## A wave-propagation method for conservation laws and balance laws with spatially varying flux functions (2002)

Venue: | SIAM J. Sci. Comput |

Citations: | 42 - 5 self |

### BibTeX

@ARTICLE{Bale02awave-propagation,

author = {Derek S. Bale and Randall J. Leveque and Sorin Mitran and James A. Rossmanith},

title = {A wave-propagation method for conservation laws and balance laws with spatially varying flux functions},

journal = {SIAM J. Sci. Comput},

year = {2002},

volume = {24},

pages = {2002}

}

### OpenURL

### Abstract

Abstract. We study a general approach to solving conservation laws of the form qt+f(q, x)x =0, where the flux function f(q, x) has explicit spatial variation. Finite-volume methods are used in which the flux is discretized spatially, giving a function fi(q) over the ith grid cell and leading to a generalized Riemann problem between neighboring grid cells. A high-resolution wave-propagation algorithm is defined in which waves are based directly on a decomposition of flux differences fi(Qi)− fi−1(Qi−1) into eigenvectors of an approximate Jacobian matrix. This method is shown to be second-order accurate for smooth problems and allows the application of wave limiters to obtain sharp results on discontinuities. Balance laws qt + f(q, x)x = ψ(q, x) are also considered, in which case the source term is used to modify the flux difference before performing the wave decomposition, and an additional term is derived that must also be included to obtain full accuracy. This method is particularly useful for quasi-steady problems close to steady state. Key words. finite-volume methods, high-resolution methods, conservation laws, source terms, discontinuous flux functions AMS subject classifications. 65M06, 35L65 PII. S106482750139738X

### Citations

638 | Approximate Riemann Solvers, Parameter Vectors and Difference Schemes
- Roe
- 1981
(Show Context)
Citation Context ...ndition (1.12) A i−1/2(Qi − Qi−1)=fi(Qi) − fi−1(Qi−1) happens to be satisfied. For an autonomous system (fi(q) ≡ f(q)) this reduces to the condition imposed in defining the “Roe average” A i−1/2 (see =-=[27]-=-), and considerable research has gone into defining averaged Jacobians with this property for specific nonlinear problems. The main novel feature of the algorithm we present here is the following. We ... |

518 |
Numerical Methods for Conservation Laws
- LeVeque
- 1992
(Show Context)
Citation Context ... is not a reasonable model of reality. To more properly model this situation we should use a nonlinear traffic model in which the velocity depends on the density, for example, the classical flux (see =-=[19]-=-, [31]) (2.7) fi(q)=ui(1 − q)q, where ui is now the maximum speed (at q = 0)but the speed drops linearly to 0 as the density increases to q = 1 (bumper-to-bumper traffic). In this case even taking ui−... |

86 | Wave propagation algorithms for multidimensional hyperbolic systems
- LeVeque
- 1997
(Show Context)
Citation Context ...(q)x =0, where f depends only on q, has been studied extensively, and a variety of numerical methods have been developed. We concentrate on extending the high-resolution wavepropagation algorithms of =-=[20]-=- to the case of a spatially varying flux function. These are finite-volume methods in which a cell average (1.3) Q n i ≈ 1 ∆x � xi+1/2 x i−1/2 q(x, tn) dx is updated in each time step. This is done us... |

78 | Balancing source terms and flux gradients in high resolution Godunov methods
- LeVeque
- 1998
(Show Context)
Citation Context ...d we would like to develop a wave-propagation method in which the waves model only the information that should propagate relative to the background steady state solution. One approach was proposed in =-=[21]-=-, but we now believe that a better approach is suggested by the f-wave algorithm presented above, even in the autonomous case, and has the additional advantage of applying also with spatially varying ... |

61 | One-dimensional transport equations with discontinuous coefficients
- Bouchut, James
- 1998
(Show Context)
Citation Context ...mpose appropriate admissibility conditions to select the physically correct solution, and this can be more subtle than in the autonomous case. For some discussions of such problems, see, for example, =-=[5, 6, 9, 10, 12, 15, 17, 25, 26]-=-. We are continuing to study extensions of our algorithm to particular problems of this form. 3. The wave-propagation algorithms. We first briefly summarize the wavepropagation algorithm described in ... |

58 |
Upwind methods for hyperbolic conservation laws with source terms. Computers Fluids
- Bermudez, Vazquez
- 1994
(Show Context)
Citation Context ...consists of a conservation law with a source term on the right-hand side. Many approaches have been studied for equations of this form, primarily in the case of an autonomous flux function f(q)(e.g., =-=[4, 8, 11, 13, 14, 28, 30]-=-). One simple approach that is often used is the fractional step method, in which one alternates between solving the homogeneous equation (1.1)and the ordinary differential equation (7.2) qt = ψ(q, x)... |

58 |
Convex conservation laws with discontinuous coecients. Existence, uniqueness and asymptotic behavior
- Klingenberg, Risebro
- 1995
(Show Context)
Citation Context ...mpose appropriate admissibility conditions to select the physically correct solution, and this can be more subtle than in the autonomous case. For some discussions of such problems, see, for example, =-=[5, 6, 9, 10, 12, 15, 17, 25, 26]-=-. We are continuing to study extensions of our algorithm to particular problems of this form. 3. The wave-propagation algorithms. We first briefly summarize the wavepropagation algorithm described in ... |

45 |
Solution of the Cauchy problem for a conservation law with a discontinuous function
- Gimse, Risebro
- 1992
(Show Context)
Citation Context ...mpose appropriate admissibility conditions to select the physically correct solution, and this can be more subtle than in the autonomous case. For some discussions of such problems, see, for example, =-=[5, 6, 9, 10, 12, 15, 17, 25, 26]-=-. We are continuing to study extensions of our algorithm to particular problems of this form. 3. The wave-propagation algorithms. We first briefly summarize the wavepropagation algorithm described in ... |

44 |
A conservation law with point source and discontinuous flux function modeling continuous sedimentation
- Diehl
- 1996
(Show Context)
Citation Context |

42 |
Measure solutions to the linear multi-dimensional transport equation with non-smooth coefficients
- Poupaud, Rascle
- 1997
(Show Context)
Citation Context |

37 |
A well-balanced flux-vector splitting scheme designed for hyperbolic systems of conservation laws with source terms
- Gosse
(Show Context)
Citation Context ...consists of a conservation law with a source term on the right-hand side. Many approaches have been studied for equations of this form, primarily in the case of an autonomous flux function f(q)(e.g., =-=[4, 8, 11, 13, 14, 28, 30]-=-). One simple approach that is often used is the fractional step method, in which one alternates between solving the homogeneous equation (1.1)and the ordinary differential equation (7.2) qt = ψ(q, x)... |

30 |
Riemann problems with a discontinuous flux function
- Gimse, Risebro
- 1990
(Show Context)
Citation Context |

27 | A wave propagation method for three-dimensional hyperbolic conservation laws
- Langseth, LeVeque
(Show Context)
Citation Context ...r dimensions. The wave-propagation algorithm developed in section 3 is easily extended to multidimensional conservation laws following the same procedure used in [20] for the standard algorithm. (See =-=[16]-=- for the threedimensional extension.)In two space dimensions a conservation law with spatially varying fluxes takes the form (6.1) qt + f(q, x, y)x + g(q, x, y)y =0, where q = q(x, y, t). The two-dime... |

27 | and Marica Pelanti. A class of approximate Riemann solvers and their relation to relaxation schemes
- LeVeque
(Show Context)
Citation Context ...ying fluxes this is a more natural decomposition for reasons discussed in section 2. This decomposition can also be related to a generalized relaxation scheme for the conservation law as discussed in =-=[23]-=-. The vectors Z p = β p r p will be called f-waves, as they are analogous to the waves W p from (1.10)but carry flux increments rather than increments in q. A potential advantage of using the decompos... |

27 |
Upwind differencing schemes for hyperbolic conservation laws with source terms
- Roe
- 1986
(Show Context)
Citation Context ...consists of a conservation law with a source term on the right-hand side. Many approaches have been studied for equations of this form, primarily in the case of an autonomous flux function f(q)(e.g., =-=[4, 8, 11, 13, 14, 28, 30]-=-). One simple approach that is often used is the fractional step method, in which one alternates between solving the homogeneous equation (1.1)and the ordinary differential equation (7.2) qt = ψ(q, x)... |

21 | Numerical approximations of one-dimensional linear conservation equations with discontinous coefficients
- Gosse, James
- 2000
(Show Context)
Citation Context |

21 |
Analysis and approximation of conservation laws with source terms
- Greenberg, LeRoux, et al.
- 1997
(Show Context)
Citation Context |

18 |
Numerical tests of evolution systems, gauge conditions, and boundary conditions for 1d colliding gravitational plane waves
- Bardeen, Buchman
(Show Context)
Citation Context ...rying flux functions, an application that is considered in a separate paper [2] and the recent theses [1] and [29]. Application to a problem in gravitational waves is discussed by Bardeen and Buchman =-=[3]-=-. Spatially varying flux functions lead to difficulties not seen in the autonomous case, some of which are discussed below. In this paper we concentrate on the case in which the eigenvalues of the flu... |

11 | High-resolution finite volume methods for acoustics in periodic or random
- Fogarty, LeVeque
- 1999
(Show Context)
Citation Context ...e wave-propagation algorithms have been extended to apply to nonconservative hyperbolic systems in [20], where the acoustics system (5.4)is used as an example. These algorithms are further studied in =-=[7]-=-, where it is observed that good results are obtained even when the coefficients K(x)and ρ(x)are rapidly varying and/or discontinuous. However, the conservative form based on (4.1)has advantages, at l... |

8 | Finite volume methods for nonlinear elasticity in heterogeneous
- LeVeque
- 2001
(Show Context)
Citation Context ...derations, since the components of flux are −u and −σ. Clearly the velocity and stress must be continuous at an interface for the continuum model to hold. This Riemann problem is discussed further in =-=[22]-=-, where a description of an approach to computing the exact solution is described. However, an approximate Riemann solver is found to work very well, based on defining the matrix Ai−1/2 in terms of it... |

7 |
Rankine-Hugoniot-Riemann solver considering source terms and multidimensional eects
- Jenny, Muller
- 1998
(Show Context)
Citation Context |

7 | Solitary waves in layered nonlinear media
- LeVeque, Yong
- 2002
(Show Context)
Citation Context ...se constant properties (a layered medium or laminate). An interesting nonlinear effect was found in this case: a smooth pulse breaks up into solitary waves that appear to behave as solitons; see also =-=[24]-=-. Here we use this and a related example with smoothly varying material properties as a test of the accuracy of the numerical method. We take (4.8) for some function φ(x), with initial data (4.9) and ... |

4 |
Explicit formula for weighted scalar nonlinear conservation laws
- LeFloch, Nedelec
- 1988
(Show Context)
Citation Context |

3 |
Wave propagation algorithms on curved manifolds with applications to relativistic hydrodynamics
- Bale
- 2002
(Show Context)
Citation Context ...terogeneous porous media. Solving conservation laws on curved manifolds also leads to spatially varying flux functions, an application that is considered in a separate paper [2] and the recent theses =-=[1]-=- and [29]. Application to a problem in gravitational waves is discussed by Bardeen and Buchman [3]. Spatially varying flux functions lead to difficulties not seen in the autonomous case, some of which... |

3 |
Conservation laws with sharp inhomogeneities
- Lyons
- 1982
(Show Context)
Citation Context |

1 |
Corberán, Construction of second-order TVD schemes for nonhomogeneous hyperbolic conservation laws
- M
(Show Context)
Citation Context |

1 |
Clawpack software, online at http://www.amath.washington.edu/˜claw
- LeVeque
(Show Context)
Citation Context .../2 . p=1 Now � Zp is a limited version of the f-wave Zp obtained in the same manner as � Wp would be obtained from Wp . The standard wave-propagation algorithm is implemented in the Clawpack software =-=[18]-=-, and this can be applied to the problems considered here by defining Wp = Zp /sp , since we assume sp �= 0. However, the modification just described appears more robust since we do not need to worry ... |