High-order well-balanced methods for systems of balance laws: a control-based approach

2025

Read High-order well-balanced methods for systems of balance laws: a control-based approach (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-02047.txt.

Introduction

Gómez-Bueno, Castro Díaz, and Parés presented this numerical methods paper in twenty twenty-five. It deals with balance laws, equations that track flows with sources, such as water depth or gas motion. Good numerical schemes should keep steady states exact rather than letting errors creep in. Earlier schemes did this by solving a hard problem in each grid cell at each step.

Those cell problems worked only when steady states were known in closed form. The authors recast each problem as a control task on a related equation system. A gradient from a companion problem then drives a Newton solver. They test the method on systems from a simple wave equation to shallow water and gas flow with gravity. Results confirm accuracy and exact steady states.

Catalog record

YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE66 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-02047

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

High-order well-balanced methods for systems of balance laws: a control-based approach Abstract In some previous works, two of the authors have introduced a strategy to develop high-order numerical methods for systems of balance laws that preserve all the stationary solutions of the system. The key ingredient of these methods is a well-balanced reconstruction operator. A strategy has been also introduced to modify any standard reconstruction operator like MUSCL, ENO, CWENO, etc. in order to be well-balanced. This strategy involves a non-linear problem at every cell at every time step that consists in finding the stationary solution whose average is the given cell value. So far this strategy has been only applied to systems whose stationary solution are known either in explicit or implicit form. The goal of this paper is to present a general implementation of this technique that can be applied to any system of balance laws. To do this, the nonlinear problems to be solved in the reconstruction procedure are interpreted as control problems: they consist in finding a solution of an ODE system whose average at the computation interval is given. These problems are written in functional form and the gradient of the functional is computed on the basis of the adjoint problem. Newton’s method is applied then to solve the problems. Special care is put to analyze the effects of computing the averages and the source terms using quadrature formulas. To test their efficiency and well-balancedness, the methods are applied to a number of systems of balance laws, ranging from easy academic systems consisting of Burgers equation with some nonlinear source terms to the shallow water equations or Euler equations of gas dynamics with gravity effects. I. Gómez, M.J. Castro, C. Parés University of Málaga. 1 Introduction Let us consider a PDE system of the form: U_{t}(x,t)+f(U(x,t))_{x}=S(U(x,t))H_{x}(x),\quad x\in\mathbb{R},\,t>0, (1) where U(x,t) takes values on an open convex set \Omega\subset\mathbb{R}^{N} , f:\Omega\longrightarrow\mathbb{R}^{N} is the flux function, S:\Omega\longrightarrow\mathbb{R}^{N} , and H is a continuous known function from \mathbb{R} to \mathbb{R} (possibly the identity function H(x)=x ). It is supposed that system ( 1 ) is strictly hyperbolic, that is, D_{f}(U)=\displaystyle\frac{\partial f}{\partial U}(U) has N real different eigenvalues and eigenvectors. Moreover, we suppose that \lambda_{i}(U)\neq 0,\,i=1,\ldots,N . Systems of the form ( 1 ) have non trivial stationary solutions that satisfy the ODE system: f(U)_{x}=S(U)H_{x}. (2) A numerical method is said to be well-balanced if it solves exactly or with enhanced accuracy all the stationary solutions of the system or, at least, a relevant family of them. The use of methods with this property is of major importance when the waves generated for small perturbations of a steady state are going to be simulated: this is the case, for instance, for tsunami waves in the Ocean. Well-balanced methods have been studied by many authors: see, for instance, [ 1 ] , [ 2 ] , [ 3 ] , [ 8 ] , [ 9 ] , [ 10 ] , [ 11 ] , [ 13 ] , [ 14 ] , [ 15 ] , [ 16 ] , [ 17 ] , [ 21 ] , [ 19 ] , [ 20 ] , [ 22 ] , [ 24 ] , [ 25 ] , [ 26 ] , [ 27 ] , [ 28 ] , [ 29 ] , [ 30 ] , [ 31 ] , [ 32 ] , [ 34 ] ,… See [ 6 ] and its references for a recent review on this topic. Recently, in [ 7 ] the following family of semidiscrete high-order well-balanced finite-volume methods for ( 1 ) has been discussed: \frac{dU_{i}}{dt}=-\frac{1}{\Delta x}\left(F_{i+\frac{1}{2}}(t)-F_{i-\frac{1}{2}}(t)\right)+\frac{1}{\Delta x}\int_{x_{i-\frac{1}{2}}}^{x_{i+\frac{1}{2}}}S(P_{i}^{t}(x))H_{x}(x)\,dx, (3) where: • I_{i}=\left[x_{i-\frac{1}{2}},x_{i+\frac{1}{2}}\right] are the computational cells, whose length \Delta x is supposed to be constant for simplicity; • U_{i}(t) is the approximation of the average of the exact solution at the i th cell at time t , that is, U_{i}(t)\cong\frac{1}{\Delta x}\int_{x_{i-\frac{1}{2}}}^{x_{i+\frac{1}{2}}}U(x,t)\,dx; • P_{i}^{t}(x) is the approximation of the solution at the i th cell given by a high-order reconstruction operator from the sequence of cell averages \{U_{i}(t)\} : P_{i}^{t}(x)=P_{i}(x;\{U_{j}(t)\}_{j\in\mathcal{S}_{i}}); where \mathcal{S}_{i} denotes the set of indexes of the cells belonging to the stencil of the i th cell. • F_{i+\frac{1}{2}}=\mathbb{F}(U_{i+\frac{1}{2}}^{t,-},U_{i+\frac{1}{2}}^{t,+}) , where U_{i+\frac{1}{2}}^{t,\pm} are the reconstructed states at the intercells, i.e. U_{i+\frac{1}{2}}^{t,-}=P_{i}^{t}(x_{i+\frac{1}{2}}),\quad U_{i+\frac{1}{2}}^{t,+}=P_{i+1}^{t}(x_{i+\frac{1}{2}}), and \mathbb{F} is a consistent first order numerical flux. It can be then easily shown that, if the reconstruction operator is well-balanced for a stationary solution U of ( 1 ) then the numerical method is also well-balanced for U according to the following definitions: Definition 1 . Given a stationary solution U of ( 1 ): • The numerical method ( 3 ) is said to be well-balanced for U if the vector of cell averages of U is an equilibrium of the ODE system ( 3 ). • The reconstruction operator is said to be well-balanced for U if P_{i}(x)=U(x),\quad\forall x\in[x_{i-\frac{1}{2}},x_{i+\frac{1}{2}}],\,\forall i, (4) where P_{i} is the approximation of U obtained by applying the reconstruction operator to the vector of cell averages of U . The following strategy to design a well-balanced reconstruction operator P_{i} on the basis of a standard operator Q_{i} was introduced in [ 4 ] : given a family of cell values \{U_{i}\} , at every cell I_{i}=[x_{i-\frac{1}{2}},x_{i+\frac{1}{2}}] : 1. Look for the stationary solution U_{i}^{*}(x) such that: \frac{1}{\Delta x}\int_{x_{i-\frac{1}{2}}}^{x_{i+\frac{1}{2}}}U_{i}^{*}(x)\,dx=U_{i}. (5) 2. Apply the reconstruction operator to the cell values \{V_{j}\}_{j\in S_{i}} given by V_{j}=U_{j}-\frac{1}{\Delta x}\int_{x_{j-\frac{1}{2}}}^{x_{j+\frac{1}{2}}}U_{i}^{*}(x)\,dx, to obtain: Q_{i}(x)=Q_{i}(x;\{V_{j}\}_{j\in\mathcal{S}_{i}}). 3. Define P_{i}(x)=U_{i}^{*}(x)+Q_{i}(x). (6) It can be then easily shown that the reconstruction operator P_{i} in ( 6 ) is well-balanced for every stationary solution provided that the reconstruction operator Q_{i} is exact for the null function. Moreover, if Q_{i} is conservative, then P_{i} is conservative, that is, \frac{1}{\Delta x}\int_{x_{i-\frac{1}{2}}}^{x_{i+\frac{1}{2}}}P_{i}(x)\,dx=U_{i},\,\forall i, and P_{i} is high-order accurate provided that the stationary solutions are smooth. The well-balanced property of the method can be lost if a quadrature formula is used to compute the integral appearing at right-hand side of ( 3 ). In order to circumvent this difficulty, the authors of [ 7 ] proposed to rewrite the methods as follows \begin{split}\frac{dU_{i}}{dt}&=-\frac{1}{\Delta x}\left(F_{i+\frac{1}{2}}(t)-f\left(U_{i}^{t,*}(x_{i+\frac{1}{2}})\right)-F_{i-\frac{1}{2}}(t)+f\left(U_{i}^{t,*}(x_{i-\frac{1}{2}})\right)\right)\\ &+\frac{1}{\Delta x}\int_{x_{i-\frac{1}{2}}}^{x_{i+\frac{1}{2}}}\left(\left[S(P_{i}^{t}(x))-S(U_{i}^{t,*}(x))\right]H_{x}(x)\right)\,dx.\end{split} (7) where U_{i}^{t,*} is the stationary solution found in ( 5 ) at the i th cell and time t . Once the method has been rewritten in this form, a quadrature formula can be applied to the integral without losing the well-balanced property. The main difficulty when this strategy is applied comes from the first step of the well-balanced reconstruction operator: a nonlinear problem of the form ( 5 ) has to be solved at every time step. Since the stationary solutions of ( 1 ) are the solutions of the ODE system f(U)_{x}=S(U)\,H_{x}, (8) then, problem ( 5 ) is equivalent to find the solution of an ODE system with prescribed average in the integration domain. In some cases, the explicit form of the general solution of the ODE is known and ( 5 ) can be solved by hand or by using standard iterative methods for nonlinear problems: this is the case, for instance, of the shallow water equations (see [ 7 ] and its references). The goal of this paper is to

Read the complete text