Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “learning dynamics”

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 1,225 records · Page 68
All sourcesNASA NTRSDOE OSTI
NASA NTRS1991

Intelligent control based on fuzzy logic and neural net theory

In the conception and design of intelligent systems, one promising direction involves the use of fuzzy logic and neural network theory to enhance such systems' capability to learn from experience and adapt to changes in an environment of uncertainty and imprecision. Here, an intelligent control scheme is explored by integrating these multidisciplinary techniques. A self-learning system is proposed as an intelligent controller for dynamical processes, employing a control policy which evolves and improves automatically. One key component of the intelligent system is a fuzzy logic-based system which emulates human decision making behavior. It is shown that the system can solve a fairly difficult control learning problem. Simulation results demonstrate that improved learning performance can be achieved in relation to previously described systems employing bang-bang control. The proposed system is relatively insensitive to variations in the parameters of the system environment.

Lee, Chuen-Chien↗
DOE OSTI2023

Machine learning methods for particle stress development in suspension Poiseuille flows

Numerical simulations are used to study the dynamics of a developing suspension Poiseuille flow with monodispersed and bidispersed neutrally buoyant particles in a planar channel, and machine learning is applied to learn the evolving stresses of the developing suspension. The particle stresses and pressure develop on a slower time scale than the volume fraction, indicating that once the particles reach a steady volume fraction profile, they rearrange to minimize the contact pressure on each particle. Here we consider how the stress development leads to particle migration, time scales for stress development, and present a new physics-informed Galerkin neural network that allows for learning the particle stresses when direct measurements are not possible. The particle fluxes are compared with the Suspension Balance Model with good agreement. We show that when stress measurements are possible, the MOR-physics operator learning method can also capture the particle stresses.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

A Machine Learning Assisted Development of a Model for the Populations of Convective and Stratiform Clouds

Abstract Traditional parameterizations of the interaction between convection and the environment have relied on an assumption that the slowly varying large‐scale environment is in statistical equilibrium with a large number of small and short‐lived convective clouds. They fail to capture nonequilibrium transitions such as the diurnal cycle and the formation of mesoscale convective systems as well as observed precipitation statistics and extremes. Informed by analysis of radar observations, cloud‐permitting model simulation, theory, and machine learning, this work presents a new stochastic cloud population dynamics model for characterizing the interactions between convective and stratiform clouds, with the goal of informing the representation of these interactions in global climate models. Fifteen wet seasons of precipitating cloud observations by a C‐band radar at Darwin, Australia are fed into a machine learning algorithm to obtain transition functions that close a set of coupled equations relating large‐scale forcing, mass flux, the convective cell size distribution, and the stratiform area. Under realistic large‐scale forcing, the derived transition functions show that, on the one hand, interactions with stratiform clouds act to dampen the variability in the size and number of convective cells and therefore in the convective mass flux. On the other, for a given convective area fraction, a larger number of smaller cells is more favorable for the growth of stratiform area than a smaller number of larger cells. The combination of these two factors gives rise to solutions with a few convective cells embedded in a large stratiform area, reminiscent of mesoscale convective systems.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRSDate not supplied

Flight Dynamics and Navigation Performance of the BioSentinel Mission

BioSentinel was one of the ten 6U CubeSats launched by the Space Launch System (SLS) as part of the Artemis-I mission on November 16th, 2022. The spacecraft was deployed after the trans-lunar injection imparted by the upper stage of SLS, and performed a lunar flyby that provided the necessary energy to escape the Earth-Moon system into heliocentric space. The BioSentinel mission required flight dynamics support prior to flight to assess the deployment conditions and the trajectory design, as well as navigation support during flight, to provide orbit determination solutions for the mission and the Deep Space Network (DSN). This paper discusses the BioSentinel mission flight dynamics campaign, from mission design to the challenges and lessons learned encountered during flight.

Astrodynamics↗
DOE OSTI2026

Machine learning inversion of interatomic force constants from single-crystal inelastic neutron scattering

Atomic vibrations govern many macroscopic properties of materials, but experiments to comprehensively probe them remain challenging. Inelastic neutron scattering (INS) is a powerful technique to map phonon dispersions in crystals, especially when leveraging modern time-of-flight (ToF) spectrometers with large detectors. However, efficiently and robustly extracting interatomic force constants (FCs) parameterizing phonon dynamics from experimental spectra remains a bottleneck due to the complexity and high dimensionality of ToF INS datasets. Here, we present a machine learning approach for the direct inversion of FCs from single-crystal INS measurements. The framework leverages synthetic training data generated using universal machine-learned force fields and an efficient physics-based forward model. We benchmark two neural architectures–one emphasizing structured latent representation learning and the other direct, supervised spectral regression–across simulated datasets for two materials under idealized and noisy conditions. The latent-representation model is subsequently applied to experimental single-crystal INS data on germanium. The model is shown to reproduce FCs derived from both first-principles simulations and from iterative optimization, and furthermore achieves reliable inference even from sparse, single-orientation measurements representing short data acquisitions. Analysis of the learned latent space reveals semantically continuous and physically interpretable encodings that support strong cross-domain generalization. By bridging theoretical and experimental domains, we establish a path toward rapid inversion of experimental spectra and data-driven interpretation of temperature-dependent lattice dynamics.

42 ENGINEERING↗
DOE OSTI2024

A safe reinforcement learning algorithm for supervisory control of power plants

Traditional control theory-based methods require tailored engineering for each system and constant fine-tuning. In power plant control, one often needs to obtain a precise representation of the system dynamics and carefully design the control scheme accordingly. Model-free Reinforcement learning (RL) has emerged as a promising solution for control tasks due to its ability to learn from trial-and-error interactions with the environment. It eliminates the need for explicitly modeling the environment’s dynamics, which is potentially inaccurate. However, the direct imposition of state constraints in power plant control raises challenges for standard RL methods. To address this, we propose a chance-constrained RL algorithm based on Proximal Policy Optimization for supervisory control. Our method employs Lagrangian relaxation to convert the constrained optimization problem into an unconstrained objective, where trainable Lagrange multipliers enforce the state constraints. In conclusion, our approach achieves the smallest distance of violation and violation rate in a load-follow maneuver for an advanced Nuclear Power Plant design.

constrained optimization↗
DOE OSTI2015

Learning Delayed Influences of Biological Systems

Boolean networks are widely used model to represent gene interactions and global dynamical behavior of gene regulatory networks. To understand the memory effect involved in some interactions between biological components, it is necessary to include delayed influences in the model. In this paper, we present a logical method to learn such models from sequences of gene expression data. This method analyzes each sequence one by one to iteratively construct a Boolean network that captures the dynamics of these observations. To illustrate the merits of this approach, we apply it to learning real data from bioinformatic literature. Using data from the yeast cell cycle, we give experimental results and show the scalability of the method. We show empirically that using this method we can handle millions of observations and successfully capture delayed influences of Boolean networks.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

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↗
DOE OSTI2019

A Hybrid Reinforcement Learning and Cellular Automata Model for Crowd Simulation on the GPU

We present a GPU-based hybrid model for crowd simulations. The model uses reinforcement learning to guide groups of pedestrians towards a goal while adapting to environmental dynamics, and a cellular automaton to describe individual pedestrians’ interactions. In contrast to traditional multi-agent reinforcement learning methods, our model encodes the learned navigation policy into a navigation map, which is used by the cellular automaton’s update rule to calculate the next simulation step. As a result, reinforcement learning is independent of the number of agents, allowing the simulation of large crowds. Implementation of this model on the GPU allows interactive simulations of several hundreds of pedestrians.

Ruiz, Sergio↗
DOE OSTI2022

An Online Approach to Solve the Dynamic Vehicle Routing Problem with Stochastic Trip Requests for Paratransit Services

Many transit agencies operating paratransit and microtransit services have to respond to trip requests that arrive in real-time, which entails solving hard combinatorial and sequential decision-making problems under uncertainty. To avoid decisions that lead to significant inefficiency in the long term, vehicles should be allocated to requests by optimizing a non-myopic utility function or by batching requests together and optimizing a myopic utility function. While the former approach is typically offline, the latter can be performed online. We point out two major issues with such approaches when applied to paratransit services in practice. First, it is difficult to batch paratransit requests together as they are temporally sparse. Second, the environment in which transit agencies operate changes dynamically (e.g., traffic conditions can change over time), causing the estimates that are learned offline to become stale. To address these challenges, we propose a fully online approach to solve the dynamic vehicle routing problem (DVRP) with time windows and stochastic trip requests that is robust to changing environmental dynamics by construction. We focus on scenarios where requests are relatively sparse—our problem is motivated by applications to paratransit services. We formulate DVRP as a Markov decision process and use Monte Carlo tree search to evaluate actions for any given state. Accounting for stochastic requests while optimizing a non-myopic utility function is computationally challenging; indeed, the action space for such a problem is intractably large in practice. To tackle the large action space, we leverage the structure of the problem to design heuristics that can sample promising actions for the tree search. Our experiments using real-world data from our partner agency show that the proposed approach outperforms existing state-of-the-art approaches both in terms of performance and robustness.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2018

Network Scale Travel Time Prediction using Deep Learning

In recent years, deep learning models have been receiving increased attention within the artificial intelligence (AI) community because of their high prediction accuracy. In this paper, two deep learning models, long short-term memory (LSTM) and convolutional neural network (CNN), are proposed to predict travel time in a road network. One major advantage of using deep learning for travel time prediction is that it can make accurate predictions for all the segments in the transportation network with a single model structure, instead of building customized models for each segment separately. The proposed models were evaluated on a transportation network in the City of Saint Louis, Missouri. The prediction results show that deep learning can provide accurate prediction for both congested and uncongested traffic conditions, and can successfully capture the traffic dynamics of unexpected incidents or special events. The study findings show that deep learning offers a promising approach to real-time prediction of travel times on a network scale.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2022

Long-time integration of parametric evolution equations with physics-informed DeepONets

Ordinary and partial differential equations (ODEs/PDEs) play a paramount role in analyzing and simulating complex dynamic processes across all corners of science and engineering. In recent years machine learning tools are aspiring to introduce new effective ways of simulating such equations, however existing approaches are not able to reliably return stable and accurate predictions across long temporal horizons. We aim to address this challenge by introducing an effective framework for learning evolution operators that map random initial conditions to associated ODE/PDE solutions within a short time interval. Such operators can be parametrized by deep neural networks that are trained in an entirely self-supervised manner without requiring one to generate any paired input-output observations. Global long-time predictions across a range of initial conditions can be then obtained by iteratively evaluating the trained model using each prediction as the initial condition for the next evaluation step. Here, this introduces a new approach to temporal domain decomposition that is shown to be effective in performing accurate long-time simulations for a wide range of parametric ODE and PDE systems, from wave propagation, to reaction-diffusion dynamics and stiff chemical kinetics, introducing a new way of rapidly emulating non-equilibrium processes in science and engineering.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

PolyODENet: Deriving mass-action rate equations from incomplete transient kinetics data

Kinetics of a reaction network that follows mass-action rate laws can be described with a system of ordinary differential equations (ODEs) with polynomial right-hand side. However, it is challenging to derive such kinetic differential equations from transient kinetic data without knowing the reaction network, especially when the data are incomplete due to experimental limitations. We introduce a program, PolyODENet, toward this goal. Based on the machine-learning method Neural ODE, PolyODENet defines a generative model and predicts concentrations at arbitrary time. As such, it is possible to include unmeasurable intermediate species in the kinetic equations. Importantly, we have implemented various measures to apply physical constraints and chemical knowledge in the training to regularize the solution space. Here, using simple catalytic reaction models, we demonstrate that PolyODENet can predict reaction profiles of unknown species and doing so even reveal hidden parts of reaction mechanisms.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2019

Structural phase transitions in a MoWSe 2 monolayer: Molecular dynamics simulations and variational autoencoder analysis

Electrical and optoelectronic properties of two-dimensional (2D) transition metal dichalcogenides (TMDCs) can be tuned by exploiting their structural phase transitions. Here semiconducting (2H) to metallic (1T) phase transition is investigated in a strained MoWSe 2 monolayer using molecular dynamics (MD) simulations. Novel intermediate structures called α and β are found between the 2H and 1T phases. These intermediate structures are similar to those observed in a 2D MoS 2 by scanning transmission electron microscopy. Here, a deep generative model, namely the variational autoencoder (VAE) trained by MD data, is used to generate novel heterostructures with α and β interfaces. Quantum simulations based on density functional theory show that these heterostructures are stable and suitable for novel nanoelectronics applications.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2018

Reinforced dynamics for enhanced sampling in large atomic and molecular systems

A new approach for efficiently exploring the configuration space and computing the free energy of large atomic and molecular systems is proposed, motivated by an analogy with reinforcement learning. There are two major components in this new approach. Like metadynamics, it allows for an efficient exploration of the configuration space by adding an adaptively computed biasing potential to the original dynamics. Like deep reinforcement learning, this biasing potential is trained on the fly using deep neural networks, with data collected judiciously from the exploration and an uncertainty indicator from the neural network model playing the role of the reward function. Parameterization using neural networks makes it feasible to handle cases with a large set of collective variables. This has the potential advantage that selecting precisely the right set of collective variables has now become less critical for capturing the structural transformations of the system. Finally, the method is illustrated by studying the full-atom explicit solvent models of alanine dipeptide and tripeptide, as well as the system of a polyalanine-10 molecule with 20 collective variables.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

Stoichiometry dependent properties of cerium hydride: An active learning developed interatomic potential study

Cerium hydride has a variety of interesting properties, including a known lattice contraction and densification with increasing hydrogen content. However, precise stoichiometric control is not experimentally straightforward and ab initio approaches are not computationally feasible for many properties such as melting and low temperature diffusion. Therefore, we develop a machine-learned interatomic potential for cerium hydride that is valid for H to Ce ratios from 2.0 to 3.0. A query-by-committee active learning approach is used to develop the training set. Leveraging classical molecular dynamics simulations, we assess a range of properties and provide fundamental mechanisms for the trends with stoichiometry. Finally, a majority of the properties follow the trend of lattice contraction, being governed by the stronger lattice binding induced by adding octahedral atoms.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Data-driven methods for diffusivity prediction in nuclear fuels

The growth rate of structural defects in nuclear fuels under irradiation is intrinsically related to the diffusion rates of the defects in the fuel lattice. The generation and growth of atomistic structural defects can significantly alter the performance characteristics of the fuel. This alteration of functionality must be accurately captured to qualify a nuclear fuel for use in reactors. Predicting the diffusion coefficients of defects and how they impact macroscale properties such as swelling, gas release, and creep is therefore of significant importance in both the design of new nuclear fuels and the assessment of current fuel types. In this article, we apply data-driven methods focusing on machine learning (ML) to determine various diffusion properties of two nuclear fuels—uranium oxide and uranium nitride. We show that using ML can increase, often significantly, the accuracy of predicting diffusivity in nuclear fuels in comparison to current analytical models. We also illustrate how ML can be used to quickly develop fuel models with parameter dependencies that are more complex and robust than what is currently available in the literature. In conclusion, these results suggest there is potential for ML to accelerate the design, qualification, and implementation of nuclear fuels.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
NASA NTRS2017

Work Term Assignment Spring 2017

My tour in the Engineering Robotics directorate exceeded my expectations. I learned lessons about Creo, manufacturing and assembly, collaboration, and troubleshooting. During my first tour, last spring, I used Creo on a smaller project, but had limited experience with it before starting in the Dynamic Systems Test branch this spring. I gained valuable experience learning assembly design, sheet metal design and designing with intent for manufacturing and assembly. These skills came from working both on the hatch and the floor. I also learned to understand the intent of other designers on models I worked with. While redesigning the floor, I was modifying an existing part and worked to understand what the previous designer had done to make it fit with the new model. Through working with the machine shop and in the mock-up, I learned much more about manufacturing and assembly. I used a Dremel, rivet gun, belt sander, and countersink for the first time. Through taking multiple safety training for different machine shops, I learned new machine shop safety skills specific to each one. This semester also gave me new collaborative opportunities. I collaborated with engineers within my branch as well as with Human Factors and the building 10 machine shop. This experience helped me learn how to design for functionality and assembly, not only for what would be easiest in my designs. In addition to these experiences, I learned many lessons in troubleshooting. I was the first person in my office to use a Windows 10 computer. This caused unexpected issues with NASA services and programs, such as the Digital Data Management Server (DDMS). Because of this, I gained experience finding solutions to lockout and freeze issues as well as Creo specific settings. These will be useful skills to have in the future and will be implemented in future rotations. This co-op tour has motivated me more to finish my degree and pursue my academic goals. I intend to take a machining Career Gateway Elective in the Fall to improve my skills in building as well as designing with manufacturing intent. I am also inspired to take more mechatronics CGE courses before I graduate to learn more about the crossover between mechanical and electrical engineering. This semester, I worked on multiple projects and had the opportunity to learn from engineers of different disciplines. I became proficient in Creo 2.0, a program I had not used significantly before. I finished modifying the hatch for 3D print and made sizeable modifications to the nose floor support design. I also gained hands-on experience that will be useful in my engineering career in the future. I would consider all of these major achievements from this spring semester. Lastly, I learned to ask more questions and to search for the right people to find answers which I know will be a valuable skill in the future. This summer, I will be completing my last rotation in the Flight Operations Directorate at Ellington Air Field. After this last tour, I will be returning to school for the Fall, Spring and Summer. I will graduate in August of 2018. I am looking forward to learning more about the different jobs available to engineers. This division works directly with many different types of aircrafts and I am excited to learn more about this focus of engineering. The Aircraft Operations Division values team work greatly and I intend to improve my interpersonal skills by working with them this summer.

Sico, Mallory↗