Engineering Papers⌕ Search

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.

At least 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Real-time inference and extrapolation with Time-Conditioned UNet: Applications in hypersonic flows, incompressible flows, and global temperature forecasting

Neural Operators are fast and accurate surrogates for nonlinear mappings between functional spaces within training domains. Extrapolation beyond the training domain remains a grand challenge across all application areas. We present Time-Conditioned UNet (TC-UNet) as an operator learning method to solve time-dependent PDEs continuously in time without any temporal discretization, including in extrapolation scenarios. TC-UNet incorporates the temporal evolution of the PDE into its architecture by combining a parameter conditioning approach with the attention mechanism from the Transformer architecture. After training, TC-UNet makes real-time inferences on an arbitrary temporal grid. We demonstrate its extrapolation capability on a climate problem by estimating the global temperature for several years and also for inviscid hypersonic flow around a double cone. We propose different training strategies involving temporal bundling and sub-sampling. We demonstrate performance improvements for several benchmarks, performing extrapolation for long time intervals and zero-shot super-resolution time.

Deep learning↗
NASA NTRS2005

Diet as a factor in behavioral radiation protection following exposure to heavy particles

Major risks associated with radiation exposures on deep space missions include carcinogenesis due to heavy-particle exposure of cancer-prone tissues and performance decrements due to neurological damage produced by heavy particles. Because exposure to heavy particles can cause oxidative stress, it is possible that antioxidants can be used to mitigate these risks (and possibly some health risks of microgravity). To assess the capacity of antioxidant diets to mitigate the effects of exposure to heavy particles, rats were maintained on antioxidant diets containing 2% blueberry or strawberry extract or a control diet for 8 weeks prior to exposure to 1.5 or 2.0 Gy of accelerated iron particles at Brookhaven National Laboratory. Following irradiation rats were tested on a series of behavioral tasks: amphetamine-induced taste aversion learning, operant responding and spatial learning and memory. The results indicated that the performance of the irradiated rats maintained on the antioxidant diets was, in general, significantly better than that of the control animals, although the effectiveness of the diets ameliorating the radiation-induced deterioration in performance varied as a function of both the specific diet and the specific endpoint. In addition, animals fed antioxidant diets prior to exposure showed reduced heavy particle-induced tumorigenesis one year after exposure compared to the animals fed the control diet. These results suggest that antioxidant diets have the potential to serve as part of a system designed to provide protection to astronauts against the effects of heavy particles on exploratory missions outside the magnetic field of the earth.

Non-NASA Center↗
NASA NTRS2012

Antarctic Exploration Parallels for Future Human Planetary Exploration: Science Operations Lessons Learned, Planning, and Equipment Capabilities for Long Range, Long Duration Traverses

The purpose for this workshop can be summed up by the question: Are there relevant analogs to planetary (meaning the Moon and Mars) to be found in polar exploration on Earth? The answer in my opinion is yes or else there would be no reason for this workshop. However, I think some background information would be useful to provide a context for my opinion on this matter. As all of you are probably aware, NASA has been set on a path that, in its current form, will eventually lead to putting human crews on the surface of the Moon and Mars for extended (months to years) in duration. For the past 50 V 60 years, starting not long after the end of World War II, exploration of the Antarctic has accumulated a significant body of experience that is highly analogous to our anticipated activities on the Moon and Mars. This relevant experience base includes: h Long duration (1 year and 2 year) continuous deployments by single crews, h Established a substantial outpost with a single deployment event to support these crews, h Carried out long distance (100 to 1000 kilometer) traverses, with and without intermediate support h Equipment and processes evolved based on lessons learned h International cooperative missions This is not a new or original thought; many people within NASA, including the most recent two NASA Administrators, have commented on the recognizable parallels between exploration in the Antarctic and on the Moon or Mars. But given that level of recognition, relatively little has been done, that I am aware of, to encourage these two exploration communities to collaborate in a significant way. [Slide 4] I will return to NASA s plans and the parallels with Antarctic traverses in a moment, but I want to spend a moment to explain the objective of this workshop and the anticipated products. We have two full days set aside for this workshop. This first day will be taken up with a series of presentations prepared by individuals with experience that extends back as far as the late 1940s and includes contemporary experience. The people presenting bring a variety of points of view, including not only U.S. but international, although most, if not all, have collaborated on international teams. The second day will consist of a series of small focused group interactions centered on those elements likely to be needed for traverse missions, such as mobility, habitation, and extravehicular activity (EVA, aka space suits). Our invited participants will be talking with people that specialize in these elements so that we can foster more direct interaction and exchange of experiences between these two exploration communities. After the workshop we will be preparing a report documenting these presentations and the essence of the focused interactions.

Hoffman, Stephen J.↗
DOE OSTI2024

Bridging Hydrological Ensemble Simulation and Learning Using Deep Neural Operators

Ensemble-based simulation and learning (ESnL) has long been used in hydrology for parameter inference, but computational demands of process-based ESnL can be quite high. To address this issue, we propose a deep neural operator learning approach. Neural operators are generic machine learning algorithms that can learn functional mappings between infinite-dimensional spaces, providing a highly flexible tool for scientific machine learning. Our approach is built upon DeepONet, a specific deep neural operator, and is designed to address several common problems in hydrology, namely, model parameter estimation, prediction at ungaged locations, and uncertainty quantification. Here we demonstrate the effectiveness of our DeepONet-based workflow using an existing large model ensemble created for an eastern U.S. watershed that is instrumented with 10 streamflow gages. Results suggest DeepONet achieves high efficiency in learning an ML surrogate model from the model ensemble, with the modified Kling-Gupta Efficiency exceeding 0.9 on holdout test sets. Parameter inference, carried out using the trained DeepONet surrogate model and genetic algorithm, also yields robust results. Additionally, we formulate and train a separate DeepONet model for physics-informed, seq-to-seq streamflow forecasting, which further reduces biases in the pre-trained DeepONet surrogate model. While this study focuses primarily on a single watershed, our approach is general and may be extended to enable learning from model ensembles across multiple basins or models. Thus, this research represents a significant contribution to the application of hybrid machine learning in hydrology.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2011

Flight Dynamics Operations: Methods and Lessons Learned from Space Shuttle Orbit Operations

The Flight Dynamics Officer is responsible for trajectory maintenance of the Space Shuttle. This paper will cover high level operational considerations, methodology, procedures, and lessons learned involved in performing the functions of orbit and rendezvous Flight Dynamics Officer and leading the team of flight dynamics specialists during different phases of flight. The primary functions that will be address are: onboard state vector maintenance, ground ephemeris maintenance, calculation of ground and spacecraft acquisitions, collision avoidance, burn targeting for the primary mission, rendezvous, deorbit and contingencies, separation sequences, emergency deorbit preparation, mass properties coordination, payload deployment planning, coordination with the International Space Station, and coordination with worldwide trajectory customers. Each of these tasks require the Flight Dynamics Officer to have cognizance of the current trajectory state as well as the impact of future events on the trajectory plan in order to properly analyze and react to real-time changes. Additionally, considerations are made to prepare flexible alternative trajectory plans in the case timeline changes or a systems failure impact the primary plan. The evolution of the methodology, procedures, and techniques used by the Flight Dynamics Officer to perform these tasks will be discussed. Particular attention will be given to how specific Space Shuttle mission and training simulation experiences, particularly off-nominal or unexpected events such as shortened mission durations, tank failures, contingency deorbit, navigation errors, conjunctions, and unexpected payload deployments, have influenced the operational procedures and training for performing Space Shuttle flight dynamics operations over the history of the program. These lessons learned can then be extended to future vehicle trajectory operations.

Cutri-Kohart, Rebecca M.↗
NASA NTRSDate not supplied

Lessons Learned in Hydrogen System Operation

NASA has learned extensive lessons on the operations of hydrogen systems during the developments and operations of space hardware. A subset of these lessons that are useful for aircraft and terrestrial applications using liquid hydrogen are presented to help disseminate the information learned. Discussion includes test data, analysis, and observations on cryogenic insulation systems, stratification and temperature gradients within tank systems, fittings to reduce leakage, and integration of refrigeration systems within a cryogenic fluids system.

Liquid Hydrogen↗
NASA NTRS1987

A learning controller for nonrepetitive robotic operation

A practical learning control system is described which is applicable to complex robotic and telerobotic systems involving multiple feedback sensors and multiple command variables. In the controller, the learning algorithm is used to learn to reproduce the nonlinear relationship between the sensor outputs and the system command variables over particular regions of the system state space, rather than learning the actuator commands required to perform a specific task. The learned information is used to predict the command signals required to produce desired changes in the sensor outputs. The desired sensor output changes may result from automatic trajectory planning or may be derived from interactive input from a human operator. The learning controller requires no a priori knowledge of the relationships between the sensor outputs and the command variables. The algorithm is well suited for real time implementation, requiring only fixed point addition and logical operations. The results of learning experiments using a General Electric P-5 manipulator interfaced to a VAX-11/730 computer are presented. These experiments involved interactive operator control, via joysticks, of the position and orientation of an object in the field of view of a video camera mounted on the end of the robot arm.

Miller, W. T., III↗
DOE OSTI2025

Karhunen–Loève deep learning method for surrogate modeling and approximate Bayesian parameter estimation

We evaluate the performance of the Karhunen-Loève Deep Neural Network (KL-DNN) framework for surrogate modeling and approximate Bayesian parameter estimation in partial differential equation models. In the surrogate model, the Karhunen-Loève (KL) expansions are used for the dimensionality reduction of the number of unknown parameters and variables, and a deep neural network is employed to relate the reduced space of parameters to that of the state variables. The KL-DNN surrogate model is used to formulate a maximum-a-posteriori-like least-squares problem, which is randomized to draw samples of the posterior distribution of the parameters. We test the proposed framework for a hypothetical unconfined aquifer via comparison with the forward MODFLOW and inverse PEST++ iterative ensemble smoother (IES) solutions as well as the state-of-the-art Fourier neural operator (FNO) and deep operator networks (DeepONets) operator learning surrogate models. Our results show that the KL-DNN surrogate model outperforms FNO and DeepONet for forward predictions. For solving inverse problems, the randomized algorithm provides the same or more accurate Bayesian predictions of the parameters than IES as evidenced by the higher log-predictive probability of both the estimated parameter field and the forecast hydraulic head. The posterior mean obtained from the randomized algorithm is closer to the reference parameter field than that obtained with FNO as the maximum a posteriori estimate.

Approximate Bayesian inference↗
DOE OSTI2026

Learning turbulent flows with generative models for super resolution and sparse flow reconstruction

Neural operators are promising surrogates for dynamical systems but when trained with standard L 2 losses they tend to oversmooth fine-scale turbulent structures. Here, we show that combining operator learning with generative modeling overcomes this limitation. We consider three practical turbulent-flow challenges where conventional neural operators fail: spatio-temporal super-resolution, forecasting, and sparse flow reconstruction. For Schlieren jet super-resolution, an adversarially trained neural operator (adv-NO) reduces the energy-spectrum error by 15 × while preserving sharp gradients at neural operator-like inference cost. For 3D homogeneous isotropic turbulence, adv-NO trained on only 160 timesteps from a single trajectory forecasts accurately for five eddy-turnover times and offers 114 × wall-clock speed-up at inference than the baseline diffusion-based forecasters, enabling near-real-time rollouts. For reconstructing cylinder wake flows from highly sparse Particle Tracking Velocimetry-like inputs, a conditional generative model infers full 3D velocity and pressure fields with correct phase alignment and statistics. These advances enable accurate reconstruction and forecasting at low compute cost, bringing near-real-time analysis and control within reach in experimental and computational fluid mechanics.

Fluid dynamics↗
DOE OSTI2025

Neural operators for stochastic modeling of nonlinear structural system response to natural hazards

Traditionally, neural networks have been employed to learn the mapping between finite-dimensional Euclidean spaces. However, recent research has opened up new horizons, focusing on the utilization of deep neural networks to learn operators capable of mapping infinite-dimensional function spaces. Here, in this work, we employ two state-of-the-art neural operators, the deep operator network (DeepONet) and the Fourier neural operator (FNO) for the prediction of the nonlinear time history response of structural systems exposed to natural hazards, such as earthquakes and windstorms. Specifically, we propose two architectures, a self-adaptive FNO and a fast Fourier transform-based DeepONet (DeepFNOnet), where we employ a FNO beyond the DeepONet to learn the discrepancy between the ground truth and the solution predicted by the DeepONet. To demonstrate the efficiency and applicability of the architectures, two problems are considered. In the first, we use the proposed model to predict the seismic nonlinear dynamic response of a six-story shear building subject to stochastic ground motions. In the second problem, we employ the operators to predict the wind-induced nonlinear dynamic response of a high-rise building while explicitly accounting for the stochastic nature of the wind excitation. In both cases, the trained metamodels achieve high accuracy while being orders of magnitude faster than their corresponding high-fidelity models.

DeepONet↗
DOE OSTI2026

U-net architected deep material network training with microstructure local field information

The Deep Material Network (DMN) has recently emerged as a powerful reduced-order modeling framework for simulating the mechanical response of heterogeneous materials such as composites. Unlike most data-driven approaches that directly learn a material’s response under prescribed loading, the DMN acts as a homogenization operator, learning the kinematic constraints and mechanical interactions of the underlying microstructure. However, traditional DMN training relies exclusively on homogenized effective properties derived from Direct Numerical Simulations (DNS), discarding the rich local field data that govern microstructural interactions. In this work, we extend the DMN framework to incorporate such local field information into the offline training process. Utilizing a U-Net architecture, we augment the DMN training objective to include the first and second statistical moments of the local stress fields obtained from linear DNS. This ensures that the learned network topology not only fits the effective stiffness but also accurately reflects the internal local stress and strain partitioning of the microstructure. The results confirm that supervising the localization process during training yields a superior surrogate model, reducing local prediction errors by an order of magnitude and significantly improving generalization to unseen nonlinear constitutive behaviors compared to traditional DMNs.

36 MATERIALS SCIENCE↗
DOE OSTI2024

MLOps for Beam Controls

Machine learning operations (MLOps) is the standardization and streamlining of the ML development lifecycle to address the challenges associated with large-scale machine learning applications. The full MLOps pipeline consists of open-source tools: DataHub, MinIO and MLflow. It is being used for dataset management and model development to handle changing data dependencies, varying business needs, reproducibility, and diverse teams working with differing tools and skills. To demonstrate the completion of an MLOps pipeline for particle accelerator operations, we are deploying a simple script that computes settings for the Booster’s gradient magnet power supply. Once the demonstration is complete, we will develop and deploy ML-based optimization algorithms to improve Booster’s overall efficiency. This MLOps pipeline opens the gate to systematically develop and deploy ML applications for accelerator controls and diagnostics.

43 PARTICLE ACCELERATORS↗
DOE OSTI2024

MLOps for Beam Controls

Machine learning operations (MLOps) is the standardization and streamlining of the ML development lifecycle to address the challenges associated with large-scale machine learning applications. The full MLOps pipeline consists of open-source tools: DataHub, MinIO and MLflow. It is being used for dataset management and model development to handle changing data dependencies, varying business needs, reproducibility, and diverse teams working with differing tools and skills. To demonstrate the completion of an MLOps pipeline for particle accelerator operations, we are deploying a simple script that computes settings for the Booster’s gradient magnet power supply. Once the demonstration is complete, we will develop and deploy ML-based optimization algorithms to improve Booster’s overall efficiency. This MLOps pipeline opens the gate to systematically develop and deploy ML applications for accelerator controls and diagnostics.

43 PARTICLE ACCELERATORS↗
DOE OSTI2024

MLOps for Beam Controls

Machine learning operations (MLOps) is the standardization and streamlining of the ML development lifecycle to address the challenges associated with large-scale machine learning applications. The full MLOps pipeline consists of open-source tools: DataHub, MinIO and MLflow. It is being used for dataset management and model development to handle changing data dependencies, varying business needs, reproducibility, and diverse teams working with differing tools and skills. To demonstrate the completion of an MLOps pipeline for particle accelerator operations, we are deploying a simple script that computes settings for the Booster’s gradient magnet power supply. Once the demonstration is complete, we will develop and deploy ML-based optimization algorithms to improve Booster’s overall efficiency. This MLOps pipeline opens the gate to systematically develop and deploy ML applications for accelerator controls and diagnostics.

43 PARTICLE ACCELERATORS↗