Leveraging Approximation Theory for Efficient Scientific Machine Learning
Explore the source record for details and available documents.
Actor, Jonas [Sandia National Laboratories (SNL-NM↗
SEARCH · Engineering Papers
Results for “state machine”
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.
Nuclear Deflection of Kilometer-Scale Asteroids Simulated using Sandia's Z Machine
Explore the source record for details and available documents.
Moore, Nathan W. [Sandia National Laboratories (SN↗
Leveraging Numerical Methods for Efficient Scientific Machine Learning
Explore the source record for details and available documents.
Actor, Jonas [Sandia National Laboratories (SNL-NM↗
Development of a Digital Twin Towards Enhanced Monitoring of the SF6 Sub-system on the Z Machine
Explore the source record for details and available documents.
McCabe, Isabel Nicole [Sandia National Laboratorie↗
Predicting Chemical Reaction Barriers with a Machine Learning Model
In the past few decades, tremendous advances have been made in the understanding of catalysis at solid surfaces. Despite this, most discoveries of materials for improved catalytic performance are made by a slow trial and error process in an experimental laboratory. Computational simulations have begun to provide a way to rationally design materials for optimizing catalytic performance, but due to the high computational expense of calculating transition state energies, simulations cannot adequately screen the phase space of materials. In this work, we attempt to mitigate this expense by using a machine learning approach to predict the most expensive and most important parameter in a catalyst’s affinity for a reaction: the reaction barrier. Previous methods which used the step reaction energy as the only parameter in a linear regression had a mean absolute error (MAE) on the order of 0.4 eV, too high to be used predictively. Here in our work, we achieve a MAE of about 0.22 eV, a marked improvement towards the goal of computational prediction of catalytic activity.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
The impact of curation errors in the PDBBind Database on machine learning predictions of protein–protein binding affinity
The PDBBind database has been widely utilized for the computational prediction of protein–protein binding affinities. While the accuracy of the PDBBind-curated equilibrium dissociation constants (K D ) has been reported for the protein–ligand subset of the PDBBind database, the curation accuracy has not been reported for the protein–protein subset. Here, we present a detailed manual analysis for the subset of PDBBind records with PubMed Central Open Access primary publications and find that ~19% of these records had K D values that were not supported by their primary publications. The impact of these putative curation errors on the machine learning-based prediction of K D from experimental protein–protein 3D structures was evaluated and correcting the curation errors improved the Pearson correlation coefficient between measured and random forest-predicted log 10 (K D ) values by ~8 percentage points. This finding underscores the importance of dataset accuracy for computational modelling and highlights the need for more stringent curation processes when extracting information from the scientific literature.
59 BASIC BIOLOGICAL SCIENCES↗
Spatiotemporal Super-Resolution with Generative Machine Learning for Creating Renewable Energy Resource Data Under Climate Change Scenarios
As we plan for a future with higher penetrations of renewables and increasing electrification, it becomes more important to understand how the electricity grid will operate under a variety of weather events. We must also consider that the weather our future grid will experience will be different and possibly more extreme than the historical weather that we have extensive data for. We can use data from global climate models (GCMs) to help understand how our climate may change over the next several decades, but there is often a significant gap between the low-resolution GCM data and the high-resolution weather data required to study power systems under specific weather events. Therefore, our objective in this work is to develop tools that can bridge this gap by using low-resolution GCM data to create realistic high-resolution weather datasets that can be used to study renewable energy generation and electricity demand. To accomplish this objective, we have developed a set of generative machine learning models that can rapidly downscale GCM daily average output data at an approximate grid resolution of 100km to hourly data at an approximate 4 km grid resolution. The models can be used to create high resolution data from nearly any GCM included in the Coupled Model Intercomparison Project (CMIP) Phase 5 or 6. Our methods include all datasets regularly used to study the integration of wind and solar power plants as well as changes in electricity demand due to heating and cooling loads. These models and datasets enable power systems modelers to study climate change-influenced weather events and their impact on the grid. We have downscaled and validated wind, solar, temperature, and humidity data with very promising results. The generative machine learning methods are computationally efficient and produce data that has similar statistical characteristics to current state-of-the-art historical datasets. We have trained initial generative models and produced an initial dataset collectively referred to as Sup3rCC: Super-Resolved Renewable Energy Resource Data with Climate Change Impacts. The data covers a (mostly) historical period from 2015-2025 and a future period from 2050-2059. We have also taken hypothetical high-electrification load data and scaled the heating and cooling loads with respect to the 2050-2059 high-resolution Sup3rCC meteorology. The results show how future levels of renewable energy generation and electrified load may be impacted by climate change, setting the stage for capacity expansion models to consider a dynamic climate through model years.
climate change↗
Accelerating Computational Materials Discovery with Machine Learning and Cloud High-Performance Computing: from Large-Scale Screening to Experimental Validation
High-throughput computational materials discovery has promised significant acceleration of the design and discovery of new materials for many years. Despite a surge in interest and activity, the constraints imposed by large-scale computational resources present a significant bottleneck. Furthermore, examples of large-scale computational discovery carried through experimental validation remain scarce, especially for materials with product applicability. In this paper, we demonstrate how this vision became reality by first combining state-of-the-art artificial intelligence (AI) models and traditional physics-based models on cloud high performance computing (HPC) resources to quickly navigate through more than 32 million candidates and predict around half a million potentially stable materials. Focusing on solid-state electrolytes for battery applications, our discovery pipeline further identified 18 promising candidates with new compositions and rediscovered a decade’s worth of collective knowledge in the field as a byproduct. By employing around one thousand virtual machines in the cloud, this process took less than 80 hours. We then synthesized and experimentally characterized the structures and conductivities of our top candidates, the Na x Li 3-x YCl 6 (0.5 ≤ x ≤ 2.5) series, demonstrating the potential of these compounds to serve as solid electrolytes. Additional candidate materials are currently under experimental investigation that could offer more examples of the computational discovery of new phases of Li- and Na-conducting solid electrolytes. We believe this unprecedented approach of synergistically integrating AI models and cloud HPC not only accelerates materials discovery but also showcases the potency of AI-guided experimentation in unlocking transformative scientific breakthroughs with real-world applications.
36 MATERIALS SCIENCE↗
Identifying Sociomarkers of Pediatric Asthma Patients at Risk of Hospital Revisiting
ObjectiveAsthma is one of the most common chronic childhood diseases in the United States [2, 3]. In addition to its pervasiveness, pediatric asthma shows high sensitivity to the environment. Combining medical-social dataset with machine learning methods we demonstrate how socio-markers play an important role in identifying patients at risk of hospital revisits due to pediatric asthma within a year.IntroductionA socio-marker is a measurable indicator of social conditions where a patient is embedded in and exposed to, being analogous with a biomarker indicating the severity or presence of some disease state. Social factors are one of the most clinical health determinants [1], which play a critical role in explaining health outcomes. Socio-markers can help medical practitioners and researchers to reliably identify high-risk individuals in a timely manner.MethodsWe collected data from three different sources: pediatric asthma encounter records from Jan 1st, 2016 to Dec 31st, 2016 at a children’s hospital, the 2010 U.S census data and neighborhood quality survey data by Memphis Property Hub. After merging these datasets we examine the effect of social features in identifying the patients who visited the hospital more than once during the observation period. We only use the first time visit (3,678 cases) to avoid over-counting of the same patients. In addition to demographic features (age, gender, insurance type, and race (African American and White)), we incorporate the social features such as the proportion of individuals living below the federal poverty level, blight prevalence, neighborhood quality, neighborhood quality inequality, trash dumping presence, the broken window pervasiveness within the zip code area of patients’ residence are included.We then implemented a Support Vector Machine (SVM) based classification model using abovementioned 11 social features. The classification outcome is either patient visits the hospital only one-time (class 0) or revisits the hospital within a year (class 1). Among 3,678 unique patients in the dataset, only 823 pediatric patients revisited hospital with asthma. So, to overcome the class imbalance issue, we have used 823 patients’ data (randomly selected in 1,000 iterations) from each class. Further, to avoid overfitting and ensure generalizability, we divided the dataset as training, test, and validation with a proportion of 60%, 20%, and 20%, respectively. The reported test (5-folds cross-validation using training and testing data) and validation accuracy of the SVM method are averaged over 1,000 iterations to avoid sampling error and bias.ResultsThe proposed socio-marker model resulted in an average classification accuracy of 63.70% for the test set and 63.67 % for the validation set. Further, the average specificity (the total true negative cases divided by the sum of true negative and false positive) and sensitivity (the total number of true positive cases divided by the sum of positive predicted cases) is found to be 62.79% and 64.77%, respectively for the test set and 62.79% and 64.83%, respectively for the validation set. Results of this study suggest that socio-marker features that are not directly related to a patient’s medical conditions can still predict whether the patient will come back to the hospital within a year or not with approximately 64% accuracy.ConclusionsBringing the socio-marker features in the surveillance system may ease the burden of detecting the patients at risk of revisiting the hospital. The results should be interpreted with caution because we only used 12-month period of observation and the visit beyond the observation window is not considered. Also the patients may have visited different hospitals which are not captured in the data.References1. Booske BC, Athens JK, Kindig DA, Park H, Remington PL: Different perspectives for assigning weights to determinants of health. University of Wisconsin: Population Health Institute 2010.2. Subbarao P, Mandhane PJ, Sears MR: Asthma: epidemiology, etiology and risk factors. Canadian Medical Association Journal 2009, 181(9):E181-E190.3. Gold DR, Wright R: Population disparities in asthma. Annu Rev Public Health 2005, 26:89-113.
Shin, Eun Kyong↗
Generalized representative structures for atomistic systems
A new method is presented to generate atomic structures that reproduce the essential characteristics of arbitrary material systems, phases, or ensembles. Previous methods allow one to reproduce the essential characteristics (e.g. the chemical disorder) of a large random alloy within a small crystal structure. The ability to generate small representations of random alloys, along with the restriction to crystal systems, results from using the fixed-lattice cluster correlations to describe structural characteristics. A more general description of the structural characteristics of atomic systems is obtained using complete sets of atomic environment descriptors. These are used within for generating representative atomic structures without restriction to fixed lattices. A general data-driven approach is provided here utilizing the atomic cluster expansion (ACE) basis. The N-body ACE descriptors are a complete set of atomic environment descriptors that span both chemical and spatial degrees of freedom and are used within for describing atomic structures. The generalized representative structure (GRS) method presented within generates small atomic structures that reproduce ACE descriptor distributions corresponding to arbitrary structural and chemical complexity. It is shown that systematically improvable representations of crystalline systems on fixed parent lattices, amorphous materials, liquids, and ensembles of atomic structures may be produced efficiently through optimization algorithms. With the GRS method, we highlight reduced representations of atomistic machine-learning training datasets that contain similar amounts of information and small 40–72 atom representations of liquid phases. The ability to use GRS methodology as a driver for informed novel structure generation is also demonstrated. The advantages over other data-driven methods and state-of-the-art methods restricted to high-symmetry systems are highlighted.
atomic cluster expansion↗
Machine learning potential assisted exploration of complex defect potential energy surfaces
Abstract Atomic-scale defects generated in materials under both equilibrium and irradiation conditions can significantly impact their physical and mechanical properties. Unraveling the energetically most favorable ground-state configurations of these defects is an important step towards the fundamental understanding of their influence on the performance of materials ranging from photovoltaics to advanced nuclear fuels. Here, using fluorite-structured thorium dioxide (ThO 2 ) as an exemplar, we demonstrate how density functional theory and machine learning interatomic potential can be synergistically combined into a powerful tool that enables exhaustive exploration of the large configuration spaces of small point defect clusters. Our study leads to several unexpected discoveries, including defect polymorphism and ground-state structures that defy our physical intuitions. Possible physical origins of these unexpected findings are elucidated using a local cluster expansion model developed in this work.
36 MATERIALS SCIENCE↗
A Machine Learning Approach to Predict Aircraft Landing Times using Mediated Predictions from Existing Systems
We developed a novel approach for predicting the landing time of airborne flights in real-time operations. The first step predicts a landing time by using mediation rules to select from among physics-based predictions (relying on the expected flight trajectory) already available in real time in the Federal Aviation Administration System Wide Information Management system data feeds. The second step uses a machine learning model built upon the mediated predictions. The model is trained to predict the error in the mediated prediction, using features describing the current state of an airborne flight. These features are calculated in real time from a relatively small number of data elements that are readily available for airborne flights. Initial results based on five months of data at six large airports demonstrate that incorporating a machine learning model on top of the mediated physics-based prediction can lead to substantial additional improvements in prediction quality.
Machine learning↗
A Machine Learning Approach to Predict Aircraft Landing Times using Mediated Predictions from Existing Systems
We developed a novel approach for predicting the landing time of airborne flights in real-time operations. The first step predicts a landing time by using mediation rules to select from among physics-based predictions (relying on the expected flight trajectory) already available in real time in the Federal Aviation Administration System Wide Information Management system data feeds. The second step uses a machine learning model built upon the mediated predictions. The model is trained to predict the error in the mediated prediction, using features describing the current state of an airborne flight. These features are calculated in real time from a relatively small number of data elements that are readily available for airborne flights. Initial results based on five months of data at six large airports demonstrate that incorporating a machine learning model on top of the mediated physics-based prediction can lead to substantial additional improvements in prediction quality.
Machine learning↗
Two-level overlapping additive Schwarz preconditioner for training scientific machine learning applications
In this work we introduce a novel two-level overlapping additive Schwarz preconditioner for accelerating the training of scientific machine learning applications. The design of the proposed preconditioner is motivated by the nonlinear two-level overlapping additive Schwarz preconditioner. The neural network parameters are decomposed into groups (subdomains) with overlapping regions. In addition, the network’s feed-forward structure is indirectly imposed through a novel subdomain-wise synchronization strategy and a coarse-level training step. Through a series of numerical experiments, which consider physicsinformed neural networks and operator learning approaches, we demonstrate that the proposed two-level preconditioner significantly speeds up the convergence of the standard (LBFGS) optimizer while also yielding more accurate machine learning models. Moreover, the devised preconditioner is designed to take advantage of model-parallel computations, which can further reduce the training time.
97 MATHEMATICS AND COMPUTING↗
High-Purity Aluminum Magnet Technology for Advanced Space Transportation Systems
Basic research on advanced plasma-based propulsion systems is routinely focused on plasmadynamics, performance, and efficiency aspects while relegating the development of critical enabling technologies, such as flight-weight magnets, to follow-on development work. Unfortunately, the low technology readiness levels (TRLs) associated with critical enabling technologies tend to be perceived as an indicator of high technical risk, and this, in turn, hampers the acceptance of advanced system architectures for flight development. Consequently, there is growing recognition that applied research on the critical enabling technologies needs to be conducted hand in hand with basic research activities. The development of flight-weight magnet technology, for example, is one area of applied research having broad crosscutting applications to a number of advanced propulsion system architectures. Therefore, NASA Marshall Space Flight Center, Louisiana State University (LSU), and the National High Magnetic Field Laboratory (NHMFL) have initiated an applied research project aimed at advancing the TRL of flight-weight magnets. This Technical Publication reports on the group's initial effort to demonstrate the feasibility of cryogenic high-purity aluminum magnet technology and describes the design, construction, and testing of a 6-in-diameter by 12-in-long aluminum solenoid magnet. The coil was constructed in the machine shop of the Department of Physics and Astronomy at LSU and testing was conducted in NHMFL facilities at Florida State University and at Los Alamos National Laboratory. The solenoid magnet was first wound, reinforced, potted in high thermal conductivity epoxy, and bench tested in the LSU laboratories. A cryogenic container for operation at 77 K was also constructed and mated to the solenoid. The coil was then taken to NHMFL facilities in Tallahassee, FL. where its magnetoresistance was measured in a 77 K environment under steady magnetic fields as high as 10 T. In addition, the temperature dependence of the coil's resistance was measured from 77 to 300 K. Following this series of tests, the coil was transported to NHMFL facilities in Los Alamos, NM, and pulsed to 2 T using an existing capacitor bank pulse generator. The coil was completely successful in producing the desired field without damage to the windings.
Goodrich, R. G.↗
Process‐Oriented Calibration of a Turbulence Scheme in the DOE's Global Storm‐Resolving Model Using Machine Learning
A process‐oriented calibration framework is developed for the Simplified Higher‐Order Closure (SHOC) turbulence scheme in DOE's Simple Cloud Resolving E3SM Atmospheric Model (SCREAM). This framework leverages machine learning surrogates and observational constraints to efficiently calibrate SHOC adjustable parameters across two convective regimes: clear‐sky dry convective boundary layer and fair‐weather shallow cumulus clouds from ARM observations. We use perturbed‐parameter ensembles of a doubly periodic version of SCREAM to train surrogates and apply Markov Chain Monte Carlo sampling guided by cost functions based on benchmarking large‐eddy simulations and observations to identify optimized parameter sets that perform well in both regimes. The calibrated SHOC parameters substantially improve boundary‐layer turbulence and cloud boundaries, and modeled cloud fraction and radiative effects align better with observations than the default. These results demonstrate that combining multiple process‐specific convective regimes with machine‐learning surrogates can reduce parametric uncertainties and yield a model more faithful to cloud–turbulence interactions.
58 GEOSCIENCES↗
Post-Event Fault Identification with Machine Learning for Protection System Validation
Power system protection devices have transitioned over the past few decades from mechanical to analog devices, then to solid state and finally digital. Relays and their associated critical network of equipment have significantly increased in complexity. Even internally, relays have gained significant intricacy, with relatively simple overcurrent or differential functions now being assisted by a myriad of other functions. This is necessary as the grid becomes more complex, but it brings increased difficulty in monitoring and upkeep. Misoperation caused by accidental improper relay settings or deliberate malicious actions is a constant challenge faced by all utilities. These improper settings can be difficult to identify and may require exhaustive post-mortem analysis, typically after a major outage event has already occurred. A mechanism is needed for monitoring the behavior of protection systems to validate that their performance falls within expectations. Relays that fail to isolate a fault or trip when there is no system disturbance can be flagged for settings review in situations where this behavior may not have been noticed due to manual restoration or backup protection operations. This work presents a concept for a machine learning (ML) system capable of validating the performance of protection systems by identifying fault events and characterizing protection system responses based solely on available current and voltage measurements. As a first step in its development, an experimental dataset is generated, and a random forest model is implemented with high accuracy in distinguishing four power system scenarios.
24 - POWER TRANSMISSION AND DISTRIBUTION↗
Learning Model Structural Uncertainty with Gaussian Processes
The advent of commercially available quantum computers has marked the beginning of quantum computing as a reality. Both quantum gate and annealing computers have been released by major computer hardware companies. In this work, the D-Wave 2XTM quantum annealing computer housed at the NASA Advanced Systems computational facility is investigated to accelerate Machine Learning (ML) for image registration. NASA collects large amounts of images over the globe remotely using space-based monitoring. Images of a fixed areas of the land surface are taken over time. Due to the orbit of the sensors, the viewing angles deviate slightly, and it is necessary to align or register the images precisely to create image time series over the land surface. Unaligned images can lead to substantial analysis errors. These time-series are then used in modeling Earth Systems models such as hydrological, weather, and carbon monitoring models. In this work, we consider the Moderate Resolution Image Spectrometer (MODIS) data collected by the NASA's terra satellite. Artificial Neural Networks (ANNs) is a natural fit for ML modelling of images. Several successes have been reported using machine learning related to image processing. We investigate the use of ML to register MODIS images. ANNs are investigated in combination with a Restricted Boltzmann Machines (RBM) as an auto-encoder. We will present results showing the accuracy and efficiency of this approach.The D-Wave 2XTM quantum annealer samples the ground-state wave-function of a spin-Ising systems with quadratic interactions between qubits and a Chimera connectivity. The system sits in a ~15 mK thermal bath. One can think of the system as being placed in the ground state initially and subject to thermal excitations governed by Boltzmann statistics. If this is assumed true, one can use the statistics from the D-Wave 2XTM to train RBMs. Generating statistics for training Boltzmann machines is an NP-hard problem and constitutes the largest compute cost. We investigate the use of the D-Wave 2XTM to accelerate the training of the RBMs in our ANNs and report on the results.
Kouatchou, Jules↗