Search NASA⌕ Search

SEARCH · Search NASA

Results for “stochastic approximation”

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 379 records · Page 21

Comprehensive modeling of turbulent particulate flows using Eulerian and Lagrangian schemes

This paper addresses turbulent particle dispersion and modulation effects in dilute gas-particle turbulent flows using the Eulerian and Lagrangian modeling approaches. Gradient diffusion approximations are employed in the Eulerian formulation, while a stochastic procedure is utilized to simulate turbulent dispersion in the Lagrangian formulation. The k-epsilon turbulence model is used to characterize the time and length scales of the continuous phase turbulence. For the particle size and loading considered, the turbulence transport equations must be modified to account for the modulation effects. Models are proposed for both Eulerian and Lagrangian schemes. Comparisons and predictions are made in fully developed gas-solid pipe flow and confined coaxial jets laden with particles. For the monodispersed system investigated, Eulerian approach is less expensive and gives more consistent results than the Lagrangian approach. The Lagrangian technique should be further developed to eliminate current inherent inconsistencies especially with regard to the symmetry boundary condition.

Adeniji-Fashola, A.↗

Modeling of confined turbulent fluid-particle flows using Eulerian and Lagrangian schemes

Two important aspects of fluid-particulate interaction in dilute gas-particle turbulent flows (the turbulent particle dispersion and the turbulence modulation effects) are addressed, using the Eulerian and Lagrangian modeling approaches to describe the particulate phase. Gradient-diffusion approximations are employed in the Eulerian formulation, while a stochastic procedure is utilized to simulate turbulent dispersion in the Lagrangina formulation. The k-epsilon turbulence model is used to characterize the time and length scales of the continuous phase turbulence. Models proposed for both schemes are used to predict turbulent fully-developed gas-solid vertical pipe flow with reasonable accuracy.

Adeniji-Fashola, A.↗

Restoring Flow Regimes after Dam Construction and Operation: Evidence from Dynamic Systems Theory

Dams provide a variety of benefits to society, including hydropower generation, flood control, recreation, and water supply for municipal, industrial, and agricultural customers. However, constructing and operating dams for these purposes can negatively impact downstream ecosystems by altering natural flow regimes and interfering with the lifecycles of aquatic and riparian species. Increased concern from environmental interests has put pressure on dam operators to quantify the flow regime changes that have resulted from dam construction and to evaluate how future modifications to dam releases may impact river systems. In this study, we present a method for analyzing the impact of dam operations by reconstructing system trajectories from hydrologic flow time series for different periods of dam operation (pre-dam, post-dam, and post-modification). We apply these methods to flow data from eight rivers in the US to determine the dimensionality of each operating period and to analyze trends based on the primary purpose of dam operation. Hydropower dams operated without flow restrictions exhibit high dimensional, stochastic behavior. However, the system dimensionality is reduced to approximately three variables when ramping restrictions are implemented, or in cases where an afterbay is present to reregulate flows. This low dimensional behavior is more similar to natural flow regimes. For dams operated primarily for flood control or water supply, the flow regimes are typically low dimensional and more deterministic.

13 HYDRO ENERGY↗

Computational Math Problems for a Clean Energy Future

Cutting edge computational mathematics are ubiquitous in renewable energy research. Problems in resilient and reliable electric grid operations, infrastructure planning, wind farm yaw control, and more demand sophisticated and scalable computational tools that enable the transition of renewable energy technologies from proof of concept to deployment into our energy system. The mission of the Computational Science Center at NREL is to lead the lab's efforts to solve energy challenges using high-performance computing (HPC), computational science, applied mathematics, scientific data management, visualization, and informatics. In this poster, we provide a short overview of three areas of computational mathematics research at NREL: wind power scenario generation for stochastic grid operations and infrastructure planning, improved rational function approximations for electromagnetic transients codes, and wind farm yaw control using a combination of the Alternating Direction Method of Multipliers (ADMM) and reinforcement learning (RL). Increasing penetrations of renewable energy into power grids motivate the investigation of new approaches to characterizing uncertainty for five-minute economic dispatch problems. Similarly, as the penetration of distributed energy resources on power grids increases, it becomes important to revisit our methods of modelling transient phenomena, i.e. electromagnetic transients programs. Finally, the combination of ADMM and RL for wind farm yaw control presented here can potentially increase the efficiency of the deployed distributed controllers by orders of magnitude.

ADMM↗

Continuation of probability density functions using a generalized Lyapunov approach

Techniques from numerical bifurcation theory are very useful to study transitions between steady fluid flow patterns and the instabilities involved. Here, we provide computational methodology to use parameter continuation in determining probability density functions of systems of stochastic partial differential equations near fixed points, under a small noise approximation. Key innovation is the efficient solution of a generalized Lyapunov equation using an iterative method involving low-rank approximations. We apply and illustrate the capabilities of the method using a problem in physical oceanography, i.e. the occurrence of multiple steady states of the Atlantic Ocean circulation.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Toward a systematic improvement of the fixed-node approximation in diffusion Monte Carlo for solids—A case study in diamond

While Diffusion Monte Carlo (DMC) is in principle an exact stochastic method for ab initio electronic structure calculations, in practice, the fermionic sign problem necessitates the use of the fixed-node approximation and trial wavefunctions with approximate nodes (or zeros). This approximation introduces a variational error in the energy that potentially can be tested and systematically improved. Here, we present a computational method that produces trial wavefunctions with systematically improvable nodes for DMC calculations of periodic solids. These trial wavefunctions are efficiently generated with the configuration interaction using a perturbative selection made iteratively (CIPSI) method. A simple protocol in which both exact and approximate results for finite supercells are used to extrapolate to the thermodynamic limit is introduced. This approach is illustrated in the case of the carbon diamond using Slater–Jastrow trial wavefunctions including up to one million Slater determinants. Fixed-node DMC energies obtained with such large expansions are much improved, and the fixed-node error is found to decrease monotonically and smoothly as a function of the number of determinants in the trial wavefunction, a property opening the way to a better control of this error. The cohesive energy extrapolated to the thermodynamic limit is in close agreement with the estimated experimental value. Interestingly, this is also the case at the single-determinant level, thus, indicating a very good error cancellation in carbon diamond between the bulk and atomic total fixed-node energies when using single-determinant nodes.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Phase space representation of quantum dynamics

We discuss a phase space representation of quantum dynamics of systems with many degrees of freedom. This representation is based on a perturbative expansion in quantum fluctuations around one of the classical limits. We explicitly analyze expansions around three such limits: (i) corpuscular or Newtonian limit in the coordinate-momentum representation, (ii) wave or Gross-Pitaevskii limit for interacting bosons in the coherent state representation, and (iii) Bloch limit for the spin systems. We discuss both the semiclassical (truncated Wigner) approximation and further quantum corrections appearing in the form of either stochastic quantum jumps along the classical trajectories or the nonlinear response to such jumps. We also discuss how quantum jumps naturally emerge in the analysis of non-equal time correlation functions. This representation of quantum dynamics is closely related to the phase space methods based on the Wigner-Weyl quantization and to the Keldysh technique. We show how such concepts as the Wigner function, Weyl symbol, Moyal product, Bopp operators, and others automatically emerge from the Feynmann's path integral representation of the evolution in the Heisenberg representation. We illustrate the applicability of this expansion with various examples mostly in the context of cold atom systems including sine-Gordon model, one- and two-dimensional Bose-Hubbard model, Dicke model and others.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Poisson-Vlasov in a strong magnetic field: A stochastic solution approach

Stochastic solutions are obtained for the Maxwell-Vlasov equation in the approximation where magnetic field fluctuations are neglected and the electrostatic potential is used to compute the electric field. This is a reasonable approximation for plasmas in a strong external magnetic field. Both Fourier and configuration space solutions are constructed.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A mixed $\ell_1$ regularization approach for sparse simultaneous approximation of parameterized PDEs

We introduce and assess a novel sparse polynomial technique for the simultaneous approximation of parameterized partial differential equations (PDEs) with deterministic and stochastic inputs. Our approach treats the numerical solution as a jointly sparse reconstruction problem through the reformulation of the standard basis pursuit denoising, where the set of jointly sparse vectors is infinite. To achieve global reconstruction of sparse solutions to parameterized elliptic PDEs over both physical and parametric domains, we combine the standard measurement scheme developed for compressed sensing in the context of bounded orthonormal systems with a novel mixed-norm based $\ell_1$ regularization method that exploits both energy and sparsity. Moreover, we are able to prove that, with minimal sample complexity, error estimates comparable to the best $s$-term and quasi-optimal approximations are achievable, while requiring only {\em a priori} bounds on polynomial truncation error with respect to the energy norm.Finally, we perform extensive numerical experiments on several high-dimensional parameterized elliptic PDE models to demonstrate the superior recovery properties of the proposed approach.

97 MATHEMATICS AND COMPUTING↗

Observability Analysis of a Power System Stochastic Dynamic Model Using a Derivative-Free Approach

Serving as a prerequisite to power system dynamic state estimation, the observability analysis of a power system dynamic model has recently attracted the attention of many power engineers. However, because this model is typically nonlinear and large-scale, the analysis of its observability is a challenge to the traditional derivative-based methods. Indeed, the linear-approximation-based approach may provide unreliable results while the nonlinear-technique-based approach inevitably faces extremely complicated derivations. Furthermore, because power systems are intrinsically stochastic, the traditional deterministic approaches may lead to inaccurate observability analyses. In this work, facing these challenges, we propose a novel polynomial-chaos-based derivative-free observability analysis approach that not only is free of any linear approximations, but also accounts for the stochasticity of the dynamic model while bringing a low implementation complexity. Furthermore, this approach enables us to quantify the degree of observability of a stochastic model, what conventional deterministic methods cannot do. The excellent performance of the proposed method has been demonstrated by performing extensive simulations using a synchronous generator model with IEEE-DC1A exciter and the TGOV1 turbine governor.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Optimization Methods for Large-Scale Machine Learning

Here this paper provides a review and commentary on the past, present, and future of numerical optimization algorithms in the context of machine learning applications. Through case studies on text classification and the training of deep neural networks, we discuss how optimization problems arise in machine learning and what makes them challenging. A major theme of our study is that large-scale machine learning represents a distinctive setting in which the stochastic gradient (SG) method has traditionally played a central role while conventional gradient-based nonlinear optimization techniques typically falter. Based on this viewpoint, we present a comprehensive theory of a straightforward, yet versatile SG algorithm, discuss its practical behavior, and highlight opportunities for designing algorithms with improved performance. This leads to a discussion about the next generation of optimization methods for large-scale machine learning, including an investigation of two main streams of research on techniques that diminish noise in the stochastic directions and methods that make use of second-order derivative approximations.

97 MATHEMATICS AND COMPUTING↗

Semiclassical Wigner distribution for a two-mode entangled state generated by an optical parametric oscillator

We derive the steady-state solution of the Fokker-Planck equation that describes the dynamics of the nondegenerate optical parametric oscillator in the truncated Wigner representation of the density operator. We assume that the pump mode is strongly damped, which permits its adiabatic elimination. When the elimination is correctly executed, the resulting stochastic equations contain multiplicative noise terms and do not admit a potential solution. However, we develop a heuristic scheme leading to a satisfactory steady-state solution. This provides a clear view of the intracavity two-mode entangled state valid in all operating regimes of the optical parametric oscillator. A non-Gaussian distribution is obtained for the above threshold solution.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Parallel Environment for the Creation of Stochastics 1.0

PECOS is a computational library for creating and manipulating realizations of stochastic quantities, including scalar uncertain variables, random fields, and stochastic processes. It offers a unified interface to univariate and multivariate polynomial approximations using either orthogonal or interpolation polynomials; numerical integration drivers for Latin hypercube sampling, quadrature, cubature, and sparse grids; and fast Fourier transforms using third party libraries. The PECOS core also offers statistical utilities and transformations between various representations of stochastic uncertainty. PECOS provides a C++ API through which users can generate and transform realizations of stochastic quantities. It is currently used by Sandia's DAKOTA, Stokhos, and Encore software packages for uncertainty quantification and verification. PECOS generates random sample sets and multi-dimensional integration grids, typically used in forward propagation of scalar uncertainty in computational models (uncertainty quantification (UQ)). PECOS also generates samples of random fields (RFs) and stochastic processes (SPs) from a set of user-defined power spectral densities (PSDs). The RF/SP may be either Gaussian or non-Gaussian and either stationary or nonstationary, and the resulting sample is intended for run-time query by parallel finite element simulation codes. Finally, PECOS supports nonlinear transformations of random variables via the Nataf transformation and extensions.

Field Jr, Richard↗

Theory of weak scattering of stochastic electromagnetic fields from deterministic and random media

The theory of scattering of scalar stochastic fields from deterministic and random media is generalized to the electromagnetic domain under the first-order Born approximation. The analysis allows for determining the changes in spectrum, coherence, and polarization of electromagnetic fields produced on their propagation from the source to the scattering volume, interaction with the scatterer, and propagation from the scatterer to the far field. An example of scattering of a field produced by a {delta}-correlated partially polarized source and scattered from a {delta}-correlated medium is provided.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Stochastic Discrete Equation Method (sDEM) for two-phase flows

A new scheme for the numerical approximation of a five-equation model taking into account Uncertainty Quantification (UQ) is presented. In particular, the Discrete Equation Method (DEM) for the discretization of the five-equation model is modified for including a formulation based on the adaptive Semi-Intrusive (aSI) scheme, thus yielding a new intrusive scheme (sDEM) for simulating stochastic two-phase flows. Some reference test-cases are performed in order to demonstrate the convergence properties and the efficiency of the overall scheme. The propagation of initial conditions uncertainties is evaluated in terms of mean and variance of several thermodynamic properties of the two phases.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Acoustic wave propagation and stochastic effects in metamaterial absorbers

We show how stochastic variations of the effective parameters of anisotropic structured metamaterials can lead to increased absorption of sound. For this, we derive an analytical model based on the Bourret approximation and illustrate the immediate connection between material disorder and attenuation of the averaged field. We demonstrate numerically that broadband absorption persists at oblique irradiation and that the influence of red noise comprising short spatial correlation lengths increases the absorption beyond what can be archived with a structured but ordered system.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

A feedback control strategy for the airfoil system under non-Gaussian colored noise excitation

The stability of a binary airfoil with feedback control under stochastic disturbances, a non-Gaussian colored noise, is studied in this paper. First, based on some approximated theories and methods the non-Gaussian colored noise is simplified to an Ornstein-Uhlenbeck process. Furthermore, via the stochastic averaging method and the logarithmic polar transformation, one dimensional diffusion process can be obtained. At last by applying the boundary conditions, the largest Lyapunov exponent which can determine the almost-sure stability of the system and the effective region of control parameters is calculated.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Financial derivative pricing under probability operator via Esscher transfomation

The problem of pricing contingent claims has been extensively studied for non-Gaussian models, and in particular, Black- Scholes formula has been derived for the NIG asset pricing model. This approach was first developed in insurance pricing{sup 9} where the original distortion function was defined in terms of the normal distribution. This approach was later studied6 where they compared the standard Black-Scholes contingent pricing and distortion based contingent pricing. So, in this paper, we aim at using distortion operators by Cauchy distribution under a simple transformation to price contingent claim. We also show that we can recuperate the Black-Sholes formula using the distribution. Similarly, in a financial market in which the asset price represented by a stochastic differential equation with respect to Brownian Motion, the price mechanism based on characteristic Esscher measure can generate approximate arbitrage free financial derivative prices. The price representation derived involves probability Esscher measure and Esscher Martingale measure and under a new complex valued measure φ (u) evaluated at the characteristic exponents φ{sub x}(u) of X{sub t} we recuperate the Black-Scholes formula for financial derivative prices.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗