Search NASA⌕ Search

SEARCH · Search NASA

Results for “linear 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 595 records · Page 33

Reinforcement Learning of Structured Stabilizing Control for Linear Systems With Unknown State Matrix

This paper delves into designing feedback control gains for a continuous-time linear quadratic regulator (LQR) problem that is constrained to certain predefined structure with unknown state matrix. We bring forth the ideas from reinforcement learning (RL) in conjunction with sufficient stability and performance guarantees in order to design these structured gains using the trajectory measurements of states and controls. Here we first formulate a model-based framework using dynamic programming (DP) to embed the structural constraint to the LQR gain computation in the continuous-time setting, and then subsequently, formulate a policy iteration RL algorithm that can alleviate the requirement of known state matrix in conjunction with maintaining the feedback gain structure. The design enables a distributed learning control design which is necessary for many large-scale cyber-physical systems. Theoretical guarantees are provided for stability and convergence of the structured reinforcement learning (SRL) algorithm. We validate our theoretical results with numerical simulations on a multi-agent networked linear time-invariant (LTI) dynamic system.

42 ENGINEERING↗

A state-of-the-art review of transportation systems evaluation techniques relevant to air transportation, volume 1

Mathematical and philosophical approaches are presented for evaluation and implementation of ground and air transportation systems. Basic decision processes are examined that are used for cost analyses and planning (i.e, statistical decision theory, linear and dynamic programming, optimization, game theory). The effects on the environment and the community that a transportation system may have are discussed and modelled. Algorithmic structures are examined and selected bibliographic annotations are included. Transportation dynamic models were developed. Citizen participation in transportation projects (i.e, in Maryland and Massachusetts) is discussed. The relevance of the modelling and evaluation approaches to air transportation (i.e, airport planning) is examined in a case study in St. Louis, Missouri.

Haefner, L. E.↗

Elastic wave propagation in variable media using a discontinuous Galerkin method.

Motivated by the needs of seismic inversion and building on our prior experience for fluid-dynamics systems, we present a high-order discontinuous Galerkin (DG) Runge-Kutta method applied to isotropic, linearized elasto-dynamics. Unlike other DG methods recently presented in the literature, our method allows for inhomogeneous material variations within each element that enables representation of realistic earth models - a feature critical for future use in seismic inversion. Likewise, our method supports curved elements and hybrid meshes that include both simplicial and nonsimplicial elements. We demonstrate the capabilities of this method through a series of numerical experiments including hybrid mesh discretizations of the Marmousi2 model as well as a modified Marmousi2 model with a oscillatory ocean bottom that is exactly captured by our discretization. A discontinuous Galerkin method for solving the equations of linear isotropic elasticity has been presented. The formulation is designed to accommodate variation of media parameters within elements, curved elements and unstructured heterogeneous meshes. We have demonstrated that each of these important features of the formulation can produce results that are significantly different from formulations that do not possess these capabilities suggesting that each of these capabilities may be important for effective full waveform inversion of elastic medium.

36 MATERIALS SCIENCE↗

Symplectic neural network and its application to charged particle dynamics in electromagnetic fields

Recently, machine learning models have shown many successes in various applications in science and technology. In this work, we focus on the charged particle dynamics, with the development of a class of symplectic neural networks, including a linear version, SympMat, and a nonlinear version, HénonNet. Both are designed to preserve the structure of Hamiltonian systems. We show that they can be used to model relevant Hamiltonian systems of interest in plasma physics and astrophysics, for linear and nonlinear charged particle dynamics, with the potential to bridge multi-scale simulations. These symplectic neural networks are adapted to the applications in plasma simulations and particle-wave interaction with parametric dependence and periodicity, where we have investigated their performance and accuracy. In particular, SympMat is shown to outperform the traditional Boris particle pusher down to the sub-gyroperiod scale in the case of charged particles in uniform magnetic fields. HénonNet successfully predicts the hot electron distribution, which is validated against theoretical results. These results highlight the potential of symplectic neural networks as a trajectory integrator for particle-in-cell simulations or a fast surrogate to replace conventional numerical schemes.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Time-delay control of a magnetic levitated linear positioning system

In this paper, a high accuracy linear positioning system with a linear force actuator and magnetic levitation is proposed. By locating a permanently magnetized rod inside a current-carrying solenoid, the axial force is achieved by the boundary effect of magnet poles and utilized to power the linear motion, while the force for levitation is governed by Ampere's Law supplied with the same solenoid. With the levitation in a radial direction, there is hardly any friction between the rod and the solenoid. The high speed motion can hence be achieved. Besides, the axial force acting on the rod is a smooth function of rod position, so the system can provide nanometer resolution linear positioning to the molecule size. Since the force-position relation is highly nonlinear, and the mathematical model is derived according to some assumptions, such as the equivalent solenoid of the permanently magnetized rod, so there exists unknown dynamics in practical application. Thus 'robustness' is an important issue in controller design. Meanwhile the load effect reacts directly on the servo system without transmission elements, so the capability of 'disturbance rejection; is also required. With the above consideration, a time-delay control scheme is chosen and applied. By comparing the input-output relation and the mathematical model, the time-delay controller calculates an estimation of unmodeled dynamics and disturbances and then composes the desired compensation into the system. Effectiveness of the linear positioning system and control scheme are illustrated with simulation results.

Tarn, J. H.↗

The locking and unlocking thresholds for tearing modes in a cylindrical tokamak

The locking and unlocking thresholds for tearing modes are in general different. In this work, the physics origin for this difference is illustrated from theory analysis, and a numerical procedure is developed to find both locking and unlocking thresholds. In particular, a new scaling law for the unlocking threshold that is valid in both weak and strong rotation regimes has been derived from the lowest amplitude of the RMP (resonant magnetic perturbation) allowed for the locked-mode solution. Above the unlocking threshold, the criterion for the phase-flip instability is extended to identify the entire locked-mode states. Two different regimes of the RMP amplitude in terms of the accessibility of the locked-mode states have been found. In the first regime, the locked-mode state may or may not be accessible depending on the initial conditions of an evolving island. In the second regime, the locked-mode state can always be reached regardless of the initial conditions of the tearing mode. The lowest RMP amplitude for the second regime is determined to be the mode-locking threshold. The different characteristics of the two regimes above the unlocking threshold reveal the underlying physics for the gap between the locking and unlocking thresholds and provide an explanation for the closely related and widely observed hysteresis phenomena in island evolution during the sweeping process of the RMP amplitude up and down across that threshold gap.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

ANALYTICAL MODELS OF EXOPLANETARY ATMOSPHERES. I. ATMOSPHERIC DYNAMICS VIA THE SHALLOW WATER SYSTEM

Within the context of exoplanetary atmospheres, we present a comprehensive linear analysis of forced, damped, magnetized shallow water systems, exploring the effects of dimensionality, geometry (Cartesian, pseudo-spherical, and spherical), rotation, magnetic tension, and hydrodynamic and magnetic sources of friction. Across a broad range of conditions, we find that the key governing equation for atmospheres and quantum harmonic oscillators are identical, even when forcing (stellar irradiation), sources of friction (molecular viscosity, Rayleigh drag, and magnetic drag), and magnetic tension are included. The global atmospheric structure is largely controlled by a single key parameter that involves the Rossby and Prandtl numbers. This near-universality breaks down when either molecular viscosity or magnetic drag acts non-uniformly across latitude or a poloidal magnetic field is present, suggesting that these effects will introduce qualitative changes to the familiar chevron-shaped feature witnessed in simulations of atmospheric circulation. We also find that hydrodynamic and magnetic sources of friction have dissimilar phase signatures and affect the flow in fundamentally different ways, implying that using Rayleigh drag to mimic magnetic drag is inaccurate. We exhaustively lay down the theoretical formalism (dispersion relations, governing equations, and time-dependent wave solutions) for a broad suite of models. In all situations, we derive the steady state of an atmosphere, which is relevant to interpreting infrared phase and eclipse maps of exoplanetary atmospheres. We elucidate a pinching effect that confines the atmospheric structure to be near the equator. Our suite of analytical models may be used to develop decisively physical intuition and as a reference point for three-dimensional magnetohydrodynamic simulations of atmospheric circulation.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Scalable Extended Dynamic Mode Decomposition Using Random Kernel Approximation

The Koopman operator is a linear, infinite-dimensional operator that governs the dynamics of system observables; extended dynamic mode decomposition (EDMD) is a data-driven method for approximating the Koopman operator using functions (features) of the system state snapshots. This paper investigates an approach to EDMD in which the features used provide random approximations to a particular kernel function. The objective of this is computational economy for large data sets: EDMD is generally ill-suited for problems with large state dimension, and its dual kernel formulation (KDMD) is well-suited for such problems only if the number of data snapshots is relatively small. We discuss two specific methods for generating features: random Fourier features and the Nÿstrom method. The first method is a data-independent method for translation-invariant kernels only and involves random sampling in feature space; the second method is a data-dependent empirical method that may be used for any kernel and involves random sampling of data. We first discuss how these ideas may be applied in an EDMD context, as well as a means for adaptively adding random Fourier features. As a result, we demonstrate these methods on two example problems and conclude with an analysis of the relative benefits and drawbacks of each method.

59 BASIC BIOLOGICAL SCIENCES↗

Strengthening the Security of Operational Technology: Understanding Contemporary Bill of Materials

The evolution of cyber-physical infrastructure has made its security more challenging. The last few years have witnessed a convergence of hardware and software segments in various domains, including operational technology (OT) which is responsible for carrying out critical tasks such as monitoring and controlling power grids, nuclear plants, transportation, and emergency services. Both hardware and software encapsulate numerous open source and proprietary subcomponents, making it crucial for end-users to understand the composition of the products they are using. For example, wind turbines incorporate thousands of lines of code (software) used for the turbine's design, planning, operation, and analytics in addition to the numerous hardware subcomponents that construct it. Due to the highly complex nature of software and hardware, knowledge of the components and subcomponents is required to mitigate cyber vulnerabilities and defend against cyberattacks. There has also been a transformation from a traditional linear supply chain into a global, dynamic, diverse, and interconnected system. The digitization of the supply chain makes it easier to find and exploit vulnerabilities. Critical infrastructures (e.g., power grids, oil, natural gas, water, and wastewater) rely on OT to function, and if the OT is compromised, equipment damage and potential interruption of services could result. A significant security measure to protect OT systems from disruption is to develop a supply chain bill of materials (BoM) corresponding to the software and hardware used in OT, along with attestations amongst vendors and asset owners. A supply chain BoM is a proactive way to understand the inherent vulnerabilities in the system and mitigate them in advance of being exploited. BoMs bolster the trust placed in the digital infrastructure and enhance software supply chain security by sustaining the management of component obsolescence and compliance, along with the seclusion of unsafe segments of a specific product. Adopting BoM tools is becoming increasingly important across various government sectors, as evidenced by the recent U.S. executive order on cybersecurity (NIST 2021). This paper aims to classify BoMs based on structure, functionality, component type, and architecture. The work also discusses case studies to further highlight the benefits of BoMs. In addition, it identifies missing pieces in existing BoM implementations so that future research may identify bounds on where it could expect to make improvements and directly enable researchers to identify promising areas for exploration. Further, the authors provide valuable recommendations to tool developers, researchers, and standardizing organizations (policymakers), additionally benefitting critical infrastructure owners and government executives. This aids in paving a path for future work, thereby, providing suggestions to determine a tool for consumers that best suit their needs.

97 MATHEMATICS AND COMPUTING↗

Dynamic Mode Decomposition of Solids

Dynamic mode decomposition (DMD) is a method that has gained notoriety in the field of turbulent fluid flows as a method for decomposing the flow field into modes that could be further deconstructed to understand their influence on the overall dynamics of the system. Forays into solids and non-linear systems were considered, but not applied. In this work, DMD was applied for the first time to the heat-diffusion and reactive heat diffusion equations on a random particle pack of uniform solid spheres. A verification of a linear heat-diffusion test problem was successful, showing equality between the normal modes and Koopman modes obtained from DMD. Further application to a non-linear reactive system revealed stability limits of the underlying modes which are dependent on microstructure and chemical kinetics. This work will enable the development of reactive material models based on further analysis using DMD to quantify the statistical dependencies of transient response on microstructural characteristics.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

On absolute stability and sensitivity.

Dynamic system absolute stability and sensitivity to parameter variations in linear part of system calculated by graphical and analytical techniques

Siljak, D.↗

Development of a rotorcraft. Propulsion dynamics interface analysis, volume 2

A study was conducted to establish a coupled rotor/propulsion analysis that would be applicable to a wide range of rotorcraft systems. The effort included the following tasks: (1) development of a model structure suitable for simulating a wide range of rotorcraft configurations; (2) defined a methodology for parameterizing the model structure to represent a particular rotorcraft; (3) constructing a nonlinear coupled rotor/propulsion model as a test case to use in analyzing coupled system dynamics; and (4) an attempt to develop a mostly linear coupled model derived from the complete nonlinear simulations. Documentation of the computer models developed is presented.

Hull, R.↗

Analytical comparison of transient and steady state visual evoked cortical potentials

To better describe the linear-dynamic properties of the human visual-cortical response system, transient and steady state Visual Evoked Response Potentials (VERP) were observed. The stimulus presentation device provided both the evoking stimulus (flickering or pulsing lights) and a video task display. The steady state stimulus was modulated by a complex, ten frequency, sum-of-sines, wave. The transient VERP was the time-locked average of the EEG to a series of narrow light pulses (pulse width of 10 msec). The Fourier transform of the averaged pulses had properties that approximate band limited white noise, i.e., a flat spectrum over the frequency region spanned by the 10 summed sines. The Fourier transform of both the steady state and the transient evoked potentials resulted in transfer that are equivalent and therefore comparable. To investigate the effects of task loading on evoked potentials, a grammatical reasoning task was provided. Results support the relevancy of continued application of a systems engineering approach for describing neurosensory functioning.

Junker, A. M.↗

An approach to CMG steering using feedback linearization

This paper presents an approach for controlling spacecraft equipped with control moment gyroscopes. A technique from feedback linearization theory is used to transform the original nonlinear problem to an equivalent linear form without approximating assumptions. In this form, the spacecraft dynamics appear linearly, and are decoupled from redundancy in the system of gyroscopes. A general approach to distributing control effort among the available actuators is described which includes provisions for redistribution of rotors, explicit bounds in gimbal rates, and guaranteed operation at or near singular configurations. A particular algorithm is developed for systems of double-gimbal devices, and demonstrated in two examples for which existing approaches fail to give adequate performance.

Dzielski, John↗

Approach to control moment gyroscope steering using feedback linearization

This paper presents an approach for controlling spacecraft equipped with control moment gyroscopes. A technique from feedback linearization theory is used to transform the original nonlinear problem to an equivalent linear form without approximating assumptions. In this form, the spacecraft dynamics appear linearly, and are decoupled from redundancy in the system of gyroscopes. A general approach to distributing control effort among the available actuators is described which includes provisions for redistribution of rotors, explicit bounds in gimbal rates, and guaranteed operation at or near singular configurations. A particular algorithm is developed for systems of double-gimbal devices, and demonstrated in two examples for which existing approaches fail to give adequate performance.

Dzielski, John↗