scholarly journals Inverses of SBP-SAT Finite Difference Operators Approximating the First and Second Derivative

2021 ◽  
Vol 89 (2) ◽  
Author(s):  
Sofia Eriksson

AbstractThe scalar, one-dimensional advection equation and heat equation are considered. These equations are discretized in space, using a finite difference method satisfying summation-by-parts (SBP) properties. To impose the boundary conditions, we use a penalty method called simultaneous approximation term (SAT). Together, this gives rise to two semi-discrete schemes where the discretization matrices approximate the first and the second derivative operators, respectively. The discretization matrices depend on free parameters from the SAT treatment. We derive the inverses of the discretization matrices, interpreting them as discrete Green’s functions. In this direct way, we also find out precisely which choices of SAT parameters that make the discretization matrices singular. In the second derivative case, it is shown that if the penalty parameters are chosen such that the semi-discrete scheme is dual consistent, the discretization matrix can become singular even when the scheme is energy stable. The inverse formulas hold for SBP-SAT operators of arbitrary order of accuracy. For second and fourth order accurate operators, the inverses are provided explicitly.

1980 ◽  
Vol 20 (03) ◽  
pp. 120-128 ◽  
Author(s):  
Stephen H. Leventhal

Abstract This paper describes the application of the fourth-order operator compact implicit (OCI) method to multiphase flow through porous media. A conservative form of the method is developed and used to solve a one-dimensional, two-phase, immiscible waterflood problem.The results of numerical experiments demonstrate the ability to track sharp fronts with relatively few mesh points. Comparisons with conventional finite-difference methods and variational methods are given. Introduction In recent years, new work in the numerical solution of boundary-value problems for differential equations has been in the direction of high-order methods - that is, methods whose order of accuracy is higher than the first- and second-order methods generally in use. These high-order methods have been developed for both finite-difference, finite-element, and collocation techniques.High-order finite-difference techniques are formed by expanding the computational molecule of the problem. For example, while a second problem. For example, while a second order-accurate formula for a second derivative involves three points, a fourth-order-accurate formula involves five points. This growth in the number of points in the approximation leads to an increase in points in the approximation leads to an increase in the bandwidth of matrices and the need to impose special difference formulas near the boundary.Variational methods achieve higher-order accuracy by using a higher-degree piecewise polynomial subspace for its approximation. The problem of special treatment near the boundary does not exist, but the increase in the degree of the polynomial also leads to an increase in the work for a given number of mesh points. However, this increase is offset by the need for fewer mesh points to achieve a given accuracy. Thus, on model test problems it has been shown that the total work is less for high-order methods.An alternative to the approach of increasing bandwidth to increase order of accuracy is the operator compact implicit (OCI) method. In this method, an approximation to the spatial part of the differential operator is sought using not only the value of the function at adjacent points but also the value of the operator at these points. By this technique, an implicit relationship is defined between the differential operator and the function on the most compact set of points possible. For example, an implicit fourth-order relationship for a second derivative may be derived using only three points.The OCI method has been applied successfully to two-point boundary-value problems, first- and second-order wave equations, diffusion-convection equations, and viscous flow problems. In this paper, a form of the OCI method problems. In this paper, a form of the OCI method applicable to equations in conservation form is presented and the method is applied to the equations presented and the method is applied to the equations of flow through porous media.The specific example considered is a one-dimensional, two-phase waterflood problem studied by Spivak et al. After presenting the form of the two-phase flow equations to be used, the OCI method is developed. In the implementation of the method it is necessary to define transmissibilities at half mesh points. Therefore, we next examine the problem of interpolation. Finally, numerical results problem of interpolation. Finally, numerical results are presented on the performance of the method for the one-dimensional waterflood problem. Development of the Differential Equations The physical problem that the OCI method is applied to in this study is a one-dimensional, compressible, two-phase, immiscible waterflood problem. SPEJ P. 120


2021 ◽  
Vol 61 (SI) ◽  
pp. 49-58
Author(s):  
Tomáš Bodnár ◽  
Philippe Fraunié ◽  
Karel Kozel

This paper presents the general modified equation for a family of finite-difference schemes solving one-dimensional advection equation. The whole family of explicit and implicit schemes working at two time-levels and having three point spatial support is considered. Some of the classical schemes (upwind, Lax-Friedrichs, Lax-Wendroff) are discussed as examples, showing the possible implications arising from the modified equation to the properties of the considered numerical methods.


2020 ◽  
Vol 23 (1-4) ◽  
Author(s):  
Martin J. Gander ◽  
Thibaut Lunet

AbstractWe develop new error estimates for the one-dimensional advection equation, considering general space-time discretization schemes based on Runge–Kutta methods and finite difference discretizations. We then derive conditions on the number of points per wavelength for a given error tolerance from these new estimates. Our analysis also shows the existence of synergistic space-time discretization methods that permit to gain one order of accuracy at a given CFL number. Our new error estimates can be used to analyze the choice of space-time discretizations considered when testing Parallel-in-Time methods.


2014 ◽  
Vol 16 (2) ◽  
pp. 345-358 ◽  
Author(s):  
Jan Nordström ◽  
Qaisar Abbas ◽  
Brittany A. Erickson ◽  
Hannes Frenander

AbstractA new weak boundary procedure for hyperbolic problems is presented. We consider high order finite difference operators of summation-by-parts form with weak boundary conditions and generalize that technique. The new boundary procedure is applied near boundaries in an extended domain where data is known. We show how to raise the order of accuracy of the scheme, how to modify the spectrum of the resulting operator and how to construct non-reflecting properties at the boundaries. The new boundary procedure is cheap, easy to implement and suitable for all numerical methods, not only finite difference methods, that employ weak boundary conditions. Numerical results that corroborate the analysis are presented.


JURTEKSI ◽  
2019 ◽  
Vol 6 (1) ◽  
pp. 65-70
Author(s):  
Havid Syafwan ◽  
Mahdhivan Syafwan ◽  
William Ramdhan ◽  
Riki Andri Yusda

Abstract: In this paper we discuss the development of computational application program in MATLAB to calculate the second derivative of a function by using finite difference method for various types (forward, backward, or central) and any order of accuracy. The MATLAB program extends the similar one which has been developed for the first derivative. The program is designed as an implementation of the explicit formulas of finite difference formulated by Khan and Ohba. Through this program, one can easily calculate the second derivative of a function numerically with any order of accuracy and visualize it in a graphical form. Error of numerical derivatives in each order of accuracy confirm the validity of the obtained results. Keywords: finite difference method, explicit formula, second derivative, MATLAB programming Abstrak: Dalam makalah ini dibahas pengembangan program aplikasi komputasi pada MATLAB untuk menghitung turunan kedua suatu fungsi dengan menggunakan metode beda hingga pada berbagai tipe (maju, mundur, atau pusat) dan sebarang orde ketelitian. Program MATLAB ini melanjutkan program serupa yang telah dibuat untuk turunan pertama. Program ini dirancang sebagai implementasi dari rumus eksplisit beda hingga yang diformulasi oleh Khan dan Ohba. Melalui program ini, seseorang dengan mudah dapat menghitung turunan kedua suatu fungsi secara numerik dengan sebarang orde ketelitian dan memvisualisasinya dalam bentuk grafik. Galat dari turunan numerik di setiap orde ketelitian mengkonfirmasi kevalidan hasil yang diperoleh. Kata kunci: metode beda hingga, rumus eksplisit, turunan kedua, pemrograman MATLAB  


Sign in / Sign up

Export Citation Format

Share Document