A Newton-gradient Method for Non-linear Problems in Hilbert Space Technical Report No. 7
Newton-gradient method for non-linear problems in Hilbert space
NEWTON THEORY↗
SEARCH · Search NASA
Results for “Newton Method”
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.
Newton-Krylov Method in Applications of Solving Two-phase Problems using Drift Flux Model
In this paper, Jacobian-free Newton-Krylov (JFNK) method is investigated in an application to implicitly solve two-phase flow problems using four-equation drift flux model. The closure models include the EPRI drift flux correlations and additional constitutive models to determine flow regimes, wall boiling and interfacial heat/mass transfer, two-phase flow wall friction, etc. Different from many traditional computer codes, fully implicit methods are used for the time integration. The resulted nonlinear discretized equation system is solved using the JFNK method. Expensive and error prone derivation and implementation of the analytical Jacobian matrix are avoided. Numerical results are successfully validated using existing experimental data on flow boiling under forced convection conditions in both pipe and rod bundle geometries.
42 ENGINEERING↗
One-Dimensional Ablation with Pyrolysis Gas Flow Using a Full Newton's Method and Finite Control Volume Procedure
The development and verification of a one-dimensional material thermal response code with ablation is presented. The implicit time integrator, control volume finite element spatial discretization, and Newton's method for nonlinear iteration on the entire system of residual equations have been implemented and verified for the thermochemical ablation of internally decomposing materials. This study is a continuation of the work presented in "One-Dimensional Ablation with Pyrolysis Gas Flow Using a Full Newton's Method and Finite Control Volume Procedure" (AIAA-2006-2910), which described the derivation, implementation, and verification of the constant density solid energy equation terms and boundary conditions. The present study extends the model to decomposing materials including decomposition kinetics, pyrolysis gas flow through the porous char layer, and a mixture (solid and gas) energy equation. Verification results are presented for the thermochemical ablation of a carbon-phenolic ablator which involves the solution of the entire system of governing equations.
Amar, Adam J.↗
Development of iterative techniques for the solution of unsteady compressible viscous flows
Efficient iterative solution methods are being developed for the numerical solution of two- and three-dimensional compressible Navier-Stokes equations. Iterative time marching methods have several advantages over classical multi-step explicit time marching schemes, and non-iterative implicit time marching schemes. Iterative schemes have better stability characteristics than non-iterative explicit and implicit schemes. Thus, the extra work required by iterative schemes can also be designed to perform efficiently on current and future generation scalable, missively parallel machines. An obvious candidate for iteratively solving the system of coupled nonlinear algebraic equations arising in CFD applications is the Newton method. Newton's method was implemented in existing finite difference and finite volume methods. Depending on the complexity of the problem, the number of Newton iterations needed per step to solve the discretized system of equations can, however, vary dramatically from a few to several hundred. Another popular approach based on the classical conjugate gradient method, known as the GMRES (Generalized Minimum Residual) algorithm is investigated. The GMRES algorithm was used in the past by a number of researchers for solving steady viscous and inviscid flow problems with considerable success. Here, the suitability of this algorithm is investigated for solving the system of nonlinear equations that arise in unsteady Navier-Stokes solvers at each time step. Unlike the Newton method which attempts to drive the error in the solution at each and every node down to zero, the GMRES algorithm only seeks to minimize the L2 norm of the error. In the GMRES algorithm the changes in the flow properties from one time step to the next are assumed to be the sum of a set of orthogonal vectors. By choosing the number of vectors to a reasonably small value N (between 5 and 20) the work required for advancing the solution from one time step to the next may be kept to (N+1) times that of a noniterative scheme. Many of the operations required by the GMRES algorithm such as matrix-vector multiplies, matrix additions and subtractions can all be vectorized and parallelized efficiently.
Sankar, Lakshmi N.↗
SOLVING CONJUGATE HEAT TRANSFER PROBLEM IN TWO-PHASE FLOW USING NEWTON-KRYLOV METHOD
In this paper, we successfully demonstrated an application of the Jacobian-free Newton Krylov (JFNK) method in solving the conjugated two-phase channel flow and heat conduction problem, in a fully coupled and fully implicit manner. Such a problem represents the most fundamental piece of safety analysis codes, for example the heat transfer between fuel rods and reactor coolant during postulated accidental scenarios. The two-phase flow is modeled with the widely accepted one-dimensional two-fluid two-phase flow model, and the heat conduction is modeled with two-dimensional heat conduction model. The coupling between these two physics is realized through the wall heat transfer term in fluid equations and heat flux boundary conditions of heat conduction model. For spatial discretization, both the first-order and second- order upwind schemes are considered for flow equations, while the standard central differencing method is used for the heat conduction model. For time integration, both the first-order backward Euler (BDF1) and second-order BDF2 schemes are considered. Several treatments were also introduced to enhance the robustness of the JFNK method in solving such a problem that is discontinuous in nature because of the discrete closure correlations. These treatments include the re-introduction of a transition region between the slug and annular mist flow regimes, and a simplification in the interfacial drag term. Numerical results have been extensively validated with experimental data in both vertical pipe and rod bundle geometries with a wide range of thermal-hydraulics conditions, including pressure, inlet subcooling, wall heat flux, outlet void fraction, and etc. Further benchmark with the RELAP5-3D code confirms the correct implementation of wall heat transfer models.
42 ENGINEERING↗
A log-barrier Newton-CG method for bound constrained optimization with complexity guarantees
Abstract We describe an algorithm based on a logarithmic barrier function, Newton’s method and linear conjugate gradients that seeks an approximate minimizer of a smooth function over the non-negative orthant. We develop a bound on the complexity of the approach, stated in terms of the required accuracy and the cost of a single gradient evaluation of the objective function and/or a matrix-vector multiplication involving the Hessian of the objective. The approach can be implemented without explicit calculation or storage of the Hessian.
O’Neill, Michael↗
A modified Newton-Raphson method for determining stability derivatives from flight data
Stability derivatives from flight data of fixed-wing aircraft determined by modified Newton-Raphson method
Iliff, K. W.↗
On Newton's method for Riccati equation solution
It is shown that the assumptions of controllability and observability in two theorems of Kleinman (1968, 1970) concerning Newton's method for the Ricatti equation can be weakened to stabilizability and detectability. Empirically, this has been known for some time.
Sandell, N. R., Jr.↗
Calculations of transonic flows with shocks using Newton's method and direct solver. II - Solution of Euler equations
Transonic flows with shocks are simulated using steady Euler equations and by simultaneously solving the resulting nonlinear algebraic equations using Newton's method. At each iteration, a direct solver computes the corrections and the process is repeated until convergence is achieved. The corrections and errors are reduced quadratically with the present method, allowing solutions of machine accuracy to be obtained in a few steps. Nonunique inviscid solutions and nonunique solutions of the Navier Stokes equations for quasi-one-dimensional flows in nozzles are presented. Calculations are also presented for steady two-dimensional inviscid flows around a cylinder in the transonic regime.
Hafez, M.↗
Larger Convergence Zones for Newton's Method
Iterative technique applies over wider range of initial guesses. New theorem describes convergence zone of Newton's iterative method for finding zeros of real function. Involves two points, Xp and Xp*, called primary conjugate points. If exact solution lies between these points (Xp is less than Xz is less than Xp*) and no other conjugate points in interval, then according to theorem, subsequent iterations will converge upon exact solution if initial guess lies in interval.
Campbell, C. W.↗
Newton-Krylov method in solving drift-flux two-phase flow problems for both pre-CHF and post-CHF conditions at steady state
Furthermore, this paper presents numerical simulations of two-phase flow heat transfer for both pre-CHF and post-CHF conditions typical in light water nuclear reactors. The conjugate heat transfer problem is modeled with one-dimensional drift-flux two-phase flow model for coolant flow and two-dimensional heat conduction equation in solid walls. Fully implicit time integration schemes were employed, and the resulted non-linear equations were solved with a Newton-Krylov method. For the drift-flux two-phase flow model, the EPRI drift-flux closure correlation is used to model the relative motion between the two phases. Additional closure correlations were adopted from REALP5-3D to determine two-phase flow regime, wall boiling heat and mass transfer, interfacial heat and mass transfer, and two-phase flow frictional pressure drop. Three sets of experiments were used for code validation, including Bartolomei subcooled flow boiling in vertical pipes, FRIGG boiling tests in vertical bundles, and Bennett heated tube tests. RELAP5-3D simulation results were also provided to perform code-to-code benchmark. Overall, numerical results of this work showed good agreements with both experimental data and RELAP5-3D simulation results. Discrepancy between them were also identified in post-CHF region, which suggests necessary further improvement.
42 ENGINEERING↗
Application of Jacobian-free Newton–Krylov method in implicitly solving two-fluid six-equation two-phase flow problems: Implementation, validation and benchmark
This work represents a first-of-its-kind successful application to employ advanced numerical methods in solving realistic two-phase flow problems with two-fluid six-equation two-phase flow model. These advanced numerical methods include high-resolution spatial discretization scheme with staggered grids (high-order) fully implicit time integration schemes, and Jacobian-free Newton–Krylov (JFNK) method as the nonlinear solver. The computer code developed in this work has been extensively validated with existing experimental flow boiling data in vertical pipes and rod bundles, which cover wide ranges of experimental conditions, such as pressure, inlet mass flux, wall heat flux and exit void fraction. Additional code-to-code benchmark with the RELAP5-3D code further verifies the correct code implementation. The combined methods employed in this work exhibit strong robustness in solving two-phase flow problems even when phase appearance (boiling) and realistic discrete flow regimes are considered. Transitional flow regimes used in existing system analysis codes, normally introduced to overcome numerical difficulty, were completely removed in this work. As a result, this in turn provides the possibility to utilize more sophisticated flow regime maps in the future to further improve simulation accuracy.
42 ENGINEERING↗
Dielectric constant extraction of graphene nanostructured on SiC substrates from spectroscopy ellipsometry measurement using Gauss–Newton inversion method
The extraction of the dielectric constant of nanostructured graphene on SiC substrates from spectroscopy ellipsometry measurement using the Gauss-Newton inversion (GNI) method has been done. This study aims to calculate the dielectric constant and refractive index of graphene by extracting the value of ψ and Δ from the spectroscopy ellipsometry measurement using GNI method and comparing them with previous result which was extracted using Drude-Lorentz (DL) model. The results show that GNI method can be used to calculate the dielectric constant and refractive index of nanostructured graphene on SiC substratesmore faster as compared to DL model. Moreover, the imaginary part of the dielectric constant values and coefficient of extinction drastically increases at 4.5 eV similar to that of extracted using known DL fitting. The increase is known due to the process of interband transition and the interaction between the electrons and electron-hole at M-points in the Brillouin zone of graphene.
71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
Implementing a unified solver for nonlinearly constrained optimization
SQP and interior-point methods (also referred to as Lagrange-Newton methods) typically share key algorithmic components, such as strategies for computing descent directions and mechanisms that promote global convergence. Building on this insight, we introduce a unifying framework with eight building blocks that abstracts the workflows of Lagrange-Newton methods. We then present Uno, a modular C++ solver that implements our unifying framework and allows the automatic combination of a wide range of strategies with no programming effort from the user. Uno is meant to (1) organize mathematical optimization strategies into a coherent hierarchy; (2) offer a wide range of efficient and robust methods that can be compared for a given instance; (3) enable researchers to experiment with novel optimization strategies; and (4) reduce the cost of development and maintenance of multiple optimization solvers. Uno’s software design allows user to compose new customized solvers for emerging optimization areas such as robust optimization or optimization problems with complementarity constraints, while building on reliable nonlinear optimization techniques. We demonstrate that Uno is highly competitive against state-of-the-art solvers filterSQP, IPOPT, SNOPT, MINOS, LANCELOT, LOQO, and CONOPT on a subset of 429 small problems from the CUTE collection. Uno is available as open-source software under the MIT license at https://github.com/cvanaret/Uno and via its C, Julia, Python, Fortran, and AMPL interfaces.
97 MATHEMATICS AND COMPUTING↗
Impacts of Hybrid Parallelism and Vectorization on the Performance of Newton-Krylov Methods in Computational Aerodynamics
Finding the numerical solution of moderate and high-fidelity aerodynamics problems on modern computer architectures involves, 1) decomposing the domain into smaller regions of nearly equal size, and 2) allocating computational resources for calculations on each domain and communication between domains. Modern computer clusters are composed from hierarchies of processing, memory, and communication resources with varying capabilities and latencies.This paper focuses on the combination of domain decomposition provided by ParMETIS [1]and Newton-Krylov Methods [2–5] for the solution of Computational Aerodynamics problems of interest to NASA. Herein, trade-offs encountered when mapping aerodynamics problems to modern computer architectures are explored through examples and discussions of trade-offs in parallelism from MPI [6], Open MP [7], and vectorization as partition sizes and computational resources are varied. An example of the impact that domain decomposition and MPI+OpenMPresource allocation can have on an adjoint calculation is presented in this abstract. The full paper will include more detailed examples, discussions of difficulties and potential methods to overcome them, and topics identified for future study.
Computational Aerodynamics, Hybrid Parallelism, Ve↗
A new flux conserving Newton's method scheme for the two-dimensional, steady Navier-Stokes equations
A new numerical method is developed for the solution of the two-dimensional, steady Navier-Stokes equations. The method that is presented differs in significant ways from the established numerical methods for solving the Navier-Stokes equations. The major differences are described. First, the focus of the present method is on satisfying flux conservation in an integral formulation, rather than on simulating conservation laws in their differential form. Second, the present approach provides a unified treatment of the dependent variables and their unknown derivatives. All are treated as unknowns together to be solved for through simulating local and global flux conservation. Third, fluxes are balanced at cell interfaces without the use of interpolation or flux limiters. Fourth, flux conservation is achieved through the use of discrete regions known as conservation elements and solution elements. These elements are not the same as the standard control volumes used in the finite volume method. Fifth, the discrete approximation obtained on each solution element is a functional solution of both the integral and differential form of the Navier-Stokes equations. Finally, the method that is presented is a highly localized approach in which the coupling to nearby cells is only in one direction for each spatial coordinate, and involves only the immediately adjacent cells. A general third-order formulation for the steady, compressible Navier-Stokes equations is presented, and then a Newton's method scheme is developed for the solution of incompressible, low Reynolds number channel flow. It is shown that the Jacobian matrix is nearly block diagonal if the nonlinear system of discrete equations is arranged approximately and a proper pivoting strategy is used. Numerical results are presented for Reynolds numbers of 100, 1000, and 2000. Finally, it is shown that the present scheme can resolve the developing channel flow boundary layer using as few as six to ten cells per channel width, depending on the Reynolds number.
Scott, James R.↗
A solution to the Navier-Stokes equations based upon the Newton Kantorovich method
An implicit finite difference scheme based on the Newton-Kantorovich technique was developed for the numerical solution of the nonsteady, incompressible, two-dimensional Navier-Stokes equations in conservation-law form. The algorithm was second-order-time accurate, noniterative with regard to the nonlinear terms in the vorticity transport equation except at the earliest few time steps, and spatially factored. Numerical results were obtained with the technique for a circular cylinder at Reynolds number 15. Results indicate that the technique is in excellent agreement with other numerical techniques for all geometries and Reynolds numbers investigated, and indicates a potential for significant reduction in computation time over current iterative techniques.
Davis, J. E.↗
Development of Segregated Thermal-Hydraulics Solvers in MOOSE
The simulation of fluid flows is an essential part of the design and analysis of nuclear systems. Algorithms able to simulate flows at different fidelity levels are available in the Multiphysics Object-Oriented Simulation Environment (MOOSE) and MOOSE-based applications such as Pronghorn \cite{novak2018pronghorn}, Pronghorn-Subchannel, RELAP-7, and SAM. Currently, significant effort is being invested in the development of coarse-mesh Computational Fluid Dynamics (CFD) capabilities within MOOSE and Pronghorn for the simulation of Generation IV nuclear reactors. Traditionally, the solution algorithms in MOOSE have relied on Newton or quasi-Newton methods (such as the preconditioned Jacobian-free Newton-Krylov method) where residuals and Jacobians (or approximations thereof) are constructed. Both Newton and quasi-Newton methods require the solution of a linear system at each nonlinear Newton iteration with the Jacobian as the system matrix. The Jacobian contains blocks originating from all variables in the problem (i.e., for thermal-hydraulics at least pressure, velocities, and temperature). Due to the formulation of the problem in a general multiphysics setting on unstructured mesh, creating a good preconditioner for the linear system can be challenging, thus many fluid applications have utilized direct solver-based methods such as LU factorization. However, with increasing system size and complexity in multi-dimensional problems, the direct solution of linear systems becomes computationally expensive both in execution time and and memory. For this reason, recent effort has focused on adapting segregated solution algorithms for CFD problems in MOOSE. These algorithms use fixed-point iteration between segregated systems whose assembly and preconditioning are easier those of the monolithic system. Initial results show that the segregated solution algorithm outperforms the monolithic approach in terms of memory usage and for large 3D problems in terms of CPU time as well.
42 ENGINEERING↗