Search NASA⌕ Search

SEARCH · Search NASA

Results for “train simulation”

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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

AN AUTOMATED MACHINE LEARNING-GENETIC ALGORITHM FRAMEWORK WITH ACTIVE LEARNING FOR DESIGN OPTIMIZATION

The use of machine learning (ML)-based surrogate models is a promising technique to significantly accelerate simulation-driven design optimization of internal combustion (IC) engines, due to the high computational cost of running computational fluid dynamics (CFD) simulations. However, training the ML models requires hyperparameter selection, which is often done using trial-and-error and domain expertise. Another challenge is that the data required to train these models are often unknown a priori. In this work, we present an automated hyperparameter selection technique coupled with an active learning approach to address these challenges. The technique presented in this study involves the use of a Bayesian approach to optimize the hyperparameters of the base learners that make up a super learner model. In addition to performing hyperparameter optimization (HPO), an active learning approach is employed, where the process of data generation using simulations, ML training, and surrogate optimization is performed repeatedly to refine the solution in the vicinity of the predicted optimum. The proposed approach is applied to the optimization of a compression ignition engine with control parameters relating to fuel injection, in-cylinder flow, and thermodynamic conditions. It is demonstrated that by automatically selecting the best values of the hyperparameters, a 1.6% improvement in merit value is obtained, compared to an improvement of 1.0% with default hyperparameters. Overall, the framework introduced in this study reduces the need for technical expertise in training ML models for optimization while also reducing the number of simulations needed for performing surrogate-based design optimization.

Owoyele, Opeoluwa↗
DOE OSTI2021

In situ Parallel Training of Analog Neural Network Using Electrochemical Random-Access Memory

In-memory computing based on non-volatile resistive memory can significantly improve the energy efficiency of artificial neural networks. However, accurate in situ training has been challenging due to the nonlinear and stochastic switching of the resistive memory elements. One promising analog memory is the electrochemical random-access memory (ECRAM), also known as the redox transistor. Its low write currents and linear switching properties across hundreds of analog states enable accurate and massively parallel updates of a full crossbar array, which yield rapid and energy-efficient training. While simulations predict that ECRAM based neural networks achieve high training accuracy at significantly higher energy efficiency than digital implementations, these predictions have not been experimentally achieved. In this work, we train a 3 × 3 array of ECRAM devices that learns to discriminate several elementary logic gates (AND, OR, NAND). We record the evolution of the network’s synaptic weights during parallel in situ (on-line) training, with outer product updates. Due to linear and reproducible device switching characteristics, our crossbar simulations not only accurately simulate the epochs to convergence, but also quantitatively capture the evolution of weights in individual devices. The implementation of the first in situ parallel training together with strong agreement with simulation results provides a significant advance toward developing ECRAM into larger crossbar arrays for artificial neural network accelerators, which could enable orders of magnitude improvements in energy efficiency of deep neural networks.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1990

Ascent/abort training in the Shuttle Mission Simulator

The Shuttle Mission Simulator is the primary tool employed in the training of orbiter flight crews in ascent and abort flight techniques at the Johnson Space Center. The simulator uses a motion system, visual systems, and a high-fidelity cockpit to enhance accuracy of training. A team concept is utilized to train flight crews and controllers for each mission to ensure a complete understanding in both orbiter systems and mission specific flight profiles. Thus, each flight crew and flight control team is prepared for the actual flight by practicing repeatedly both the nominal ascent and each type of possible abort situation for that Shuttle mission.

Sims, John T.↗
NASA NTRS1992

Helicopter simulation: Making it work

The opportunities for improved training and checking by using helicopter simulators are greater than they are for airplane pilot training. Simulators permit the safe creation of training environments that are conducive to the development of pilot decision-making, situational awareness, and cockpit management. This paper defines specific attributes required in a simulator to meet a typical helicopter operator's training and checking objectives.

Payne, Barry↗
NASA NTRS1987

Flight simulator platform motion and air transport pilot training

The effect of a flight simulator platform motion on the performance and training of a pilot was evaluated using subjective ratings and objective performance data obtained on experienced B-727 pilots and pilots with no prior heavy aircraft flying experience flying B-727-200 aircraft simulator used by the FAA in the upgrade and transition training for air carrier operations. The results on experienced pilots did not reveal any reliable effects of wide variations in platform motion design. On the other hand, motion variations significantly affected the behavior of pilots without heavy-aircraft experience. The effect was limited to pitch attitude control inputs during the early phase of landing training.

Lee, Alfred T.↗
NASA NTRS2007

Development and Assessment of a Novel Training Package for Basic Maneuvering Tasks on a Flight Simulator Using Self Instruction Methods and Above Real Time Training (ARTT)

One result of the relatively recent advances in computing technology has been the decreasing cost of computers and increasing computational power. This has allowed high fidelity airplane simulations to be run on personal computers (PC). Thus, simulators are now used routinely by pilots to substitute real flight hours for simulated flight hours for training for an aircraft type rating thereby reducing the cost of flight training. However, FAA regulations require that such substitution training must be supervised by Certified Flight Instructors (CFI). If the CFI presence could be reduced or eliminated for certain tasks this would mean a further cost savings to the pilot. This would require that the flight simulator have a certain level of 'intelligence' in order to provide feedback on pilot performance similar to that of a CFI. The 'intelligent' flight simulator would have at least the capability to use data gathered from the flight to create a measure for the performance of the student pilot. Also, to fully utilize the advances in computational power, the simulator would be capable of interacting with the student pilot using the best possible training interventions. This thesis reports on the two studies conducted at Tuskegee University investigating the effects of interventions on the learning of two flight maneuvers on a flight simulator and the robustness and accuracy of calculated performance indices as compared to CFI evaluations of performance. The intent of these studies is to take a step in the direction of creating an 'intelligent' flight simulator. The first study deals with the comparisons of novice pilot performance trained at different levels of above real-time to execute a level S-turn. The second study examined the effect of out-of-the-window (OTW) visual cues in the form of hoops on the performance of novice pilots learning to fly a landing approach on the flight simulator. The reliability/robustness of the computed performance metrics was assessed by comparing them with the evaluations of the landing approach maneuver by a number of CFIs.

Ali, Syed Firasat↗
NASA NTRS2005

Return to Flight: Crew Activities Resource Reel 1 of 2

The crew of the STS-114 Discovery Mission is seen in various aspects of training for space flight. The crew activities include: 1) STS-114 Return to Flight Crew Photo Session; 2) Tile Repair Training on Precision Air Bearing Floor; 3) SAFER Tile Inspection Training in Virtual Reality Laboratory; 4) Guidance and Navigation Simulator Tile Survey Training; 5) Crew Inspects Orbital Boom and Sensor System (OBSS); 6) Bailout Training-Crew Compartment; 7) Emergency Egress Training-Crew Compartment Trainer (CCT); 8) Water Survival Training-Neutral Buoyancy Lab (NBL); 9) Ascent Training-Shuttle Motion Simulator; 10) External Tank Photo Training-Full Fuselage Trainer; 11) Rendezvous and Docking Training-Shuttle Engineering Simulator (SES) Dome; 12) Shuttle Robot Arm Training-SES Dome; 13) EVA Training Virtual Reality Lab; 14) EVA Training Neutral Buoyancy Lab; 15) EVA-2 Training-NBL; 16) EVA Tool Training-Partial Gravity Simulator; 17) Cure in Place Ablator Applicator (CIPAA) Training Glove Vacuum Chamber; 16) Crew Visit to Merritt Island Launch Area (MILA); 17) Crew Inspection-Space Shuttle Discovery; and 18) Crew Inspection-External Tank and Orbital Boom and Sensor System (OBSS). The crew are then seen answering questions from the media at the Space Shuttle Landing Facility.

Source record↗
DOE OSTI2025

DeepLensSBI: Deep inference of simulated strong lenses in ground-based surveys

This code is used to train and test machine learning models and generate results and plots presented in 2501.08524 [astro-ph.IM]. The code is written in python. The goal of this work is to train ML models trained on simulated images of strong gravitational lenses. The trained model can then quickly infer properties of the lensed objects with uncertainty quantification.

Poh, Jason [Univ. of Chicago, IL (United States)] ↗
DOE OSTI2020

Machine learning approaches for structural and thermodynamic properties of a Lennard-Jones fluid

Predicting the functional properties of many molecular systems relies on understanding how atomistic interactions give rise to macroscale observables. However, current attempts to develop predictive models for the structural and thermodynamic properties of condensed-phase systems often rely on extensive parameter fitting to empirically selected functional forms whose effectiveness is limited to a narrow range of physical conditions. Here, we illustrate how these traditional fitting paradigms can be superseded using machine learning. Specifically, we use the results of molecular dynamics simulations to train machine learning protocols that are able to produce the radial distribution function, pressure, and internal energy of a Lennard-Jones fluid with increased accuracy in comparison to previous theoretical methods. The radial distribution function is determined using a variant of the segmented linear regression with the multivariate function decomposition approach developed by Craven et al. [J. Phys. Chem. Lett. 11, 4372 (2020)]. The pressure and internal energy are determined using expressions containing the learned radial distribution function and also a kernel ridge regression process that is trained directly on thermodynamic properties measured in simulation. The presented results suggest that the structural and thermodynamic properties of fluids may be determined more accurately through machine learning than through human-guided functional forms.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

HYPHY: Deep Generative Conditional Posterior Mapping of Hydrodynamical Physics

Generating large-volume hydrodynamical simulations for cosmological observables is a computationally demanding task necessary for next-generation observations. In this work, we construct a novel fully convolutional variational autoencoder (VAE) to synthesize hydrodynamic fields conditioned on dark matter fields from N -body simulations. After training the model on a single hydrodynamical simulation, we are able to probabilistically map new dark-matter-only simulations to corresponding full hydrodynamical outputs. By sampling over the latent space of our VAE, we can generate posterior samples and study the variance of the mapping. We find that our reconstructed field provides an accurate representation of the target hydrodynamical fields as well as reasonable variance estimates. This approach has promise for the rapid generation of mocks as well as for implementation in a full inverse model of observed data.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS2021

Feasibility of Earthbound Motion in Lunar Gravity

BACKGROUND: Marginal stability of astronaut movement while performing lunar surface tasks has been well documented, and is clearly demonstrated in videos of falls, and near falls, during Apollo Lunar Extravehicular Activities. Referencing mission reports from Apollo 15 and 16 [1, 2], suspected causes for falls include: surface conditions, visibility, and gravitational effects (hypogravity). In this preliminary test, we employ the open-source biomechanical tool OpenSim [3, 4] to analyze the impact of lunar gravity (Lg) on two object-pickup motions performed by a single shirt-sleeved subject. Specifically, we attempt to answer the following questions based on an estimation of the Center of Mass Projection (CoMP) and ground reaction force Center of Pressure (COP) as it relates to the astronaut Base of Support (BOS) for 1g and Lg conditions: 1. Is the task motion, as performed in 1g, dynamically feasible in 1g and Lg? 2. Can we make the motion dynamically feasible in Lg by slowing it down? 3. Is the Lg COP equal to the 1g COP at a theoretically predicted reduction in motion speed? METHODS AND RESULTS: To answer the first question, the gravitational acceleration in the OpenSim model is modified from a nominal 9.81 m/s2 to 1.64 m/s2, and the 1g joint trajectory is input to an OpenSim based method [5] for estimating ground reaction forces and moments. From this method, the position of the CoMP and COP can be estimated and checked to see if they remain within a simulated BOS formed from the footprint of the OpenSim model to determine whether the motion is dynamically feasible. As expected, both of the motions were estimated to be feasible in a 1g environment, however, both motions had periods of infeasibility in Lg. It is well known that crew members make adjustments to motion trajectories in altered gravity fields to maintain balance. As a first step, we considered the simple adjustment of slowing the motion in Lg by a constant factor. This was accomplished by scaling the time stamps in the motion trajectory file by that factor. For the two motions considered, it was found that scale factors of 1.3 and 1.4 kept the COP just within the subject BOS. The CoMP is unchanged by the gravity field. Simple analysis of an inverted pendulum in the Lg environment, which generalizes to a general multibody system, leads to a theoretical prediction that a reduction in speed factor of √1g/Lg, or 2.445, will make the COP trajectory in Lg equivalent to that in 1g. When the above procedure was performed with a factor of 2.445, the estimated COP in Lg, was observed to be very close to that in 1g. In summary, we have developed a method for estimating the CoMP and COP in Lg, for subject motion collected in 1g. We believe this method can prove to be a valuable check and balance for simulated Lg training and testing by exposing potential simulator-induced artifacts that make the simulated task motion seem possible, when in fact, it would violate the above criteria. We also note that a reduction in task speed should tend the task motion towards stability, with a theoretical slowdown factor of √1g/Lg making the motion stability equal to that in 1g according to the CoMP and COP criteria.

R K Huffman↗
DOE OSTI2024

Resilient Control of Networked Microgrids Using Vertical Federated Reinforcement Learning: Designs and Real-Time Test-Bed Validations

Improving system-level resiliency of networked microgrids against adversarial cyber-attacks is an important aspect in the current regime of increased inverter-based resources (IBRs). To achieve that, this paper contributes in designing a hierarchical control layer, in conjunction with the existing control layers, resilient to adversarial attack signals. Considering model complexities, unknown dynamical behaviors of IBRs, and privacy issues regarding data sharing in multi-party-owned microgrids, designing such a control layer is non-trivial. Here, to tackle these issues, a novel federated reinforcement learning (Fed-RL) method is proposed. To grasp the interconnected dynamics of networked microgrids, the paper develops Federated Soft Actor-Critic (FedSAC) algorithm following the vertical structure of implementing Fed-RL. Next, utilizing the OpenAI Gym interface, we built a custom set-up in GridLAB-D/HELICS co-simulation platform, named Resilient RL Co-simulation (ResRLCoSIM), to train the RL agents with IEEE 123-bus benchmark comprising 3 interconnected microgrids. Finally, the learned policies in the simulation are transferred to the real-time hardware-in-the-loop (HIL) test-bed developed using the high-fidelity Hypersim platform. Finally, experiments show that the simulator-trained RL controllers achieve desirable performance with the test-bed platform, validating the minimization of the sim-to-real gap.

24 POWER TRANSMISSION AND DISTRIBUTION↗
NASA NTRS1981

Line-oriented flight training: Northwest Airlines

An exemption from certain FAA regulations which stereotype simulator flight training was obtained and pilots with current line experience were used to prepare and develop scenarios for a program in which each crew member would be trained to recognize and properly use all available resouces. The development of the scenarios for training to proficiency and pilot reaction to the training sessions are discussed.

Nunn, H. T.↗
DOE OSTI2022

Adversarial autoencoder ensemble for fast and probabilistic reconstructions of few-shot photon correlation functions for solid-state quantum emitters

Second-order photon correlation measurements [g (2) (τ) functions] are widely used to classify single-photon emission purity in quantum emitters or to measure the multiexciton quantum yield of emitters that can simultaneously host multiple excitations – such as quantum dots – by evaluating the value of g (2) (τ = 0). Accumulating enough photons to accurately calculate this value is time consuming and could be accelerated by fitting of few-shot photon correlations. Here, we develop an uncertainty-aware, deep adversarial autoencoder ensemble (AAE) that reconstructs noise-free g (2) (τ) functions from noise-dominated, few-shot inputs. The model is trained with simulated g (2) (τ) functions that are facilely generated by Poisson sampling time bins. The AAE reconstructions are performed orders-of-magnitude faster, with reconstruction errors and estimates of g (2) (τ = 0) that are lower in variance and similar in accuracy compared to Maximum likelihood estimation and Levenberg-Marquardt least-squares fitting approaches, for simulated and experimentally measured few-shot g (2) (τ) functions (~100 two-photon events) of InP/ZnS/ZnSe and CdS/CdSe/CdS quantum dots. The deep-ensemble model comprises eight individual autoencoders, allowing for probabilistic reconstructions of noise-free g (2) (τ) functions, and we show that the predicted variance scales inversely with number of shots, with comparable uncertainties to computationally intensive Markov chain Monte Carlo sampling. Furthermore, this work demonstrates the advantage of machine learning models to perform uncertainty-aware, fast, and accurate reconstructions of simple Poisson-distributed photon correlation functions, allowing for on-the-fly reconstructions and accelerated materials characterization of solid-state quantum emitters.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
NASA NTRS1973

Apollo experience report: Simulation of manned space flight for crew training

Through space-flight experience and the development of simulators to meet the associated training requirements, several factors have been established as fundamental for providing adequate flight simulators for crew training. The development of flight simulators from Project Mercury through the Apollo 15 mission is described. The functional uses, characteristics, and development problems of the various simulators are discussed for the benefit of future programs.

Woodling, C. H.↗
DOE OSTI2021

Condensed Phase Water Molecular Multipole Moments from Deep Neural Network Models Trained on Ab Initio Simulation Data

Ionic solvation phenomena in liquids involve intense interactions in the inner solvation shell. For interactions beyond the first shell, the ion–solvent interaction energies result from the sum of many smaller-magnitude contributions that can still include polarization effects. Deep neural network (DNN) methods have recently found wide application in developing efficient molecular models that maintain near-quantum accuracy. In this work, we extend the DeePMD-kit code to produce accurate molecular multipole moments in the bulk and near interfaces. The new method is validated by comparing the DNN moments with those generated by ab initio simulations. The moments are used to compute the electrostatic potential at the center of a molecular-sized hydrophobic cavity in water. The results show that the fields produced by the DNN models are in quantitative agreement with the AIMD-derived values. These efficient methods will open the door to more accurate solvation models for large solutes such as proteins.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Deep learning enhanced super-resolution x-ray fluorescence microscopy by a dual-branch network

X-ray fluorescence (XRF) microscopy is a powerful technique for quantifying the distribution of elements in complex materials, which makes it a crucial imaging technique across a wide range of disciplines in physical and biological sciences, including chemistry, materials science, microbiology, and geosciences. However, as a scanning microscopy technique, the spatial resolution of XRF imaging is inherently constrained by the x-ray probe profile and scanning step size. Here we propose a dual-branch machine learning (ML) model, which can extract scale-variant features and bypass abundant low-frequency information separately, to enhance the spatial resolution of the XRF images by mitigating the effects of blurring from the probe profile. The model is trained by simulated natural images, and a two-stage training strategy is used to overcome the domain gap between the natural images and experimental data. The tomography reconstruction from enhanced XRF projections shows an improvement in resolution by a scale factor of four and reveals distinct internal features invisible in low-resolution XRF within a battery sample. This study offers a promising approach for obtaining high-resolution XRF imaging from its low-resolution version, paving the way for future investigations in a broader range of disciplines and materials.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Power System Event Classification and Localization Using a Convolutional Neural Network

Detection and timely identification of power system disturbances are essential for situation awareness and reliable electricity grid operation. Because records of actual events in the system are limited, ensemble simulation-based events are needed to provide adequate data for building event-detection models through deep learning; e.g., a convolutional neural network (CNN). An ensemble numerical simulation-based training data set have been generated through dynamic simulations performed on the Polish system with various types of faults in different locations. Such data augmentation is proven to be able to provide adequate data for deep learning. The synchronous generators’ frequency signals are used and encoded into images for developing and evaluating CNN models for classification of fault types and locations. With a time-domain stacked image set as the benchmark, two different time-series encoding approaches, i.e., wavelet decomposition-based frequency-domain stacking and polar coordinate system-based Gramian Angular Field (GAF) stacking, are also adopted to evaluate and compare the CNN model performance and applicability. The various encoding approaches are suitable for different fault types and spatial zonation. With optimized settings of the developed CNN models, the classification and localization accuracies can go beyond 84 and 91%, respectively.

24 POWER TRANSMISSION AND DISTRIBUTION↗