21 attributes at 120 locations in Tularosa Basin, New Mexico; data comes from past play fairway analyses in this region. ML analyses identified geothermal signatures associated with modestly, moderately, and highly hydrothermal systems. We also defined dominant attributes and spatial distribution of the geothermal signatures. The fifth, sixth, seventh, and eighth datasets include Tohatchi Springs, New Mexico, Hawaii, Brady site, Nevada, and EGS Collab, respectively. Moreover, we coupled GeothermalCloud and magnetotellurics data to pinpoint drilling locations for developing geothermal projects in the Tularosa Basin, New Mexico. GeothermalCloud found potential prospective locations for geothermal resources near White Sands Missile Range and McGregor Range at Fort Bliss. Magnetotellurics data determined the potential depth (~1800m) of geothermal prospects at McGregor Range based on apparent resistivity structures/layers in the subsurface. The McGregor Range consists of three resistivity layers and two resistivity structures. Magnetotellurics data also helps identify that the western portion of the McGregor Range has thick and low-resistivity earth materials. The low resistivity to the west is most likely for a fault system. Assuming temperature is consistent with a geothermal reservoir, the west-central part of the McGregor Range has the highest geothermal potential because of the increase in porosity and associated permeability attributed to the interpreted fault system. Also, we devised a coupling strategy between a process model and GeothermalCloud to characterize hydrogeological conditions and geothermal conditions, respectively. The process model characterizes hydrogeological and geothermal conditions on highly prospective geothermal sites provided by GeothermalCloud. We developed a physics-informed neural network (PINN) version of the Burns equation that can be easily coupled with GeothermalCloud. Furthermore, we performed an optimal design decision maximizing the economic value of an EGS power plant. This study optimized the range of well spacing between injection and production wells maximizing net present value in dollars (NPV). For this task, we used the GeoDT to simulate the Utah FORGE EGS development cycle from the initial well design to the end of production. Next, we accomplished another crucial task, which is predicting permeability of geothermal reservoirs. Predicting permeability of geothermal reservoirs is a non-trivial task because of huge computational runtime of simulation and lack of measurements. To avoid these limitations, we used easy-to-measure chemical concentrations in the subsurface as measurement data and convolutional neural network based ML model of a high-fidelity model. Next, we predicted permeability using Markov chain Monte Carlo simulation. We found that Markov chain Monte Carlo simulation predicts permeability with a high certainty if the prediction zone in the simulation area has chemical concentration data. Finally, we analyzed the DOE funded INGENIOUS and GeoDAWN projects data. For discovering hidden geothermal systems in the Great Basin, the INGENIOUS project accumulated old data, collected new data, and released them in 2022. The dataset includes a total of 24 geological, geophysical, and geochemical attributes. Data resolution and scale significantly vary prohibiting an appropriate usage. To avoid such limitations, we brought all data in the same resolution and scale by applying the inverse distance weighting interpolation technique for predicting data in unsampled locations. Subsequently, we analyzed LiDAR data of the GeoDAWN project. We received data in tiles format. The DOE’s overarching goal is to use ML on LiDAR data for finding favorable geological structures (e.g., step up faults in Brady, Nevada). To serve the purpose, we need to label favorable geologic structures that correspond to LiDAR data. We wrote an algorithm to label the LiDAR data with the favorable geologic structures.

15 GEOTHERMAL ENERGY↗

Efficient Interdependent Systems Recovery Modeling with DeepONets

Modeling the recovery of interdependent critical infrastructure is a key component of quantifying and optimizing societal resilience to disruptive events. However, simulating the recovery of large-scale interdependent systems under random disruptive events is computationally expensive. Therefore, we propose the application of Deep Operator Networks (DeepONets) in this paper to accelerate the recovery modeling of interdependent systems. DeepONets are ML architectures which identify mathematical operators from data. The form of governing equations DeepONets identify and the governing equation of interdependent systems recovery model are similar. Therefore, we hypothesize that DeepONets can efficiently model the interdependent systems recovery with little training data. We applied DeepONets to a simple case of four interdependent systems with sixteen states. DeepONets, overall, performed satisfactorily in predicting the recovery of these interdependent systems for out of training sample data when compared to reference results.

97 MATHEMATICS AND COMPUTING↗

Exploring model complexity in machine learned potentials for simulated properties

Abstract Machine learning (ML) enables the development of interatomic potentials with the accuracy of first principles methods while retaining the speed and parallel efficiency of empirical potentials. While ML potentials traditionally use atom-centered descriptors as inputs, different models such as linear regression and neural networks map descriptors to atomic energies and forces. This begs the question: what is the improvement in accuracy due to model complexity irrespective of descriptors? We curate three datasets to investigate this question in terms of ab initio energy and force errors: (1) solid and liquid silicon, (2) gallium nitride, and (3) the superionic conductor Li $$_{10}$$ 10 Ge(PS $$_{6}$$ 6 ) $$_{2}$$ 2 (LGPS). We further investigate how these errors affect simulated properties and verify if the improvement in fitting errors corresponds to measurable improvement in property prediction. By assessing different models, we observe correlations between fitting quantity (e.g. atomic force) error and simulated property error with respect to ab initio values. Graphical abstract

Rohskopf, A. (ORCID:0000000227128296)↗

Classifying and analyzing small-angle scattering data using weighted k nearest neighbors machine learning techniques

A consistent challenge for both new and expert practitioners of small-angle scattering (SAS) lies in determining how to analyze the data, given the limited information content of said data and the large number of models that can be employed. Machine learning (ML) methods are powerful tools for classifying data that have found diverse applications in many fields of science. Here, ML methods are applied to the problem of classifying SAS data for the most appropriate model to use for data analysis. The approach employed is built around the method of weighted k nearest neighbors (wKNN), and utilizes a subset of the models implemented in the SasView package (https://www.sasview.org/) for generating a well defined set of training and testing data. The prediction rate of the wKNN method implemented here using a subset of SasView models is reasonably good for many of the models, but has difficulty with others, notably those based on spherical structures. A novel expansion of the wKNN method was also developed, which uses Gaussian processes to produce local surrogate models for the classification, and this significantly improves the classification accuracy. Further, by integrating a stochastic gradient descent method during post-processing, it is possible to leverage the local surrogate model both to classify the SAS data with high accuracy and to predict the structural parameters that best describe the data. The linking of data classification and model fitting has the potential to facilitate the translation of measured data into results for both novice and expert practitioners of SAS.

97 MATHEMATICS AND COMPUTING↗

Climate-invariant machine learning

Projecting climate change is a generalization problem: We extrapolate the recent past using physical models across past, present, and future climates. Current climate models require representations of processes that occur at scales smaller than model grid size, which have been the main source of model projection uncertainty. Recent machine learning (ML) algorithms hold promise to improve such process representations but tend to extrapolate poorly to climate regimes that they were not trained on. To get the best of the physical and statistical worlds, we propose a framework, termed “climate-invariant” ML, incorporating knowledge of climate processes into ML algorithms, and show that it can maintain high offline accuracy across a wide range of climate conditions and configurations in three distinct atmospheric models. Our results suggest that explicitly incorporating physical knowledge into data-driven models of Earth system processes can improve their consistency, data efficiency, and generalizability across climate regimes.

54 ENVIRONMENTAL SCIENCES↗

Interpretable Machine Learning Models for Autonomous Characterization of Analogue Ocean World Seawater Chemistry and Biosignature Potential Using Isotope Ratio Data

Background: Future missions to ocean worlds, such as Enceladus and Europa, will attempt to characterize the subsurface seawater chemistry and assess the potential for life. Such missions will be equipped with capabilities to precisely measure volatile isotopes in plumes, atmospheres, and exospheres. Motivation: While large isotopic fractionations can indicate a biological source, there are signatures resulting from abiotic geochemical processes that mimic isotopic biosignatures. While machine learning (ML) has the potential to disentangle competing effects and biotic mimicry, high-dimensional isotope ratio mass spectrometry (IRMS) data is likely to contain noise/irrelevant features and involve complex statistical interactions that make human inference and interpretation difficult. Further, ML predictions with as far-reaching implications as an extraterrestrial biosignature on an ocean world requires the use of interpretable models (i.e., not “black box” models) with physically and mathematically meaningful feature spaces along with false positive diagnostics. Methods: We use volatile CO2 IRMS data of analogue ocean world seawaters to validate an ML approach to provide biogeochemical context for biosignature detection. We employ a feature selection method called nearest-neighbor projected distance regression (NPDR) that detects statistical interactions and helps elucidate the mechanisms of the Random Forest classification models. Results: We train and validate predictive ML models on volatile CO2 IRMS data of analogue ocean world seawaters to predict major salt components (e.g., MgSO4, NaHCO3), pH, ionic strength, and the presence of biosignatures. Features derived from IRMS measurements are augmented with extracted time-series features. Our results show high test accuracy and interpretability, which is increased by interaction network visualization, sample-wise variable importance scores, and single-sample class probability estimates. We demonstrate an ML mission software solution that triggers autonomous data transmission and biogeochemical sample prediction.

geochemistry↗

Surface Complexation/Ion Exchange Hybrid Model for Radionuclide Sorption to Clay Minerals (M4SF-23LL010301062)

This progress report (Level 4 Milestone Number M4SF-23LL010301062) summarizes research conducted at Lawrence Livermore National Laboratory (LLNL) within the Argillite International Collaborations Activity Number SF-23LL01030106. The activity is focused on our long-term commitment to engaging our partners in international nuclear waste repository research. The focus of this milestone is the establishment of international collaborations for surface complexation modeling and the associated impacts of unlocking larger, community-based datasets. More specifically, we are developing a database framework for Spent Fuel and Waste and Science Technology (SFWST) that is aligned with the Helmholtz Zentrum Dresden Rossendorf (HZDR) sorption database development group in support of the database needs of the SFWST program. In our FY22 effort, we described a detailed analysis of U(VI) sorption to quartz through both traditional surface complexation modeling and through a hybrid ML framework. In FY23, effort was placed on publication of these results and expansion of the LLNL surface complexation and ion exchange database (L-SCIE) in order to assess mineral-based radionuclide retardation under a wider variety of geochemical conditions (e.g., ionic strength, varying electrolyte compositions). Efforts were initiated to expand L-SCIE to include radionuclide surface complexation and ion exchange to clays that are relevant to subsurface geochemical processes occurring at nuclear waste repositories. In particular, a large source of sorption data for clays resides at the Paul Scherrer Institute (PSI) (work primarily by Bradbury and Baeyens) and we initiated discussions on how to retrieve those data and apply FAIR principles to those datasets. In addition to L-SCIE development, two hybrid models that incorporate AI/ML were investigated and compared to discern the most promising approaches for accurate and precise estimations of radionuclide retardation. Key considerations for future model development include (1) the ability to reduce computational burden on determining retardation coefficients for PA and (2) the ability to quantify and predict radionuclide-mineral partitioning at a more efficient, rapid pace due to automated workflows. Upon the careful consideration of the most effective modeling approaches, we are identifying ways to implement these approaches into PA. Ultimately, the data science-based workflows will provide a major incentive for other institutions to adopt a FAIR-formatted, interoperable database. LLNL will play a key role in disseminating sorption data and acting as good data stewards by updating the database in a consistent format and assessing the quality of the newly assimilated data in an organized fashion. To this end, all data and workflows are open access and made available on the LLNL Seaborg research website (https://seaborg.llnl.gov/resources/geochemical-databases-modeling-codes).

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗

A Decadal Hybrid GCM Simulation Using Deep‐Learning‐Based Cloud and Convection Parameterization Generalized to a Warm Climate

A critical challenge for machine‐learning (ML) parameterization in global climate models (GCMs) is to achieve stable, accurate simulations under climates not seen during training. Previous studies have demonstrated promising offline performance and year‐long online stability in aquaplanet simulations but have encountered difficulties in real geography and under climate warming. Here we report that a GCM with real geography configuration using neural‐network‐based cloud and convection parameterization, trained exclusively with present‐day climate data, successfully performs a stable, decade‐long simulation of a warm climate with +4 K sea surface temperature (SST). The neural network (NN) is based on Han et al. (2023, https://doi.org/10.1029/2022ms003508 ) with additional inputs. The simulation captures the global precipitation distribution, surface temperatures, vertical atmospheric structures, and extreme precipitation very well, closely matching simulations from both the superparameterized CAM (SPCAM) and the conventional CAM5 in the warm climate without accuracy degradation compared to those in the baseline climate. Moreover, it produces a climate response to +4 K SST in atmospheric thermodynamic states and circulations similar to those from SPCAM and CAM5. Prognostic ablation tests on NN input variables show that the NN without convective memory as input suffers from numerical instability, and the NN without considering radiative variables and land fraction as input, or with reduced training samples produce less accurate results. To our knowledge, this is the first time an ML parameterization successfully achieves online extrapolation to a warm climate without using additional warm‐climate data for training. It demonstrates the potential of ML‐driven parameterizations for credible long‐term climate projections.

Atmosphere model↗

Classification of machine learning frameworks for data-driven thermal fluid models

Thermal fluid processes are inherently multi-physics and multi-scale, involving mass-momentum-energy transport phenomena at multiple scales. Thermal fluid simulation (TFS) is based on solving conservative equations, for which – except for “first-principles” direct numerical simulation – closure relations (CRs) are required to provide microscopic interactions or so-called sub-grid-scale physics. In practice, TFS is realized through reduced-order modeling, and its CRs as low-fidelity models can be informed by observations and data from relevant and adequately evaluated experiments and high-fidelity simulations. This paper is focused on data-driven TFS models, specifically on their development using machine learning (ML). Five ML frameworks are introduced including physics-separated ML (PSML or Type I ML), physics-evaluated ML (PEML or Type II ML), physics-integrated ML (PIML or Type III ML), physics-recovered (PRML or Type IV ML), and physics-discovered ML (PDML or Type V ML). The frameworks vary in their performance for different applications depending on the level of knowledge of governing physics, source, type, amount and quality of available data for training. Notably, outlined for the first time in this paper, Type III models present stringent requirements on modeling, substantial computing resources for training, and high potential in extracting value from “big data” in thermal fluid research. Here the current paper demonstrates and investigates ML frameworks in three examples. First, we utilize the heat diffusion equation with a nonlinear conductivity model formulated by convolutional neural networks (CNNs) and feedforward neural networks (FNNs) to illustrate the applications of Type I, Type II, Type III, and Type V ML. The results indicate a preference for Type II ML under deficient data support. Type III ML can effectively utilize field data, potentially generating more robust predictions than Type I and Type II ML. CNN-based closures exhibit more predictability than FNN-based closures, but CNN-based closures require more training data to obtain accurate predictions. Second, we illustrate how to employ Type I ML and Type II ML frameworks for data-driven turbulence modeling using reference works. Third, we demonstrate Type I ML by building a deep FNN-based slip closure for two-phase flow modeling. The results show that deep FNN-based closures exhibit a bounded error in the prediction domain.

42 ENGINEERING↗

Normal Tissue Complication Probability Analysis of Acute Gastrointestinal Toxicity in Cervical Cancer Patients Undergoing Intensity Modulated Radiation Therapy and Concurrent Cisplatin

Purpose: To test the hypothesis that increased bowel radiation dose is associated with acute gastrointestinal (GI) toxicity in cervical cancer patients undergoing concurrent chemotherapy and intensity-modulated radiation therapy (IMRT), using a previously derived normal tissue complication probability (NTCP) model. Methods: Fifty patients with Stage I-III cervical cancer undergoing IMRT and concurrent weekly cisplatin were analyzed. Acute GI toxicity was graded using the Radiation Therapy Oncology Group scale, excluding upper GI events. A logistic model was used to test correlations between acute GI toxicity and bowel dosimetric parameters. The primary objective was to test the association between Grade {>=}2 GI toxicity and the volume of bowel receiving {>=}45 Gy (V{sub 45}) using the logistic model. Results: Twenty-three patients (46%) had Grade {>=}2 GI toxicity. The mean (SD) V{sub 45} was 143 mL (99). The mean V{sub 45} values for patients with and without Grade {>=}2 GI toxicity were 176 vs. 115 mL, respectively. Twenty patients (40%) had V{sub 45} >150 mL. The proportion of patients with Grade {>=}2 GI toxicity with and without V{sub 45} >150 mL was 65% vs. 33% (p = 0.03). Logistic model parameter estimates V50 and {gamma} were 161 mL (95% confidence interval [CI] 60-399) and 0.31 (95% CI 0.04-0.63), respectively. On multivariable logistic regression, increased V{sub 45} was associated with an increased odds of Grade {>=}2 GI toxicity (odds ratio 2.19 per 100 mL, 95% CI 1.04-4.63, p = 0.04). Conclusions: Our results support the hypothesis that increasing bowel V{sub 45} is correlated with increased GI toxicity in cervical cancer patients undergoing IMRT and concurrent cisplatin. Reducing bowel V{sub 45} could reduce the risk of Grade {>=}2 GI toxicity by approximately 50% per 100 mL of bowel spared.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Small Angle Scattering Data Analysis Assisted by Machine Learning Methods

Small angle scattering (SAS) is a widely used technique for characterizing structures of wide ranges of materials. For such wide ranges of applications of SAS, there exist a large number of ways to model the scattering data. While such analysis models are often available from various suites of SAS data analysis software packages, selecting the right model to start with poses a big challenge for beginners to SAS data analysis. Here, we present machine learning (ML) methods that can assist users by suggesting scattering models for data analysis. A series of one-dimensional scattering curves have been generated by using different models to train the algorithms. The performance of the ML method is studied for various types of ML algorithms, resolution of the dataset, and the number of the dataset. The degree of similarities among selected scattering models is presented in terms of the confusion matrix. The scattering model suggestions with prediction scores provide a list of scattering models that are likely to succeed. Therefore, if implemented with extensive libraries of scattering models, this method can speed up the data analysis workflow by reducing search spaces for appropriate scattering models.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Artificial Intelligence for Multiphysics Nuclear Design Optimization with Additive Manufacturing

The geometric flexibility of additively manufactured metals and ceramics generates a very large and open design space that requires advanced modeling and simulation tools for physics simulations and the rigorous definition of design problems. This effort deploys artificial intelligence (AI) and machine learning (ML) algorithms to understand the design space, evaluate potential designs, and more efficiently generate optimized results. The Transformational Challenge Reactor (TCR) program is leveraging advances in several scientific areas—including materials, manufacturing, sensors and control systems, data analytics, and high-fidelity modeling and simulation—to accelerate the design, manufacturing, qualification, and deployment of advanced nuclear energy systems. Through a manufacturing-informed design approach, the TCR program seeks to integrate digital data for rapid nuclear innovation; accelerate the adoption of advances in manufacturing, materials, and computational sciences for nuclear applications; and dramatically reduce deployment costs and timelines for new nuclear reactor technologies. This report documents efforts under the TCR program to leverage advanced modeling and simulation techniques driven by AI/ML algorithms on high-performance computing (HPC) systems to yield more optimized TCR core designs. A multiphysics ML surrogate model was developed to run on the HPC architectures. The surrogate model is trained on high-fidelity simulation data of coupled neutronics and thermofluidics and is used to quickly evaluate thousands of candidate core designs in parallel, which drives the evolution of the cooling channel shapes to minimize temperature peaking and material stress. Outcomes from these activities provide design information and feedback into the core design efforts.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗