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 703 records · Page 39
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Machine Learning Potentials with the Iterative Boltzmann Inversion: Training to Experiment

Methodologies for training machine learning potentials (MLPs) with quantum-mechanical simulation data have recently seen tremendous progress. Experimental data have a very different character than simulated data, and most MLP training procedures cannot be easily adapted to incorporate both types of data into the training process. Here, we investigate a training procedure based on iterative Boltzmann inversion that produces a pair potential correction to an existing MLP using equilibrium radial distribution function data. By applying these corrections to an MLP for pure aluminum based on density functional theory, we observe that the resulting model largely addresses previous overstructuring in the melt phase. Interestingly, the corrected MLP also exhibits improved performance in predicting experimental diffusion constants, which are not included in the training procedure. The presented method does not require autodifferentiating through a molecular dynamics solver and does not make assumptions about the MLP architecture. Our results suggest a practical framework for incorporating experimental data into machine learning models to improve the accuracy of molecular dynamics simulations.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Computationally efficient CFD prediction of bubbly flow using physics-guided deep learning

To realize efficient computational fluid dynamics (CFD) prediction of two-phase flow, a multi-scale framework was proposed in this paper by applying a physics-guided data-driven approach. Instrumental to this framework, Feature Similarity Measurement (FSM) technique was developed for error estimation in two-phase flow simulation using coarse-mesh CFD, to achieve a comparable accuracy as fine-mesh simulations with fast-running feature. In this work, by defining physics-guided parameters and variable gradients as physical features, FSM has the capability to capture the underlying local patterns in the coarse-mesh CFD simulation. Massive low-fidelity data and respective high-fidelity data are used to explore the underlying information relevant to the main simulation errors and the effects of phenomenological scaling. By learning from previous simulation data, a surrogate model using deep feedforward neural network (DFNN) can be developed and trained to estimate the simulation error of coarse-mesh CFD. In a demonstration case of two-phase bubbly flow, the DFNN model well captured and corrected the unphysical “peaks” in the velocity and void fraction profiles near the wall in the coarse-mesh configuration, even for extrapolative predictions. The research documented supports the feasibility of the physics-guided deep learning methods for coarse mesh CFD simulations which has a potential for the efficient industrial design.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2020

DEMONSTRATION OF A DATA-DRIVEN PHYSICS-BASED APPROACH FOR COMPUTATIONALLY EFFICIENT CFD PREDICTION OF TWO-PHASE BUBBLY FLOW

To realize efficient computational fluid dynamics (CFD) prediction of two-phase flow, a multi-scale physics-guided data-driven approach, Feature Similarity Measurement (FSM) technique was developed for error estimation in two-phase flow simulation using coarse-mesh CFD, to achieve a comparable accuracy as fine-mesh simulations with fast-running feature. By defining physics-guided parameters and variable gradients as physical features, FSM has the capability to capture the underlying local patterns in the coarse-mesh CFD simulation. Massive low-fidelity data and respective high-fidelity data are used to explore the underlying information relevant to the main simulation errors and the effects of phenomenological scaling. By learning from previous simulation data, a surrogate model using deep feedforward neural network (DFNN) can be developed and trained to estimate the simulation error of coarse-mesh CFD. In a demonstration case of two-phase bubbly flow, the DFNN model well captured and corrected the unphysical “peaks” in the velocity and void fraction profiles near the wall in the coarse-mesh configuration, even for extrapolative predictions. The research documented supports the feasibility of the physics-guided deep learning methods for coarse mesh CFD simulations which has a potential for the efficient industrial design.

42 ENGINEERING↗
NASA NTRS1993

Training high performance skills using above real-time training

The Above Real-Time Training (ARTT) concept is a unique approach to training high performance skills. ARTT refers to a training paradigm that places the operator in a simulated environment that functions at faster than normal time. Such a training paradigm represents a departure from the intuitive, but not often supported, feeling that the best practice is determined by the training environment with the highest fidelity. This approach is hypothesized to provide greater 'transfer value' per simulation trial, by incorporating training techniques and instructional features into the simulator. These techniques allow individuals to acquire these critical skills faster and with greater retention. ARTT also allows an individual trained in 'fast time' to operate at what appears to be a more confident state, when the same task is performed in a real-time environment. Two related experiments are discussed. The findings appear to be consistent with previous findings that show positive effects of task variation during training. Moreover, ARTT has merit in improving or maintaining transfer with sharp reductions in training time. There are indications that the effectiveness of ARTT varies as a function of task content and possibly task difficulty. Other implications for ARTT are discussed along with future research directions.

Guckenberger, Dutch↗
NASA NTRS1974

Simulator verification techniques study. Integrated simulator self test system concepts

Software and hardware requirements for implementing hardware self tests are presented in support of the development of training and procedures development simulators for the space shuttle program. Self test techniques for simulation hardware and the validation of simulation performance are stipulated. The requirements of an integrated simulator self system are analyzed. Readiness tests, fault isolation tests, and incipient fault detection tests are covered.

Montoya, G.↗
DOE OSTI2024

Accurate numerical simulations of open quantum systems using spectral tensor trains

Decoherence between qubits is a major bottleneck in quantum computations. Decoherence results from intrinsic quantum and thermal fluctuations as well as noise in the external fields that perform the measurement and preparation processes. With prescribed colored noise spectra for intrinsic and extrinsic noise, we present a numerical method, Quantum Accelerated Stochastic Propagator Evaluation (Q-ASPEN), to solve the time-dependent noise-averaged reduced density matrix in the presence of intrinsic and extrinsic noise. Q-ASPEN is arbitrarily accurate and can be applied to provide estimates for the resources needed to error-correct quantum computations. We employ spectral tensor trains, which combine the advantages of tensor networks and pseudospectral methods, as a variational ansatz to the quantum relaxation problem and optimize the ansatz using methods typically used to train neural networks. Here, the spectral tensor trains in Q-ASPEN make accurate calculations with tens of quantum levels feasible. We present benchmarks for Q-ASPEN on the spin-boson model in the presence of intrinsic noise and on a quantum chain of up to 32 sites in the presence of extrinsic noise. In our benchmark, the memory cost of Q-ASPEN scales as a low-order polynomial in the size of the system once the number of system states surpasses the number of basis functions used in the spectral expansion.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

Calorimetry with deep learning: particle simulation and reconstruction for collider physics

Using detailed simulations of calorimeter showers as training data, we investigate the use of deep learning algorithms for the simulation and reconstruction of single isolated particles produced in high-energy physics collisions. We train neural networks on single-particle shower data at the calorimeter-cell level, and show significant improvements for simulation and reconstruction when using these networks compared to methods which rely on currently-used state-of-the-art algorithms. We define two models: an end-to-end reconstruction network which performs simultaneous particle identification and energy regression of particles when given calorimeter shower data, and a generative network which can provide reasonable modeling of calorimeter showers for different particle types at specified angles and energies. We investigate the optimization of our models with hyperparameter scans. Furthermore, we demonstrate the applicability of the reconstruction model to shower inputs from other detector geometries, specifically ATLAS-like and CMS-like geometries. These networks can serve as fast and computationally light methods for particle shower simulation and reconstruction for current and future experiments at particle colliders.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
NASA NTRS1977

Analytical evaluation of two motion washout techniques

Practical tools were developed which extend the state of the art of moving base flight simulation for research and training purposes. The use of visual and vestibular cues to minimize the actual motion of the simulator itself was a primary consideration. The investigation consisted of optimum programming of motion cues based on a physiological model of the vestibular system to yield 'ideal washout logic' for any given simulator constraints.

Young, L. R.↗
NASA NTRS1999

Landsat-7 Simulation and Testing Environments

A spacecraft Attitude Control and Determination Subsystem (ACDS) is heavily dependent upon simulation throughout its entire development, implementation and ground test cycle. Engineering simulation tools are typically developed to design and analyze control systems to validate the design and software simulation tools are required to qualify the flight software. However, the need for simulation does not end here. Operating the ACDS of a spacecraft on the ground requires the simulation of spacecraft dynamics, disturbance modeling and celestial body motion. Sensor data must also be simulated and substituted for actual sensor data on the ground so that the spacecraft will respond by sending commands to the actuators as they will on orbit. And finally, the simulators is the primary training tool and test-bed for the Flight Operations Team. In this paper various ACDS simulation, developed for or used by the Landsat 7 project will be described. The paper will include a description of each tool, its unique attributes, and its role in the overall development and testing of the ACDS. Finally, a section is included which discusses how the coordinated use of these simulation tools can maximize the probability of uncovering software, hardware and operations errors during the ground test process.

Holmes, E.↗
NASA NTRS2006

Demonstration of Self-Training Autonomous Neural Networks in Space Vehicle Docking Simulations

Neural Networks have been under examination for decades in many areas of research, with varying degrees of success and acceptance. Key goals of computer learning, rapid problem solution, and automatic adaptation have been elusive at best. This paper summarizes efforts at NASA's Marshall Space Flight Center harnessing such technology to autonomous space vehicle docking for the purpose of evaluating applicability to future missions.

Patrick, M. Clinton↗
NASA NTRS2013

Origins and Early History of Underwater Neutral Buoyancy Simulation of Weightlessness for EVA Procedures Development and Training: Winnowing and Regrowth - Part 2

The technique of neutral buoyancy during water immersion was applied to a variety of questions pertaining to human performance factors in the early years of the space age. It was independently initiated by numerous aerospace contractors at nearly the same time, but specific applications depended on the problems that the developers were trying to solve. Those problems dealt primarily with human restraint and maneuverability and were often generic across extravehicular activity (EVA) and intravehicular activity (IVA) worksites. The same groups often also considered fractional gravity as well as weightless settings and experimented with ballasting to achieve lunar and Mars-equivalent loads as part of their on-going research and development. Dr. John Charles reviewed the association of those tasks with contemporary perceptions of the direction of NASA's future space exploration activities and with Air Force assessments of the military value of man in space.

Charles, John B.↗
NASA NTRS2003

Iced Aircraft Flight Data for Flight Simulator Validation

NASA is developing and validating technology to incorporate aircraft icing effects into a flight training device concept demonstrator. Flight simulation models of a DHC-6 Twin Otter were developed from wind tunnel data using a subscale, complete aircraft model with and without simulated ice, and from previously acquired flight data. The validation of the simulation models required additional aircraft response time histories of the airplane configured with simulated ice similar to the subscale model testing. Therefore, a flight test was conducted using the NASA Twin Otter Icing Research Aircraft. Over 500 maneuvers of various types were conducted in this flight test. The validation data consisted of aircraft state parameters, pilot inputs, propulsion, weight, center of gravity, and moments of inertia with the airplane configured with different amounts of simulated ice. Emphasis was made to acquire data at wing stall and tailplane stall since these events are of primary interest to model accurately in the flight training device. Analyses of several datasets are described regarding wing and tailplane stall. Key findings from these analyses are that the simulated wing ice shapes significantly reduced the C , max, while the simulated tail ice caused elevator control force anomalies and tailplane stall when flaps were deflected 30 deg or greater. This effectively reduced the safe operating margins between iced wing and iced tail stall as flap deflection and thrust were increased. This flight test demonstrated that the critical aspects to be modeled in the icing effects flight training device include: iced wing and tail stall speeds, flap and thrust effects, control forces, and control effectiveness.

Ratvasky, Thomas P.↗
DOE OSTI2022

CFD-trained ANN Model for Approximating Near-occupant Condition in Real-time Simulations

The main drawback of Computational Fluid Dynamics (CFD) simulations has been the time and resource consuming nature which is not suitable for real-time applications. In this work, we first generated numerous CFD models of a given indoor space to obtain airspeed, temperature, and mean radiant temperature near an occupant as training data. Several artificial neural networks (ANN) models were trained using this CFD simulated data to approximate near real-time environmental conditions for a given occupant. This trained ANN model approach is a part of a real-time simulation of building operations using a combination of software and real hardware (HVAC equipment) approaches. The preliminary results suggest that the CFD- generated training data and the trained ANN model can accurately approximate such conditions in a real-time application, a method that has great potential in building simulation and building digital twin areas of research.

Zhang, Yun↗
DOE OSTI2025

Reinforcement Learning Control for Buildings Co-Optimizing Energy, Comfort, and Indoor Air Quality: An Annual Assessment

Efficient control of Heating, Ventilation, and Air Conditioning (HVAC) systems is crucial for optimizing energy use and maintaining indoor comfort in buildings. Traditional control methods, such as PID control, cannot handle energy use trade-offs among multiple components in the building energy system at a supervisory level. Reinforcement learning (RL) presents a promising solution, offering adaptive and data-driven control strategies that optimize performance over time. However, RL also faces several challenges, including the conflicts encountered in co-optimizing energy savings, occupant comfort, and indoor air quality, and the requirement for extensive interactions with the environment in training. We proposed a flexible simulation platform that integrates a hybrid model for RL training and designed an RL agent to control the entire central HVAC system, focusing on co-optimizing energy consumption, thermal comfort, and indoor air quality ($\text{CO}_{2}$ and PM2.5 concentrations). Finally, we evaluated the RL agent's performance over an annual cycle. Our findings indicate that the RL agent can effectively manage the HVAC system with 14.7 % energy savings annually and balance multiple objectives, which demonstrates significant potential for improving HVAC system control and sustainability in buildings.

Guo, Fangzhou↗
DOE OSTI2026

Foundation models for atomistic simulation of chemistry and materials

Conventional computational methods for modeling chemical and materials systems are limited by system size and timescale, forcing a trade-off between quantum-mechanical accuracy and the sampling needed for realistic observables. Large language and vision foundation models — pre-trained on massive datasets using transformer architectures — have revolutionized many fields. It is thus interesting to ask whether a foundation model — subject to suitable data, parameter scaling and training — could enable learned simulations of chemistry and materials. Here, in this study, we review the field of machine-learned interatomic potentials (MLIPs) and posit that scaling up large and diverse chemical and materials datasets and highly expressive architectures using advanced training strategies should result in models that are: more efficient, transferable, robust to out-of-distribution scenarios, and easier to fine-tune to a variety of downstream physical observables than models trained from scratch on small datasets corresponding to specific, targeted atomistic simulation tasks. We provide specific criteria for creating such large-scale MLIP foundation models, coordinated strategies for their development, evaluation and deployment, and highlight potential emergent capabilities that could transform predictive simulations in chemistry and materials science and accelerate discovery across multiple technological domains.

Yuan, Eric C.-Y. [University of California, Berkel↗
DOE OSTI2024

Rapid Inverse Parameter Inference Using Physics-Informed Neural Network

As Li-ion batteries become more essential in today's economy, tools need to be developed to accurately and rapidly diagnose a battery's internal state-of-health. Using a Li-ion battery's (high-rate) voltage response, it is proposed to determine a battery's internal state through Bayesian calibration. However, Bayesian calibration is notoriously slow and requires thousands of model runs. To accelerate parameter inference using Bayesian calibration, a surrogate model is developed to replace the underlying physics-based Li-ion model. Developing a surrogate model for rapid Bayesian calibration analysis is discussed for both the single particle model (SPM) and the pseudo two-dimensional (P2D) model. Surrogate models are constructed using physics-informed neural networks (PINNs) that encode the influence of internal properties on observed voltage responses. In practice, a neural network can be trained by: 1) using simulation results of the physics-based model (i.e., a data-loss approach); 2) using the residuals of the governing equations themselves (i.e., a physics-loss approach); or 3) using a combination of simulation results and governing equation residuals. In the present work, PINNs are developed using a variety of training losses and neural network architectures. In this analysis, it is shown that a PINN surrogate model can be reliably trained with only physics-informed loss. However, using a coupled data-informed and physics-loss approach produced the most accurate PINNs.

Bayesian calibration↗