Physics guided neural networks for modelling of non-linear dynamics
Not Available
97 MATHEMATICS AND COMPUTING↗
SEARCH · Search NASA
Results for “Dynamic Linear Model”
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.
A non-linear dynamic lumped-parameter model of a rectangular plate.
A lumped-parameter model of a rectangular plate is developed by assuming fundamental mode solutions and using Hamilton's Principle and the Euler equations to set up the differential equation of motion for the system. The plate theory used may be described as the dynamic analogue of the von Karman large-deflection theory. Four sets of symmetrical boundary conditions are considered with the restriction of uniform pressure dynamic loads. The model takes the form of a mass on a cubic-hardening spring with each term defined by algebraic expressions of the plate parameters. The results for some specific problems are compared with two previous solutions. This method is less accurate but simpler to develop and apply.
Bayles, D. J.↗
Integration of Linear Dynamic Emission and Climate Models with Air Traffic Simulations
Future air traffic management systems are required to balance the conflicting objectives of maximizing safety and efficiency of traffic flows while minimizing the climate impact of aviation emissions and contrails. Integrating emission and climate models together with air traffic simulations improve the understanding of the complex interaction between the physical climate system, carbon and other greenhouse gas emissions and aviation activity. This paper integrates a national-level air traffic simulation and optimization capability with simple climate models and carbon cycle models, and climate metrics to assess the impact of aviation on climate. The capability can be used to make trade-offs between extra fuel cost and reduction in global surface temperature change. The parameters in the simulation can be used to evaluate the effect of various uncertainties in emission models and contrails and the impact of different decision horizons. Alternatively, the optimization results from the simulation can be used as inputs to other tools that monetize global climate impacts like the FAA s Aviation Environmental Portfolio Management Tool for Impacts.
Sridhar, Banavar↗
The response of an equatorial ocean to simple wind stress patterns. I - Model formulation and analytic results. II - Numerical results
A time-dependent, primitive equation, beta plane model that is two-dimensional in the horizontal has been developed to model wind-driven equatorial ocean circulation. A simple vertical structure consisting of two layers above the thermocline with the same constant density permits a steady-state undercurrent in the model. An analytical study of the linear dynamics of the model suggests that the addition of inertial effects is needed to simulate the undercurrent properly. Also, both linear and nonlinear dynamics of the model are investigated numerically. Such nonlinear response to wind stress as a strong eastward equatorial undercurrent and an intense eastward 'countercurrent' at three deg N are noted in the numerical results.
Cane, M. A.↗
Methods of sequential estimation for determining initial data in numerical weather prediction
Numerical weather prediction (NWP) is an initial-value problem for a system of nonlinear differential equations, in which initial values are known incompletely and inaccurately. Observational data available at the initial time must therefore be supplemented by data available prior to the initial time, a problem known as meteorological data assimilation. A further complication in NWP is that solutions of the governing equations evolve on two different time scales, a fast one and a slow one, whereas fast scale motions in the atmosphere are not reliably observed. This leads to the so called initialization problem: initial values must be constrained to result in a slowly evolving forecast. The theory of estimation of stochastic dynamic systems provides a natural approach to such problems. For linear stochastic dynamic models, the Kalman-Bucy (KB) sequential filter is the optimal data assimilation method, for linear models, the optimal combined data assimilation-initialization method is a modified version of the KB filter.
Cohn, S. E.↗
Stability Assessment and Tuning of an Adaptively Augmented Classical Controller for Launch Vehicle Flight Control
Recently, a robust and practical adaptive control scheme for launch vehicles [ [1] has been introduced. It augments a classical controller with a real-time loop-gain adaptation, and it is therefore called Adaptive Augmentation Control (AAC). The loop-gain will be increased from the nominal design when the tracking error between the (filtered) output and the (filtered) command trajectory is large; whereas it will be decreased when excitation of flex or sloshing modes are detected. There is a need to determine the range and rate of the loop-gain adaptation in order to retain (exponential) stability, which is critical in vehicle operation, and to develop some theoretically based heuristic tuning methods for the adaptive law gain parameters. The classical launch vehicle flight controller design technics are based on gain-scheduling, whereby the launch vehicle dynamics model is linearized at selected operating points along the nominal tracking command trajectory, and Linear Time-Invariant (LTI) controller design techniques are employed to ensure asymptotic stability of the tracking error dynamics, typically by meeting some prescribed Gain Margin (GM) and Phase Margin (PM) specifications. The controller gains at the design points are then scheduled, tuned and sometimes interpolated to achieve good performance and stability robustness under external disturbances (e.g. winds) and structural perturbations (e.g. vehicle modeling errors). While the GM does give a bound for loop-gain variation without losing stability, it is for constant dispersions of the loop-gain because the GM is based on frequency-domain analysis, which is applicable only for LTI systems. The real-time adaptive loop-gain variation of the AAC effectively renders the closed-loop system a time-varying system, for which it is well-known that the LTI system stability criterion is neither necessary nor sufficient when applying to a Linear Time-Varying (LTV) system in a frozen-time fashion. Therefore, a generalized stability metric for time-varying loop=gain perturbations is needed for the AAC.
VanZwieten, Tannen↗
Progress Towards Integrating the Finite-Volume Cubed-Sphere (FV3) Dynamical Core Tangent Linear and Adjoint Models into JEDI
The Joint Effort for Data assimilation Integration (JEDI) -- led by the Joint Center for Satellite Data Assimilation (JCSDA) -- is an inter-organizational endeavor to develop a common framework for performing data assimilation. This extensive framework will ultimately provide solvers, observation operators, interpolation and model interfaces using object oriented modeling. Two partners involved in JEDI use or plan to use the Finite Volume Cubed-Sphere (FV3) dynamical core to produce weather forecasts; these are NASA's Global Modeling and Assimilation Office and NOAA's National Center for Environment Prediction. In this work we present an update on ongoing efforts to integrate the FV3 tangent linear and adjoint models into the prototype JEDI framework. We setup and run a simple cycled data assimilation experiment using 4DVAR on the cubed sphere grid and with the FV3 tangent linear and adjoint models. Development of the observation operators for JEDI is separately underway. Instead of using real observations a simplified set of simulated observations will be used. We discuss the steps required to bring the FV3 linearized model into the object oriented framework and consider what would be the computational requirements of running this configuration for an operational system. FV3 uses a small time-step to ensure that small scales are well resolved, however this presents design challenges when running 4DVAR with the adjoint. An approach to storing the FV3 model trajectory has been developed that maintains the flexibility of using automatic differentiation. We discuss how this approach is incorporated into the framework. Other important uses of adjoint models include computing observation impacts and singular vectors, we consider how these tools can be included in JEDI.
Holdaway, Daniel↗
Unsteady aerodynamics and flutter of propfans using a three-dimensional Full-Potential Solver
A full-potential solver coupled with a linear structural dynamics model is used to calculate the unsteady aerodynamics and aeroelasticity of propfans. The solver allows calculations for arbitrary interblade phase angles. Results are presented for two propfan configurations. Good agreement is seen between the full-potential results and results from linear theory since the flow is subsonic and the thickness of the propfan blades is small. Some difficulty is encountered due to wave reflections from outer computational boundaries; however, this does not affect the results in the range of frequencies of aeroelastic interest.
Bakhle, Milind A.↗
On Kinetics Modeling of Vibrational Energy Transfer
Two models of vibrational energy exchange are compared at equilibrium to the elementary vibrational exchange reaction for a binary mixture. The first model, non-linear in the species vibrational energies, was derived by Schwartz, Slawsky, and Herzfeld (SSH) by considering the detailed kinetics of vibrational energy levels. This model recovers the result demanded at equilibrium by the elementary reaction. The second model is more recent, and is gaining use in certain areas of computational fluid dynamics. This model, linear in the species vibrational energies, is shown not to recover the required equilibrium result. Further, this more recent model is inconsistent with its suggested rate constants in that those rate constants were inferred from measurements by using the SSH model to reduce the data. The non-linear versus linear nature of these two models can lead to significant differences in vibrational energy coupling. Use of the contemporary model may lead to significant misconceptions, especially when integrated in computer codes considering multiple energy coupling mechanisms.
Gilmore, John O.↗
Unsteady aerodynamics and flutter based on the potential equation
A time-domain three-dimensional full-potential solver is coupled with a linear structural dynamics model to investigate the unsteady aerodynamics and aeroelasticity of propfans. The solver allows calculations in multiple blade passages with independent blade motions. Aeroelastic calculations are performed in both frequency and time domains. Results are presented for two propfan configurations. Good agreement is seen between the full-potential results and results from linear theory since the flow is subsonic and the thickness of the propfan blades is small; the agreement is not as good for the cases in which the angle of attack is high. Some difficulty is encountered due to wave reflections from outer computational boundaries; however, this does not affect the results in the range of frequencies of interest.
Bakhle, Milind A.↗
Design of a helicopter autopilot by means of linearizing transformations
An automatic flight control systems design methods for aircraft that have complex characteristics and operational requirements, such as the powered lift STOL and V/STOL configurations are discussed. The method is effective for a large class of dynamic systems that require multiaxis control and that have highly coupled nonlinearities, redundant controls, and complex multidimensional operational envelopes. The method exploits the possibility of linearizing the system over its operational envelope by transforming the state and control. The linear canonical forms used in the design are described, and necessary and sufficient conditions for linearizability are stated. The control logic has the structure of an exact model follower with linear decoupled model dynamics and possibly nonlinear plant dynamics. The design method is illustrated with an application to a helicopter autopilot design.
Meyer, G.↗
A Textbook for a First Course in Computational Fluid Dynamics
This paper describes and discusses the textbook, Fundamentals of Computational Fluid Dynamics by Lomax, Pulliam, and Zingg, which is intended for a graduate level first course in computational fluid dynamics. This textbook emphasizes fundamental concepts in developing, analyzing, and understanding numerical methods for the partial differential equations governing the physics of fluid flow. Its underlying philosophy is that the theory of linear algebra and the attendant eigenanalysis of linear systems provides a mathematical framework to describe and unify most numerical methods in common use in the field of fluid dynamics. Two linear model equations, the linear convection and diffusion equations, are used to illustrate concepts throughout. Emphasis is on the semi-discrete approach, in which the governing partial differential equations (PDE's) are reduced to systems of ordinary differential equations (ODE's) through a discretization of the spatial derivatives. The ordinary differential equations are then reduced to ordinary difference equations (O(Delta)E's) using a time-marching method. This methodology, using the progression from PDE through ODE's to O(Delta)E's, together with the use of the eigensystems of tridiagonal matrices and the theory of O(Delta)E's, gives the book its distinctiveness and provides a sound basis for a deep understanding of fundamental concepts in computational fluid dynamics.
Zingg, D. W.↗
Dynamics and Control of Flexible Space Vehicles
The purpose of this report is twofold: (1) to survey the established analytic procedures for the simulation of controlled flexible space vehicles, and (2) to develop in detail methods that employ a combination of discrete and distributed ("modal") coordinates, i.e., the hybrid-coordinate methods. Analytic procedures are described in three categories: (1) discrete-coordinate methods, (2) hybrid-coordinate methods, and (3) vehicle normal-coordinate methods. Each of these approaches is described and analyzed for its advantages and disadvantages, and each is found to have an area of applicability. The hybrid-coordinate method combines the efficiency of the vehicle normal-coordinate method with the versatility of the discrete-coordinate method, and appears to have the widest range of practical application. The results in this report have practical utility in two areas: (1) complex digital computer simulation of flexible space vehicles of arbitrary configuration subject to realistic control laws, and (2) preliminary control system design based on transfer functions for linearized models of dynamics and control laws.
Likins, P. W.↗
Frequency domain approach for evaluation of stochastic control of elastic spacecraft
An attempt is made to develop a reduced-order filter for an elastic spacecraft. The frequency-domain approach is resorted to both for arriving at an appropriate ROF and for evaluating the resulting stochastic flight controller. The frequency-domain approach employed is contrasted to the time-domain approach previously utilized by the authors. A linearized deterministic dynamical model of a solar electric propulsion spacecraft is analyzed. Flexibility modes were incorporated profitably into the ROF model, and the stochastic controllers exhibit notching of structural resonances reminiscent of classical notch filters. Accurate estimates of modal parameters (damping, natural frequency) are deemed crucial, and use of worst-case values in the ROF design is recommended.
Larson, V.↗
Simplified Linear Multivariable Control Of Robots
Simplified method developed to design control system that makes joints of robot follow reference trajectories. Generic design includes independent multivariable feedforward and feedback controllers. Feedforward controller based on inverse of linearized model of dynamics of robot and implements control law that contains only proportional and first and second derivatives of reference trajectories with respect to time. Feedback controller, which implements control law of proportional, first-derivative, and integral terms, makes tracking errors converge toward zero as time passes.
Seraji, Homayoun↗
Benefits of variable rotor speed in integrated helicopter/engine control
Current helicopter flight and propulsion controls are typically designed with the assumption that rotor speed will be held to a constant setpoint. A new flight and propulsion control system using a continuously variable rotor speed command is proposed to improve the maneuverability and agility of helicopter systems. In this new approach, the flight control system generates an optimal variable rotor speed command in addition to conventional control commands in a framework of integrated flight/propulsion control. The benefits (i.e. improved maneuverability and agility) of varying rotor speed during transient maneuvers are demonstrated using a bob-up maneuver as an example. In particular, two types of benefits are identified in different maneuver conditions. One comes from a thrust augmentation, while the other comes from an exchange of rotational and translational energy. In the example, a simple linear dynamic hover model is used with an optimal control design method to generate the optimal rotor speed command.
Iwata, Takanori↗
Differential adaptation of the linear and nonlinear components of the horizontal vestibuloocular reflex in squirrel monkeys
Previous work in squirrel monkeys has demonstrated the presence of linear and nonlinear components to the horizontal vestibuloocular reflex (VOR) evoked by high-acceleration rotations. The nonlinear component is seen as a rise in gain with increasing velocity of rotation at frequencies more than 2 Hz (a velocity-dependent gain enhancement). We have shown that there are greater changes in the nonlinear than linear component of the response after spectacle-induced adaptation. The present study was conducted to determine if the two components of the response share a common adaptive process. The gain of the VOR, in the dark, to sinusoidal stimuli at 4 Hz (peak velocities: 20-150 degrees /s) and 10 Hz (peak velocities: 20 and 100 degrees /s) was measured pre- and postadaptation. Adaptation was induced over 4 h with x0.45 minimizing spectacles. Sum-of-sines stimuli were used to induce adaptation, and the parameters of the stimuli were adjusted to invoke only the linear or both linear and nonlinear components of the response. Preadaptation, there was a velocity-dependent gain enhancement at 4 and 10 Hz. In postadaptation with the paradigms that only recruited the linear component, there was a decrease in gain and a persistent velocity-dependent gain enhancement (indicating adaptation of only the linear component). After adaptation with the paradigm designed to recruit both the linear and nonlinear components, there was a decrease in gain and no velocity-dependent gain enhancement (indicating adaptation of both components). There were comparable changes in the response to steps of acceleration. We interpret these results to indicate that separate processes drive the adaptation of the linear and nonlinear components of the response.
Non-NASA Center↗
Numerical Study of Outlet Boundary Conditions for Unsteady Turbulent Internal Flows Using the NCC
This paper presents the results of studies on the outlet boundary conditions for turbulent internal flow simulations. Several outlet boundary conditions have been investigated by applying the National Combustion Code (NCC) to the configuration of a LM6000 single injector flame tube. First of all, very large eddy simulations (VLES) have been performed using the partially resolved numerical simulation (PRNS) approach, in which both the nonlinear and linear dynamic subscale models were employed. Secondly, unsteady Reynolds averaged Navier- Stokes (URANS) simulations have also been performed for the same configuration to investigate the effects of different outlet boundary conditions in the context of URANS. Thirdly, the possible role of the initial condition is inspected by using three different initial flow fields for both the PRNS/VLES simulation and the URANS simulation. The same grid is used for all the simulations and the number of mesh element is about 0.5 million. The main purpose of this study is to examine the long-time behavior of the solution as determined by the imposed outlet boundary conditions. For a particular simulation to be considered as successful under the given initial and boundary conditions, the solution must be sustainable in a physically meaningful manner over a sufficiently long period of time. The commonly used outlet boundary condition for steady Reynolds averaged Navier-Stokes (RANS) simulation is a fixed pressure at the outlet with all the other dependent variables being extrapolated from the interior. The results of the present study suggest that this is also workable for the URANS simulation of the LM6000 injector flame tube. However, it does not work for the PRNS/VLES simulation due to the unphysical reflections of the pressure disturbances at the outlet boundary. This undesirable situation can be practically alleviated by applying a simple unsteady convection equation for the pressure disturbances at the outlet boundary. The numerical results presented in this paper suggest that this unsteady convection of pressure disturbances at the outlet works very well for all the unsteady simulations (both PRNS/VLES and URANS) of the LM6000 single injector flame tube.
Liu, Nan-Suey↗