Bridging scales in multiscale bubble growth dynamics with correlated fluctuations using neural operator learning
Explore the source record for details and available documents.
Rayleigh-Plesset equation↗
SEARCH · Engineering Papers
Results for “operator learning”
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.
Hydrogen-Recovery Using an AI-Arc-Plasma Learning Operational System for Produced Water (HALO)
recovery of hydrogen from produced water waste stream
02 PETROLEUM↗
Anisotropic Permeability Prediction at the pore scale: A Lattice Boltzmann and Operator Learning Approach
Explore the source record for details and available documents.
Bhattacharjee, Soumya [The University of Texas at ↗
Development of a model for human operator learning: A progress report
Explore the source record for details and available documents.
Levison, William H.↗
Space Operations Learning Center (SOLC) iPhone/iPad Application
This iPhone application, Space Junk Sammy, is intended to be an educational application designed for Apple iPhones and iPads. This new concept educates kids in an innovative way about how orbital debris affects space missions. Orbital debris is becoming a very significant concern for NASA and all Earthorbiting space missions. Spacecraft in low-Earth orbit are in constant danger of being potentially damaged or destroyed by debris. High-profile spacecraft such as the International Space Station (ISS) and Hubble Space Telescope are dealing with orbital debris on a regular basis. Other basic educational concepts that are portrayed are low-Earth orbits, satellites, ISS, attitude control, and other facts that can be presented in betweenlevel popup screens. The Orbital Debris Cleanup game is relatively simple from the user s technical standpoint. It is a 2D game where the user s avatar is a satellite buddy, named Sammy, in orbit around Earth. Sammy is controlled by the user with the device s gyroscope as well as touchscreen controls. It has equipment used for taking care of the space debris objects on the screen. Sammy also has a claw, a laser deflector, and hydrazine rockets to grab or push the debris objects into a higher orbit or into a lower orbit to burn up in the Earth s atmosphere. The user interface shows Sammy and space debris objects constantly moving from left to right, where Sammy is trying to catch the debris objects before they move off the right side of the screen. Everything will be in constant motion to increase fun and add to the realism of orbiting the Earth. The satellite buddy is used to clean up the space debris and protect other satellites. Later levels will include a laser deflector and hydrazine rockets instead of a robotic claw to push the orbital debris into a higher orbit and out of the path of other satellites
Binebrink, Daniel↗
Hydrogen-Recovery Using an AI-Arc-Plasma Learning Operational System For Produced Water (HALO)
Explore the source record for details and available documents.
02 PETROLEUM↗
Managing Risk in Safety Critical Operations - Lessons Learned from Space Operations
The Mission Control Center (MCC) at Johnson Space Center (JSC) has a rich legacy of supporting Human Space Flight operations throughout the Apollo, Shuttle and International Space Station eras. Through the evolution of ground operations and the Mission Control Center facility, NASA has gained a wealth of experience of what it takes to manage the risk in Safety Critical Operations, especially when human life is at risk. The focus of the presentation will be on the processes (training, operational rigor, team dynamics) that enable the JSC/MCC team to be so successful. The presentation will also share the evolution of the Mission Control Center architecture and how the evolution was introduced while managing the risk to the programs supported by the team. The details of the MCC architecture (e.g., the specific software, hardware or tools used in the facility) will not be shared at the conference since it would not give any additional insight as to how risk is managed in Space Operations.
Gonzalez, Steven A.↗
Weighted Composition Operators for Learning Nonlinear Dynamics
Operator theoretic methods in dynamical system have been dominated by the use of Koopman operators and their continuous time counterparts, such as Koopman Generators and Liouville Operators. The advantage gained from their use primarily stems from the ability to extract subspaces and eigenfunctions within a space of observables that are invariant with respect to the Koopman operator over that space. When this occurs, a dynamic mode decomposition of the systems state provides a linear model for the dynamical system. Not all Koopman operators have eigenfunctions that may be exploited in this manner. However, the framework can still be leveraged for approximations using other operators. In this setting, we present a different operator for the study of dynamical systems, the weighted composition operator. These operators are compact for a wide range of dynamics and spaces, and through their interactions with occupation kernels and vector valued kernels, they admit an estimation of the underlying dynamics. Here, this manuscript presents a new algorithm for the data driven study of dynamical systems from data, and also provides two numerical experiments where convergence is achieved as a proof of concept.
97 MATHEMATICS AND COMPUTING↗
Why Learning From All Operations Is Imperative
Learning from All Operations – Panel Discussion Introduction: Tzvetomir Blajev, Flight Safety Foundation Why Learning from All Operations is Imperative: Jon Holbrook, NASA Implemented Approaches to Learning from All Operations American Airlines Japan Airlines Southwest Airlines Delta Air Lines Why Learning from All Operations is Imperative (Summary): Immanuel Barshi, NASA
Jon Holbrook↗
Scaling Field-Theoretic Simulation for Multicomponent Mixtures with Neural Operators
Multicomponent polymer mixtures are ubiquitous in biological self-organization but are notoriously difficult to study computationally. Plagued by both slow single molecule relaxation times and slow equilibration within dense mixtures, molecular dynamics simulations are typically infeasible at the spatial scales required to study the stability of mesophase structure. Polymer field theories offer an attractive alternative, but analytical calculations are only tractable for mean-field theories and nearby perturbations, constraints that become especially problematic for fluctuation-induced effects such as coacervation. Here, we show that a recently developed technique for obtaining numerical solutions to partial differential equations based on operator learning, neural operators, lends itself to a highly scalable training strategy by parallelizing per-species operator maps. We illustrate the efficacy of our approach on six-component mixtures with randomly selected compositions and that it significantly outperforms the state-of-the-art pseudospectral integrators for field-theoretic simulations, especially as polymer lengths become long.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Advancing NASA SatCORPS Global Data Products with Cloud Computing and Machine Learning
Operational satellite imager radiances are valuable for deriving many different physical parameters that can be used for a variety of weather, aviation, and energy applications. The NASA Satellite ClOud and Radiation Property retrieval System (SatCORPS) applies a suite of algorithms to meteorological satellite data to provide cloud properties, radiative fluxes and other parameters on a global scale. The use of cloud computing has enabled recent enhancements to process a constellation of geostationary satellites at higher spatiotemporal resolutions than previously possible that meet low latency and near real-time needs. Data taken from Meteosat-8 and -11, Himawari, GOES-16, and -17 are processed and combined with operational polar orbiting satellite data and composited on a 3-km grid to provide global coverage. To improve the utility of the data products, machine learning and other innovative methods are applied in various ways to help minimize data product uncertainties under the most challenging conditions and to improve their consistency at all times of day. An update on recent SatCORPS enhancements is presented, highlighting the community benefits achieved with the use of cloud computing and machine learning.
William L Smith↗
CloudSat Anomaly Recovery and Operational Lessons Learned
Nov 2011: NASA/JPL declared CloudSat fully operational in DO ]OP Mode . CloudSat collects science data during sunlit portion of orbit, stable spin hibernation in eclipse . New CONOPS requires constant monitoring of thermal and power profiles, while allowing collection of 54 mins of science data per sunlit orbit
spacecraft systems↗
One-shot learning for solution operators of partial differential equations
Learning and solving governing equations of a physical system, represented by partial differential equations (PDEs), from data is a central challenge in many areas of science and engineering. Traditional numerical methods can be computationally expensive for complex systems and require complete governing equations. Existing data-driven machine learning methods require large datasets to learn a surrogate solution operator, which could be impractical. Here, we propose a solution operator learning method that requires only one PDE solution, i.e., one-shot learning, along with suitable initial and boundary conditions. Leveraging the locality of derivatives, we define a local solution operator in small local domains, train it using a neural network, and use it to predict solutions of new input functions via mesh-based fixed-point iteration or meshfree neural-network based approaches. We test our method on various PDEs, complex geometries, and a practical spatial infection spread application, demonstrating its effectiveness and generalization capabilities.
97 MATHEMATICS AND COMPUTING↗
Application of Artificial Intelligence/Machine Learning to Operations Research
This report examines the transformative impact of Artificial Intelligence (AI) and Machine Learning (ML) on operations research, private industry, and government sectors, highlighting their applications in automating processes, enhancing decision-making, and optimizing complex systems. AI/ML technologies have revolutionized industries through predictive maintenance, supply chain optimization, and autonomous systems, while also advancing public safety and defense operations. However, challenges such as data integrity, model transparency, and the need for human oversight persist, particularly in high-consequence environments. The report emphasizes the critical role of explainable AI (XAI) and human-computer interaction models like Human-in-the-Loop (HITL) and Human-on-the-Loop (HOTL) in fostering trust and accountability. Balancing automation with ethical responsibility and transparency is essential for the continued successful integration of AI/ML into operational and strategic decision-making frameworks.
97 MATHEMATICS AND COMPUTING↗
Nonintrusive projection-based reduced order modeling using stable learned differential operators
Nonintrusive projection-based reduced order models (ROMs) are essential for dynamics prediction in multi-query applications where underlying governing equations are known but the access to the source of the underlying full order model (FOM) is unavailable; that is, FOM is a glass-box. This article proposes a learn-then-project approach for nonintrusive model reduction. In the first step of this approach, high-dimensional stable sparse learned differential operators (S-LDOs) are determined using the generated data. In the second step, the ordinary differential equations, comprising these S-LDOs, are used with suitable dimensionality reduction and low-dimensional subspace projection methods to provide equations for the evolution of reduced states. This approach allows easy integration into the existing intrusive ROM framework to enable nonintrusive model reduction while allowing the use of Petrov–Galerkin projections. The applicability of the proposed approach is demonstrated for Galerkin and LSPG projection-based ROMs through four numerical experiments: 1-D scalar advection, 1-D Burgers, 2-D scalar advection and 1-D scalar advection–diffusion–reaction equations. In conclusion, the results indicate that the proposed nonintrusive ROM strategy provides accurate and stable dynamics prediction.
42 ENGINEERING↗
Operations for Learning with Graphical Models
This paper is a multidisciplinary review of empirical, statistical learning from a graphical model perspective. Well-known examples of graphical models include Bayesian net- works, directed graphs representing a Markov chain, and undirected networks representing a Markov field. These graphical models are extended to model data analysis and empirical learning using the notation of plates. Graphical operations for simplifying and manipulating a problem are provided including decomposition, differentiation, and the manipulation of probability models from the exponential family. These operations adapt existing techniques from statistics and automatic differentiation to graphs. Two standard algorithm schemes for learning are reviewed in a graphical framework: Gibbs sampling and the expectation maximization algorithm. Some algorithms are developed in this graphical framework including a generalized version of linear regression, techniques for feed-forward networks, and learning Gaussian and discrete Bayesian networks from data. The paper concludes by sketching some implications for data analysis and summarizing some popular algorithms that fall within the framework presented. The main original contributions here are the decomposition techniques and the demonstration that graphical models provide a framework for understanding and developing complex learning algorithms.
Buntine, Wray L.↗
Turning Operational Lessons Learned into Design Reality
The capabilities and limitations of a particular system design are well known by the people who operate it. Operational workarounds, operational notes and lessons learned are traditional methods for dealing with and documenting design shortcomings. The beginning of each new program brings the hope that hard-learned lessons will be incorporated into the next new system. But often operations personnel find their well-intentioned efforts frustrated by an inability to have their inputs considered by design personnel who have strictly-scoped requirements that are coupled with ambitious cost and schedule targets. There is a way for operational inputs to make it into the design, but the solution involves a combination of organizational culture and technical data. Any organization that utilizes this approach can realize significant benefits over the life cycle of their project.
Brady, David A.↗
Deep nonparametric estimation of operators between infinite dimensional spaces
Learning operators between infinitely dimensional spaces is an important learning task arising in machine learning, imaging science, mathematical modeling and simulations, etc. This paper studies the nonparametric estimation of Lipschitz operators using deep neural networks. Non-asymptotic upper bounds are derived for the generalization error of the empirical risk minimizer over a properly chosen network class. Under the assumption that the target operator exhibits a low dimensional structure, our error bounds decay as the training sample size increases, with an attractive fast rate depending on the intrinsic dimension in our estimation. Our assumptions cover most scenarios in real applications and our results give rise to fast rates by exploiting low dimensional structures of data in operator estimation. We also investigate the influence of network structures (e.g., network width, depth, and sparsity) on the generalization error of the neural network estimator and propose a general suggestion on the choice of network structures to maximize the learning efficiency quantitatively.
97 MATHEMATICS AND COMPUTING↗