Search NASA⌕ Search

SEARCH · Search NASA

Results for “Complex dynamical systems”

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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

On fast simulation of dynamical system with neural vector enhanced numerical solver

The large-scale simulation of dynamical systems is critical in numerous scientific and engineering disciplines. However, traditional numerical solvers are limited by the choice of step sizes when estimating integration, resulting in a trade-off between accuracy and computational efficiency. To address this challenge, we introduce a deep learning-based corrector called Neural Vector (NeurVec), which can compensate for integration errors and enable larger time step sizes in simulations. Our extensive experiments on a variety of complex dynamical system benchmarks demonstrate that NeurVec exhibits remarkable generalization capability on a continuous phase space, even when trained using limited and discrete data. NeurVec significantly accelerates traditional solvers, achieving speeds tens to hundreds of times faster while maintaining high levels of accuracy and stability. Moreover, NeurVec’s simple-yet-effective design, combined with its ease of implementation, has the potential to establish a new paradigm for fast-solving differential equations based on deep learning.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

On Building Predictive Digital Twin Incorporating Wave Predicting Capabilities: Case Study on UMaine Experimental Campaign - FOCAL

The response of floating wind turbines (FWT) are susceptible to stochastic wave variations. For the optimal operation of FWT, a comprehensive understanding of the phaseresolved wave dynamics and the consequential system response is crucial for real-time monitoring and control. A multi-variate, multi-step, long short term memory (MLSTM), a type of recurrent neural network (RNN) is used to capture complex system dynamics for real-time application. Results indicate that the integration of a wave prediction-reconstruction (WRP) model substantially enhances prediction accuracy by 50% on average relative to the baseline model. The improvement is consistent across various wave extremity and prediction horizons, thereby significantly broadening the scope for timely and precise predictive capabilities.

17 WIND ENERGY↗
NASA NTRS1973

Pole and zero placement in multivariable control systems

A method is proposed for designing multivariable systems based on an alternate derivation of Davison's theorem on pole placement and the solution of the nonlinear equations for the feedback gains by the least square error method. Output feedback is used to control a complex dynamical system. The freedom in design, after allocating poles, is used to place zeros and/or satisfy other design objectives. This method results in algorithms which are computationally attractive. However, this is done at a considerable sacrifice in terms of the design freedom available. For a system with m inputs and p outputs only m + p variables are available instead of mp variables.

Sridhar, B.↗
NASA NTRS2020

Machine Learning for Dynamical Modeling of a Flexible Inverted Pendulum System

The inverted pendulum system, a canonical example of an unstable mechanical system, is often used to model the control problems encountered in the flight of rockets in the initial stages of launch, when the airspeed is too small for aerodynamic stability. A system with a flexible pendulum is a variant that more accurately simulates rocket flight nonlinearities (particularly, the flex modes of the rocket). To increase NASA capability of modeling dynamical systems for which closed form solutions are not clear or easily developed, this project aims to provide a machine learning approach that produces a learned dynamical model of the PENNY robot (a rover with a flexible inverted pendulum) from operational data. The developed approach can then be generalized to other complex dynamical systems, including but not limited to rockets and other robotic systems.

M. A. DuPuis↗
NASA NTRS1990

Experiments in cooperative-arm object manipulation with a two-armed free-flying robot

Developing computed-torque controllers for complex manipulator systems using current techniques and tools is difficult because they address the issues pertinent to simulation, as opposed to control. A new formulation of computed-torque (CT) control that leads to an automated computer-torque robot controller program is presented. This automated tool is used for simulations and experimental demonstrations of endpoint and object control from a free-flying robot. A new computed-torque formulation states the multibody control problem in an elegant, homogeneous, and practical form. A recursive dynamics algorithm is presented that numerically evaluates kinematics and dynamics terms for multibody systems given a topological description. Manipulators may be free-flying, and may have closed-chain constraints. With the exception of object squeeze-force control, the algorithm does not deal with actuator redundancy. The algorithm is used to implement an automated 2D computed-torque dynamics and control package that allows joint, endpoint, orientation, momentum, and object squeeze-force control. This package obviates the need for hand-derivation of kinematics and dynamics, and is used for both simulation and experimental control. Endpoint control experiments are performed on a laboratory robot that has two arms to manipulate payloads, and uses an air bearing to achieve very-low drag characteristics. Simulations and experimental data for endpoint and object controllers are presented for the experimental robot - a complex dynamic system. There is a certain rather wide set of conditions under which CT endpoint controllers can neglect robot base accelerations (but not motions) and achieve comparable performance including base accelerations in the model. The regime over which this simplification holds is explored by simulation and experiment.

Koningstein, Ross↗
NASA NTRS1977

Robustness of linear quadratic state feedback designs in the presence of system uncertainty

The paper deals with the problem of expressing the robustness (stability) property of a linear quadratic state feedback (LQSF) design quantitatively in terms of bounds on the perturbations (modeling errors or parameter variations) in the system matrices so that the closed-loop system remains stable. Nonlinear time-varying and linear time-invariant perturbations are considered. The only computation required in obtaining a measure of the robustness of an LQSF design is to determine the eigenvalues of two symmetric matrices determined when solving the algebraic Riccati equation corresponding to the LQSF design problem. Results are applied to a complex dynamic system consisting of the flare control of a STOL aircraft. The design of the flare control is formulated as an LQSF tracking problem.

Patel, R. V.↗
DOE OSTI2024

Dynamic Control of Sodium Cold Trap Purification Temperature Using LSTM System Identification

This study investigates the dynamic regulation of the sodium cold trap purification temperature at Argonne National Laboratory’s liquid sodium test facility, employing long short-term memory (LSTM) system identification techniques. The investigation introduces an innovative hybrid approach by integrating model predictive control (MPC) based on first principles dynamic models with a multi-step time–frequency LSTM model in predicting the temperature profiles of a sodium cold trap purification system. The long short-term memory–model predictive controller (LSTM-MPC) model employs a sliding window scheme to gather training samples for multi-step prediction, leveraging historical data to construct predictive models that capture the non-linearities of the complex system dynamics without explicitly modeling the underlying physical processes. The performance of the LSTM-MPC and MPC were evaluated through simulation experiments, where both models were assessed on their capacity to maintain the cold trap temperature within predefined set-points while minimizing deviations and overshoots. Results obtained show how the data-driven LSTM-MPC model demonstrates stability and adaptability. In contrast, the traditional MPC model exhibits irregularities, particularly evident as overshoots around set-point limits, which can potentially compromise its effectiveness over long prediction time intervals. The findings obtained offer valuable insights into integrating data-driven techniques for enhancing real-time monitoring systems.

LSTM-MPC↗
DOE OSTI2016

X-ray diffraction from nonuniformly stretched helical molecules

The fibrous proteins in living cells are exposed to mechanical forces interacting with other subcellular structures. X-ray fiber diffraction is often used to assess deformation and movement of these proteins, but the analysis has been limited to the theory for fibrous molecular systems that exhibit helical symmetry. However, this approach cannot adequately interpret X-ray data from fibrous protein assemblies where the local strain varies along the fiber length owing to interactions of its molecular constituents with their binding partners. To resolve this problem a theoretical formulism has been developed for predicting the diffraction from individual helical molecular structures nonuniformly strained along their lengths. This represents a critical first step towards modeling complex dynamical systems consisting of multiple helical structures using spatially explicit, multi-scale Monte Carlo simulations where predictions are compared with experimental data in a `forward' process to iteratively generate ever more realistic models. Here the effects of nonuniform strains and the helix length on the resulting magnitude and phase of diffraction patterns are quantitatively assessed. Examples of the predicted diffraction patterns of nonuniformly deformed double-stranded DNA and actin filaments in contracting muscle are presented to demonstrate the feasibly of this theoretical approach.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2018

Quantum coherence–driven self-organized criticality and nonequilibrium light localization

Self-organized criticality emerges in dynamical complex systems driven out of equilibrium and characterizes a wide range of classical phenomena in physics, geology, and biology. We report on a quantum coherence–controlled self-organized critical transition observed in the light localization behavior of a coherence-driven nanophotonic configuration. Our system is composed of a gain-enhanced plasmonic heterostructure controlled by a coherent drive, in which photons close to the stopped-light regime interact in the presence of the active nonlinearities, eventually synchronizing their dynamics. In this system, on the basis of analytical and corroborating full-wave Maxwell-Bloch computations, we observe quantum coherence–controlled self-organized criticality in the emergence of light localization arising from the synchronization of the photons. It is associated with two first-order phase transitions: one pertaining to the synchronization of the dynamics of the photons and the second pertaining to an inversionless lasing transition by the coherent drive. The so-attained light localization, which is robust to dissipation, fluctuations, and many-body interactions, exhibits scale-invariant power laws and absence of finely tuned control parameters. We also found that, in this nonequilibrium dynamical system, the effective critical “temperature” of the system drops to zero, whereupon one enters the quantum self-organized critical regime.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Technical note: Using long short-term memory models to fill data gaps in hydrological monitoring networks

Abstract. Quantifying the spatiotemporal dynamics in subsurface hydrological flows over a long time window usually employs a network of monitoring wells. However, such observations are often spatially sparse with potential temporal gaps due to poor quality or instrument failure. In this study, we explore the ability of recurrent neural networks to fill gaps in a spatially distributed time-series dataset. We use a well network that monitors the dynamic and heterogeneous hydrologic exchanges between the Columbia River and its adjacent groundwater aquifer at the U.S. Department of Energy's Hanford site. This 10-year-long dataset contains hourly temperature, specific conductance, and groundwater table elevation measurements from 42 wells with gaps of various lengths. We employ a long short-term memory (LSTM) model to capture the temporal variations in the observed system behaviors needed for gap filling. The performance of the LSTM-based gap-filling method was evaluated against a traditional autoregressive integrated moving average (ARIMA) method in terms of error statistics and accuracy in capturing the temporal patterns of river corridor wells with various dynamics signatures. Our study demonstrates that the ARIMA models yield better average error statistics, although they tend to have larger errors during time windows with abrupt changes or high-frequency (daily and subdaily) variations. The LSTM-based models excel in capturing both high-frequency and low-frequency (monthly and seasonal) dynamics. However, the inclusion of high-frequency fluctuations may also lead to overly dynamic predictions in time windows that lack such fluctuations. The LSTM can take advantage of the spatial information from neighboring wells to improve the gap-filling accuracy, especially for long gaps in system states that vary at subdaily scales. While LSTM models require substantial training data and have limited extrapolation power beyond the conditions represented in the training data, they afford great flexibility to account for the spatial correlations, temporal correlations, and nonlinearity in data without a priori assumptions. Thus, LSTMs provide effective alternatives to fill in data gaps in spatially distributed time-series observations characterized by multiple dominant frequencies of variability, which are essential for advancing our understanding of dynamic complex systems.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Uncertainty Aware Deep Learning for Particle Accelerators

Standard deep learning models for classification and regression applications are ideal for capturing complex system dynamics. However, their predictions can be arbitrarily inaccurate when the input samples are not similar to the training data. Implementation of distance aware uncertainty estimation can be used to detect these scenarios and provide a level of confidence associated with their predictions. In this paper, we present results from using Deep Gaussian Process Approximation (DGPA) methods for errant beam prediction at Spallation Neutron Source (SNS) accelerator (classification) and we provide an uncertainty aware surrogate model for the Fermi National Accelerator Lab (FNAL) Booster Accelerator Complex (regression).

Rajput, Kishansingh↗
DOE OSTI2023

MODELING A SPENT NUCLEAR FUEL CASK SEISMIC TEST

The US Department of Energy Spent Fuel and Waste Science and Technology (SFWST) program is planning to conduct a series of full-scale shake table tests to simulate hypothetical earthquake conditions and record the response of surrogate spent nuclear fuel (SNF) assemblies in a canister dry storage system mockup. The shake table motions will represent a range of hypothetical earthquake conditions at hypothetical locations in the continental US to generally define the range of mechanical loads that SNF can be expected to experience during extended dry storage periods. The earthquake conditions will represent seismic hazards in the 2,000-to-20,000-year return period range. The test will use instrumented pressurized water reactor fuel assemblies with surrogate mass inside the fuel rods instead of radioactive fuel pellets. Dummy assemblies with similar mass and dimensions to actual fuel assemblies will occupy the rest of the fuel assembly locations in a SNF canister that holds 32 fuel assemblies. The canister will be located inside a mockup vertical concrete cask. Instrumentation will record the motion of the major components of the complex dynamic system, and strain gauges will be used to record the cladding strain at select locations. Preparations for the test require modeling predictions to identify the range of response of the system and to help select specific earthquake cases to be simulated on the shake table from a large set of potential cases. This paper describes the pretest nonlinear finite element modeling efforts that have been completed to date, including cask system level modeling and fuel assembly modeling in LS-DYNA. The cask system level models are critical for anticipating sliding or tipping of an unanchored cask during the test. The fuel assembly model is needed to estimate the range of cladding strain response and fuel assembly structural response to be expected during the test and cladding strain measurements will be key metrics for model validation and the development of modeling best practices after the test is concluded.

Klymyshyn, Nicholas A.↗
DOE OSTI2017

Energy prediction using spatiotemporal pattern networks

This paper presents a novel data-driven technique based on the spatiotemporal pattern network (STPN) for energy/power prediction for complex dynamical systems. Built on symbolic dynamical filtering, the STPN framework is used to capture not only the individual system characteristics but also the pair-wise causal dependencies among different sub-systems. To quantify causal dependencies, a mutual information based metric is presented and an energy prediction approach is subsequently proposed based on the STPN framework. To validate the proposed scheme, two case studies are presented, one involving wind turbine power prediction (supply side energy) using the Western Wind Integration data set generated by the National Renewable Energy Laboratory (NREL) for identifying spatiotemporal characteristics, and the other, residential electric energy disaggregation (demand side energy) using the Building America 2010 data set from NREL for exploring temporal features. In the energy disaggregation context, convex programming techniques beyond the STPN framework are developed and applied to achieve improved disaggregation performance.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Quantifying uncertainty for deep learning based forecasting and flow-reconstruction using neural architecture search ensembles

Classical problems in computational physics such as data-driven forecasting and signal reconstruction from sparse sensors have recently seen an explosion in deep neural network (DNN) based algorithmic approaches. However, most DNN models do not provide uncertainty estimates, which are crucial for establishing the trustworthiness of these techniques in downstream decision making tasks and scenarios. In recent years, ensemble-based methods have achieved significant success for the uncertainty quantification in DNNs on a number of benchmark problems. However, their performance on real-world applications remains under-explored. In this work, we present an automated approach to DNN discovery and demonstrate how this may also be utilized for ensemble-based uncertainty quantification. Specifically, we propose the use of a scalable neural and hyperparameter architecture search for discovering an ensemble of DNN models for complex dynamical systems. We highlight how the proposed method not only discovers high-performing neural network ensembles for our tasks, but also quantifies uncertainty seamlessly. This is achieved by using genetic algorithms and Bayesian optimization for sampling the search space of neural network architectures and hyperparameters. Subsequently, a model selection approach is used to identify candidate models for an ensemble set construction. Afterwards, a variance decomposition approach is used to estimate the uncertainty of the predictions from the ensemble. We demonstrate the feasibility of this framework for two tasks — forecasting from historical data and flow reconstruction from sparse sensors for the sea-surface temperature. In conclusion, we demonstrate superior performance from the ensemble in contrast with individual high-performing models and other benchmarks.

Deep ensembles↗
DOE OSTI2013

THE ANATOMY OF AN EXTREME STARBURST WITHIN 1.3 Gyr OF THE BIG BANG REVEALED BY ALMA

We present further analysis of the [C II] 158 {mu}m fine structure line and thermal dust continuum emission from the archetype extreme starburst/active galactic nucleus (AGN) group of galaxies in the early universe, BRI 1202-0725 at z = 4.7, using the Atacama Large Millimeter Array. The group has long been noted for having a closely separated (26 kpc in projection) FIR-hyperluminous quasar host galaxy and an optically obscured submillimeter galaxy (SMG). A short ALMA test observation reveals a rich laboratory for the study of the myriad processes involved in clustered massive galaxy formation in the early universe. Strong [C II] emission from the SMG and the quasar have been reported earlier by Wagg et al. based on these observations. In this paper, we examine in more detail the imaging results from the ALMA observations, including velocity channel images, position-velocity plots, and line moment images. We present detections of [C II] emission from two Ly{alpha}-selected galaxies in the group, demonstrating the relative ease with which ALMA can detect the [C II] emission from lower star formation rate galaxies at high redshift. Imaging of the [C II] emission shows a clear velocity gradient across the SMG, possibly indicating rotation or a more complex dynamical system on a scale {approx}10 kpc. There is evidence in the quasar spectrum and images for a possible outflow toward the southwest, as well as more extended emission (a {sup b}ridge{sup )}, between the quasar and the SMG, although the latter could simply be emission from Ly{alpha}-1 blending with that of the quasar at the limited spatial resolution of the current observations. These results provide an unprecedented view of a major merger of gas-rich galaxies driving extreme starbursts and AGN accretion during the formation of massive galaxies and supermassive black holes within 1.3 Gyr of the big bang.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Adaptive machine learning for time-varying systems: low dimensional latent space tuning

Machine learning (ML) tools such as encoder-decoder convolutional neural networks (CNN) can represent incredibly complex nonlinear functions which map between combinations of images and scalars. For example, CNNs can be used to map combinations of accelerator parameters and images which are 2D projections of the 6D phase space distributions of charged particle beams as they are transported between various particle accelerator locations. Despite their strengths, applying ML to time-varying systems, or systems with shifting distributions, is an open problem, especially for large systems for which collecting new data for re-training is impractical or interrupts operations. Particle accelerators are one example of large time-varying systems for which collecting detailed training data requires lengthy dedicated beam measurements which may no longer be available during regular operations. We present a novel method of adaptive ML for time-varying systems. Our approach is to map very high (N ≈ 100k) dimensional inputs (a combination of scalar parameters and images) into the low dimensional (N ≈ 2) latent space at the output of the encoder section of an encoder-decoder CNN. We then actively tune the low dimensional latent space-based representation of complex system dynamics by the addition of an adaptively tuned feedback vector directly before the decoder sections builds back up to our image-based high-dimensional phase space density representations. This method allows us to learn correlations within and to quickly tune the characteristics of incredibly large parameter space systems and to track their evolution in real time based on feedback without massive new data sets for re-training. We demonstrate that our method can accurately predict and track the phase space of charged particle beams at various locations in a particle accelerator by adaptively adjusting in real-time while the unknown input beam distribution of the accelerator is changing in shape, charge, and offset and while the RF system of the accelerator itself is also changing in an unpredictable way. For FACET-II we demonstrate that such an approach has the potential to use transverse deflecting cavity and energy spread spectrum beam measurements to accurately predict 2D projections of the 6D phase space of the electron beam at the plasma wakefield acceleration interaction point where such diagnostics are unavailable.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Physics-constrained superresolution diffusion for six-dimensional phase space diagnostics

Adaptive physics-constrained superresolution diffusion is developed for noninvasive virtual diagnostics of the six-dimensional (6D) phase space density of charged particle beams. An adaptive variational autoencoder embeds initial beam condition images and scalar measurements to a low-dimensional latent space from which a 32 6 pixel 6D tensor representation of the beam's 6D phase space density is generated. Projecting from a 6D tensor generates physically consistent two-dimensional projections. Physics-guided superresolution diffusion transforms low-resolution images of the 6D density to high resolution 256 × 256 pixel images. Unsupervised adaptive latent space tuning enables tracking of time-varying beams without knowledge of time-varying initial conditions. The method is demonstrated with experimental data and multiparticle simulations at the HiRES UED. The general approach is applicable to a wide range of complex dynamic systems evolving in high-dimensional phase space. The method is shown to be robust to distribution shift without retraining. Published by the American Physical Society 2025

43 PARTICLE ACCELERATORS↗
DOE OSTI2018

NSLS-II: Turn-by-Turn BPM Data Analysis – A Use Case Approach

A modern ring-based light source is a complex dynamic system for which the design is guided by a realistic computer model with predictable results. State-of-the-arts beam requirements translate into tight engineering tolerances. So, quality assurance for the equipment, e.g. ring magnets, girder system, power supplies, RF system, etc. is crucial to deliver on the design goals for the beam quality. For a streamlined commissioning and an effective approach for integration of insertion devices, the latter being an activity that is typically on-going for the entire life cycle of the facility, it is equally important to be able to calibrate the beam dynamics design model against the real system and pursue e.g. “gap analysis”, linear and nonlinear optics corrections, improvements to the baseline design, etc. In particular, for high-end insertion devices. Conceptually, it can be thought of as pursuing Model-Based and Adaptive Control; instead of “flying blind”. A straightforward approach is to use a pinger, a ~30 cm long hor/ver kicker that can kick the beam to about 1 mrad in each plane [1-2] and a Turn-by-Turn BPM data acquisition system that records the betatron motion on a Turn-by-Turn basis. Examples include: ALS, SLS, DIAMOND, and SOLEIL. We first outline how to measure the Beam Impulse Response (BIR) function and related Beam Transfer Function (BTF) with a Turn-by-Turn BPM data acquisition system and a pinger. In particular, how to obtain the amplitude and phase of the spectral components in the betatron spectrum with high precision by using signal processing. We also quantify how the impact of noise affects the accuracy of the measurements.

97 MATHEMATICS AND COMPUTING↗