Search NASA⌕ Search

SEARCH · Search NASA

Results for “Machine learning prediction”

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 1,747 records · Page 97

Forecasting Megaelectron-Volt Electrons Inside Earth's Outer Radiation Belt: PreMevE 2.0 Based on Supervised Machine Learning Algorithms

We present the recent progress in upgrading a predictive model for megaelectron-volt (MeV) electrons inside the Earth's outer Van Allen belt. This updated model, called PreMevE 2.0, provides improved forecasts, particularly at outer L-shells, by adding upstream solar wind speeds to the model's input parameter list that originally includes precipitating electrons observed at low Earth orbits and MeV electron fluxes in situ measured by a geosynchronous satellite. Furthermore, based on several kinds of linear and artificial neural networks algorithms, a list of models was constructed, trained, validated, and tested with 42-month MeV electron observations from Van Allen Probes. Out-of-sample test results from these models show that, with optimized model hyperparameters and input parameter combinations, the top performer from each category of models has the similar capability of making reliable 1-day (2-day) forecasts of 1-MeV electron flux distributions with performance efficiency values ~0.87 (~0.82) averaged over the L-shell range of 2.8–6.6, significantly outperforming the previous version of PreMevE particularly at L-shells > ~4.5. Interestingly, the linear regression model is often the most successful when compared to other models, which suggests the relationship between dynamics of trapped 1-MeV electrons and precipitating electrons is dominated by linear components. Results also show that PreMevE 2.0 can reasonably well predict the onsets of MeV electron events in 2-day forecasts. PreMevE 2.0 is designed to be driven by observations from longstanding space infrastructure to make high-fidelity forecasts for MeV electrons, thus can be an invaluable space weather forecasting tool for the future.

79 ASTRONOMY AND ASTROPHYSICS↗

Evaluation of top-down and bottom-up global terrestrial respiration estimates and their mismatch with model simulations

Terrestrial respiration is one of the most poorly understood processes in the global carbon cycle, making respiration predictions uncertain. However, expanding observations and machine learning approaches have led to a proliferation of estimates. We compiled total ecosystem and heterotrophic respiration estimates derived from top-down atmospheric inversions and bottom-up upscaling of ecosystem observations and compared them with dynamic vegetation models (DGVM) simulations over the 1980-2020 period. Our analysis revealed a convergence in mean annual global total ecosystem respiration estimates between top-down 97.1 (± SD 6.8) PgC yr-1 and bottom-up 98.5 (+/-13.4) PgC yr-1, which were both significantly lower than the ensemble mean from DGVMs estimates 133.7 (±4.7) PgC yr-1. We also found similar temporal trends between top-down estimates with a mean of 0.075 (±0.05) PgC yr-2, and bottom-up estimates of 0.05 (±0.05) PgC yr-2, which were 5 to 7 times smaller than the ensemble mean trend of 0.34 PgC yr-2 simulated by DGVMs. Global heterotrophic respiration showed much less agreement, ranging from top-down estimates of 42.7 (±4.0) PgC yr-1 to bottom-up estimates of 51.5 (±4.0) PgC yr-1 and a significantly larger ensemble model mean estimate of 60.8 PgC yr-1 (±1.9). The temporal trends in observation-based bottom-up estimates of heterotrophic respiration of 0.03 PgC yr-2 were five times lower than the model ensemble mean 0.15 PgC yr-2. Large regional disagreements in heterotrophic respiration estimates and simulations were evident in tropical and boreal latitudes. Therefore, improved regional and heterotrophic respiration estimates are necessary to reduce uncertainties regarding the future vulnerability of soil carbon.

Ballantyne, Ashley↗

DIII-D High Field Side Lower Hybrid Current Drive Experiment Overview

In preparation for high field side lower hybrid current drive (HFS LHCD) experiments in DIII-D, the HFS LHCD was to be commissioned and physics experiments commence once the system operated up to 300 kW for 0.5 s. The initial physics experiments sought to characterize coupling, wave propagation, and driven current measurements. In HFS LHCD first campaign, the maximum power was limited to <200 kW due to waveguide pressure leaks limiting the number of available modules and power per module. Here we summarize commissioning progress and initial physics observations. The HFS LHCD coupler was optimized for high qmin, DIII-D discharges where efficient off-axis current at r/a~0.6-0.8 is desired. The coupler n|| spectrum is peaked at 2.7 and is predicted to generate ~0.14 MA/MW coupled for 1.6 MW injected. In preparation, the HFS scrape-off layer density profile was characterized and found to have steeper profiles and lower fluctuation levels than the low field side. Furthermore, the HFS SOL density profile can be accurately predicted using global plasma quantities using machine learning. Thus far, one module has injected ~100 kW for 0.5 s with <5% reflected power. Nonthermal electrons have been observed on lower frequency channels of the electron cyclotron emission radiometer correlated with the LH power indicating core wave absorption. To avoid 30R neutral beam heat flux, a split launcher is proposed to avoid the high heat flux region while maintaining power spectrum and directivity.

Wukitch, Stephen J. [Massachusetts Inst. of Techno↗

MAGNETIC NONPOTENTIALITY IN PHOTOSPHERIC ACTIVE REGIONS AS A PREDICTOR OF SOLAR FLARES

Based on several magnetic nonpotentiality parameters obtained from the vector photospheric active region magnetograms obtained with the Solar Magnetic Field Telescope at the Huairou Solar Observing Station over two solar cycles, a machine learning model has been constructed to predict the occurrence of flares in the corresponding active region within a certain time window. The Support Vector Classifier, a widely used general classifier, is applied to build and test the prediction models. Several classical verification measures are adopted to assess the quality of the predictions. We investigate different flare levels within various time windows, and thus it is possible to estimate the rough classes and erupting times of flares for particular active regions. Several combinations of predictors have been tested in the experiments. The True Skill Statistics are higher than 0.36 in 97% of cases and the Heidke Skill Scores range from 0.23 to 0.48. The predictors derived from longitudinal magnetic fields do perform well, however, they are less sensitive in predicting large flares. Employing the nonpotentiality predictors from vector fields improves the performance of predicting large flares of magnitude {>=}M5.0 and {>=}X1.0.

79 ASTRONOMY AND ASTROPHYSICS↗

Divergence in land surface modeling: linking spread to structure

Divergence in land carbon cycle simulation is persistent and widespread. Regardless of model intercomparison project, results from individual models diverge significantly from each other and, in consequence, from reference datasets. Here we link model spread to structure using a 15-member ensemble of land surface models from the Multi-scale synthesis and Terrestrial Model Intercomparison Project (MsTMIP) as a test case. Our analysis uses functional benchmarks and model structure as predicted by model skill in a machine learning framework to isolate discrete aspects of model structure associated with divergence. We also quantify how initial conditions prejudice present-day model outcomes after centennial-scale transient simulations. Overall, the functional benchmark and machine learning exercises emphasize the importance of ecosystem structure in correctly simulating carbon and water cycling, highlight uncertainties in the structure of carbon pools, and advise against hard parametric limits on ecosystem function. We also find that initial conditions explain 90% of variation in global satellite-era values—initial conditions largely predetermine transient endpoints, historical environmental change notwithstanding. As MsTMIP prescribes forcing data and spin-up protocol, the range in initial conditions and high levels of predetermination are also structural. Our results suggest that methodological tools linking divergence to discrete aspects of model structure would complement current community best practices in model development.

58 GEOSCIENCES↗

Diffuse scattering from dynamically compressed single-crystal zirconium following the pressure-induced α → ω phase transition

The prototypical α → ω phase transition in zirconium is an ideal test bed for our understanding of polymorphism under extreme loading conditions. After half a century of study, a consensus had emerged that the transition is realized via one of two distinct displacive mechanisms, depending on the nature of the compression path. However, recent dynamic-compression experiments equipped with diffraction diagnostics performed in the past few years have revealed new transition mechanisms, demonstrating that our understanding of the underlying atomistic dynamics and transition kinetics is in fact far from complete. We present classical molecular dynamics simulations of the α → ω phase transition in single-crystal zirconium shock compressed along the [0001] axis using a machine-learning-class potential. The transition is predicted to proceed primarily via a modified version of the two-stage Usikov-Zilberstein mechanism, whereby the high-pressure ω phase heterogeneously nucleates at boundaries between grains of an intermediate β phase. We further observe the fomentation of atomistic disorder at the junctions between β grains, leading to the formation of highly defective interstitial material between the ω grains. We directly compare synthetic x-ray diffraction patterns generated from our simulations with those obtained using femtosecond diffraction in recent dynamic-compression experiments, and show that the simulations produce the same unique, anisotropic diffuse scattering signal unlike any previously seen from an elemental metal. Our simulations suggest that the diffuse signal arises from a combination of thermal diffuse scattering, nanoparticlelike scattering from residual kinetically stabilized α and β grains, and scattering from interstitial defective structures. Published by the American Physical Society 2024

36 MATERIALS SCIENCE↗

High-pressure melt dynamics in shock-compressed titanium

In this work we study the high-pressure melting behavior of titanium using laser-driven shock compression with in situ femtosecond x-ray diffraction and molecular-dynamics simulations based on a machine-learned interatomic potential. The MD simulations predict the solid-liquid coexistence on the Hugoniot in the ∼111−124GPa range. Experimentally, we observe the first evidence of liquid at 86 GPa. We also observe pronounced microstructural changes with pressure, with strong grain refinement associated with the emergence of liquid, within the solid-liquid coexistence (∼110−126GPa). Above 126 GPa, we observe the persistence of residual levels of highly textured crystalline Ti to ∼180GPa, well above the expected melt completion pressure. We discuss the accuracy that current laser-shock experimental platforms have at determining the melt onset and completion pressures.

36 MATERIALS SCIENCE↗

Neural Networks for Modeling and Control of Particle Accelerators

Particle accelerators are host to myriad nonlinear and complex physical phenomena. They often involve a multitude of interacting systems, are subject to tight performance demands, and should be able to run for extended periods of time with minimal interruptions. Often times, traditional control techniques cannot fully meet these requirements. One promising avenue is to introduce machine learning and sophisticated control techniques inspired by artificial intelligence, particularly in light of recent theoretical and practical advances in these fields. Within machine learning and artificial intelligence, neural networks are particularly well-suited to modeling, control, and diagnostic analysis of complex, nonlinear, and time-varying systems, as well as systems with large parameter spaces. Consequently, the use of neural network-based modeling and control techniques could be of significant benefit to particle accelerators. For the same reasons, particle accelerators are also ideal test-beds for these techniques. Many early attempts to apply neural networks to particle accelerators yielded mixed results due to the relative immaturity of the technology for such tasks. The purpose of this paper is to re-introduce neural networks to the particle accelerator community and report on some work in neural network control that is being conducted as part of a dedicated collaboration between Fermilab and Colorado State University (CSU). We describe some of the challenges of particle accelerator control, highlight recent advances in neural network techniques, discuss some promising avenues for incorporating neural networks into particle accelerator control systems, and describe a neural network-based control system that is being developed for resonance control of an RF electron gun at the Fermilab Accelerator Science and Technology (FAST) facility, including initial experimental results from a benchmark controller.

43 PARTICLE ACCELERATORS↗

Evaluation of Power Transmission Lines Hardening Scenarios Using a Machine Learning Approach

The power transmission infrastructure is vulnerable to extreme weather events, particularly hurricanes and tropical storms. A recent example is the damage caused by Hurricane Maria (H-Maria) in the archipelago of Puerto Rico in September 2017, where major failures in the transmission infrastructure led to a total blackout. Numerous studies have been conducted to examine strategies to strengthen the transmission system, including burying the power lines underground or increasing the frequency of tree trimming. However, few studies focus on the direct hardening of the transmission towers to accomplish an increase in resiliency. This machine learning-based study fills this need by analyzing three direct hardening scenarios and determining the effectiveness of these changes in the context of H-Maria. A methodology for estimating transmission tower damage is presented here in this study as well as an analysis of impact of replacing structures with a high failure rate with more resilient ones. We found the steel self-support-pole to be the best replacement option for the towers with high failure rate. Furthermore, the third hardening scenario, where all wooden poles were replaced, exhibited a maximum reduction in damaged towers in a single line of 66% while lowering the mean number of damaged towers per line by 10%.

54 ENVIRONMENTAL SCIENCES↗

Optimizing transmit field inhomogeneity of parallel RF transmit design in 7T MRI using deep learning

Ultrahigh field (UHF) Magnetic Resonance Imaging (MRI) provides a higher signal-to-noise ratio and, thereby, higher spatial resolution. However, UHF MRI introduces challenges such as transmit radiofrequency (RF) field (B+1) inhomogeneities, leading to uneven flip angles and image intensity anomalies. These issues can significantly degrade imaging quality and its medical applications. This study addresses B+1 field homogeneity through a novel deep learning-based strategy. Traditional methods like Magnitude Least Squares (MLS) optimization have been effective but are time-consuming and dependent on the patient’s presence. Recent machine learning approaches, such as RF Shim Prediction by Iteratively Projected Ridge Regression and deep learning frameworks, have shown promise but face limitations like extensive training times and oversimplified architectures. We propose a two-step deep learning strategy. First, we obtain the desired reference RF shimming weights from multi-channel B+1 fields using random-initialized Adaptive Moment Estimation. Then, we employ Residual Networks (ResNets) to train a model that maps B+1 fields to target RF shimming outputs. Our approach does not rely on pre-calculated reference optimizations for the testing process and efficiently learns residual functions. Comparative studies with traditional MLS optimization demonstrate our method’s advantages in terms of speed and accuracy. The proposed strategy achieves a faster and more efficient RF shimming design, significantly improving imaging quality at UHF. This advancement holds potential for broader applications in medical imaging and diagnostics.

Lu, Zhengyi [Vanderbilt University]↗

Microbial vitamin biosynthesis links gut microbiota dynamics to chemotherapy toxicity

ABSTRACT Dose-limiting toxicities pose a major barrier to cancer treatment. While preclinical studies show that the gut microbiota influences and is influenced by anticancer drugs, data from patients paired with careful side effect monitoring remains limited. Here, we investigate capecitabine (CAP)-microbiome interactions through longitudinal metagenomic sequencing of stool from 56 advanced colorectal cancer patients. CAP significantly altered the gut microbiome, enriching for menaquinol (vitamin K2) biosynthesis genes. Transposon library screens, targeted gene deletions, and media supplementation revealed that menaquinol biosynthesis protectsEscherichia colifrom drug toxicity. Stool menaquinol gene and metabolite levels were associated with decreased peripheral sensory neuropathy. Machine learning models trained in this cohort predicted toxicities in an independent cohort. Taken together, these results suggest treatment-associated increases in microbial vitamin biosynthesis serve a chemoprotective role for bacterial and host cells. Further, our findings provide a foundation for in-depth mechanistic dissection, human intervention studies, and extension to other cancer treatments. IMPORTANCE Side effects are common during the treatment of cancer. The trillions of microbes found within the human gut are sensitive to anticancer drugs, but the effects of treatment-induced shifts in gut microbes for side effects remain poorly understood. We profiled gut microbes in colorectal cancer patients treated with capecitabine and carefully monitored side effects. We observed a marked expansion in genes for producing vitamin K2 (menaquinone). Vitamin K2 rescued gut bacterial growth and was associated with decreased side effects in patients. We then used information about gut microbes to develop a predictive model of drug toxicity that was validated in an independent cohort. These results suggest that treatment-associated increases in bacterial vitamin production protect both bacteria and host cells from drug toxicity, providing new opportunities for intervention and motivating the need to better understand how dietary intake and bacterial production of micronutrients like vitamin K2 influence cancer treatment outcomes.

Microbiology↗

listenr

SAND2024-13902O listenr is an R package containing code that allows users to fit echo state networks (a machine learning model) on data to obtain predictions. The code also allows users to compute spatio-temporal feature importance on the echo state network, as described in “Characterizing Climate Pathways Using Feature Importance on Echo State Networks.” The package also provides a function for computing principal components.

Ries, Daniel↗

ChemEcho v1.0

ChemEcho is a tool that converts tandem mass spectra into embeddings used to build machine learning (ML) models with fully explainable predictions. It provides an API for transforming raw tandem mass spectral data into embeddings, along with functions for training and validating ML models. Additionally, it includes utilities for retrieving and cleaning training data. ChemEcho is broadly applicable in ML pipelines that use tandem mass spectra for a variety of tasks, such as chemical classification or bioactivity mining. While there are existing methods to generate embeddings from fragmentation data, ChemEcho's approach ensures that predictions remain interpretable, enabling experts to evaluate results and generate hypotheses about the underlying data.

Harwood, Thomas [Lawrence Berkeley National Labora↗

Using the concordance of in vitro and in vivo data to evaluate extrapolation assumptions

Linking in vitro bioactivity and in vivo toxicity on a dose basis enables the use of highthroughput in vitro assays as an alternative to traditional animal studies. In this study, we evaluated assumptions in the use of a high-throughput, physiologically based toxicokinetic (PBTK) model to relate in vitro bioactivity and rat in vivo toxicity data. The fraction unbound in plasma (fup) and intrinsic hepatic clearance (Clint) were measured for rats (for 67 and 77 chemicals, respectively), combined with fup and Clint literature data for 97 chemicals, and incorporated in the PBTK model. Of these chemicals, 84 had corresponding in vitro ToxCast bioactivity data and in vivo toxicity data. For each possible comparison of in vitro and in vivo endpoint, the concordance between the in vivo and in vitro data was evaluated by a regression analysis. For a base set of assumptions, the PBTK results were more frequently better associated than either the results from a “random” model parameterization or direct comparison of the “untransformed” values of AC50 and dose (performed best in 51%, 28%, and 21% of cases, respectively). We also investigated several assumptions in the application of PBTK for IVIVE, including clearance and internal dose selection. One of the better assumptions sets–restrictive clearance and comparing free in vivo venous plasma concentration with free in vitro concentration–outperformed the random and untransformed results in 71% of the in vitro-in vivo endpoint comparisons. These results demonstrate that applying PBTK improves our ability to observe the association between in vitro bioactivity and in vivo toxicity data in general. This suggests that potency values from in vitro screening should be transformed using in vitro-in vivo extrapolation (IVIVE) to build potentially better machine learning and other statistical models for predicting in vivo toxicity in humans

59 BASIC BIOLOGICAL SCIENCES↗

Optimizing collimator positions using bayesian optimization in the Fermilab MI-8 transfer line

Collimators are used to minimize losses and to remove particles that would otherwise get lost downstream and irradiate the machine. Finding the optimal jaw positions is time consuming and with the upstream beam properties changing, the collimation settings would need to be readjusted each time. Therefore, a method to optimize collimator positions and to operate them at full capacity in a short time is required for loss control downstream. A study of collimator positions was conducted and a machine learning (ML) model was developed to predict optimal collimator positions. Bayesian Optimization (BO) was used to calculate new jaw positions from the ML model. The results of BO and usage of ML for better performance of the collimation system are presented in this paper.

Babacan, Betiay [Fermilab]↗

Transforming our understanding of chloroplast-associated genes through comprehensive characterization of protein localizations and protein-protein interactions

Bioenergy crops are a renewable source of fuels and are a critical base for building a carbon-neutral economy. Rational engineering of bioenergy crops has the potential to enhance the yields. However, our ability to engineer plants is limited because the functions of most genes remain unknown. Systematic characterization of gene function in plants thus has the potential to greatly accelerate bioenergy research. Here, we focus on the chloroplast, an underexplored energy-producing organelle that is a hallmark of plants. The chloroplast is one of the promising targets of biofuel crop engineering efforts because of its central role in photosynthesis, metabolism, and intracellular signaling. However, the protein composition of the chloroplast and the functions of most of its proteins remain poorly characterized. At the core of this project, we sought to comprehensively determine the localization of chloroplast-associated proteins and generate a spatially defined protein-protein interaction network for chloroplast. For this purpose, we used the leading model alga Chlamydomonas reinhardtii, which greatly increased experimental speed and throughput. We illustrated the value of our findings to land plants by determining the localization of Arabidopsis thaliana land plant homologs of the Chlamydomonas proteins. Altogether, we were successful in determining the localization of 1,034 chloroplast-associated proteins in Chlamydomonas. The localizations provide numerous insights into the spatial organization of chloroplasts and how they function to support photosynthesis. The localization patterns of distinct proteins revealed new chloroplast structures and revealed new spatial organization inside the chloroplast. We also identified new components of known chloroplast structures, such as the chloroplast envelope, nucleoid, plastoglobuli, and pyrenoid. We identified these new components by investigating the interacting partners of known proteins. Many proteins localized in both the chloroplast and other cellular structures, thereby hinting at new functions and communication between cellular structures. We also applied machine learning on the atlas to generate predictions for the location of all of the proteins in Chlamydomonas. This enabled us to assign putative functions to many uncharacterized proteins based on their cellular location. Altogether, this research establishes a rich resource that opens new avenues of investigation and guides future work in deciphering and manipulating chloroplast function. Next, we developed an extensive protein-protein interaction network for the chloroplast by performing affinity purification-mass spectrometry on ~1,150 tagged chloroplast-associated proteins, the first such large-scale study in any photosynthetic organism. This dataset reveals 4,694 high-confidence protein-protein interactions, offering insights into the functions of thousands of conserved poorly-characterized chloroplast proteins. This systematic identification of protein-protein interactions in the chloroplast also provides multiple exciting new research directions and a detailed blueprint of the chloroplast's operation. This research lays the groundwork to decipher the inner workings of the chloroplast, the cell structure at the heart of photosynthesis. The spatial atlas and protein-protein interactions reveal chloroplast organizational features that would not have been accessible with traditional approaches. The localization mapping, insights into the function, and research materials generated further provide a rich resource for the research community to advance the understanding of how the chloroplast is organized to enable engineering of enhanced photosynthetic organisms.

59 BASIC BIOLOGICAL SCIENCES↗