Novel Adaptive Schemes for Hyperbolic Conservation Laws
Shaoshuai Chu, Pingyao Feng, Vadim A. Kolotilov, Alexander Kurganov, and Vladimir V. Ostapenko · 2025
Read Novel Adaptive Schemes for Hyperbolic Conservation Laws by Shaoshuai Chu, Pingyao Feng, Vadim A. Kolotilov, Alexander Kurganov, and Vladimir V. Ostapenko (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-2509-18908.txt.
Introduction
Shaoshuai Chu and four co-authors design new numerical schemes for equations that govern flows with sharp jumps, such as shock waves. A single fixed method struggles: sharp methods can add fake wiggles near jumps, while safe methods can blur fine detail. Their schemes switch methods by region, guided by a smoothness indicator that flags rough parts of the solution.
In rough zones the scheme uses a second-order method that keeps shocks and contact edges crisp. In smooth zones it switches to a fifth-order method for better accuracy. The indicator also tells contact edges apart from other rough spots, so each gets the limiter that fits it best. Tests on hard example problems show the combined scheme stays both sharp and clean.
Catalog record
| CREATOR | Shaoshuai Chu, Pingyao Feng, Vadim A. Kolotilov, Alexander Kurganov, and Vladimir V. Ostapenko |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 62 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-18908
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Novel Adaptive Schemes for Hyperbolic Conservation Laws Shaoshuai Chu, Pingyao Feng, Vadim A. Kolotilov, Alexander Kurganov, and Vladimir V. Ostapenko Abstract We introduce new adaptive schemes for the one- and two-dimensional hyperbolic systems of conservation laws. Our schemes are based on an adaption strategy recently introduced in [ S. Chu, A. Kurganov, and I. Menshov , Appl. Numer. Math., 209 (2025)]. As there, we use a smoothness indicator (SI) to automatically detect “rough” parts of the solution and employ in those areas the second-order finite-volume low-dissipation central-upwind scheme with an overcompressive limiter, which helps to sharply resolve nonlinear shock waves and linearly degenerate contact discontinuities. In smooth parts, we replace the limited second-order scheme with a quasi-linear fifth-order (in space and third-order in time) finite-difference scheme, recently proposed in [ V. A. Kolotilov, V. V. Ostapenko, and N. A. Khandeeva , Comput. Math. Math. Phys., 65 (2025)]. However, direct application of this scheme may generate spurious oscillations near “rough” parts, while excessive use of the overcompressive limiter may cause staircase-like nonphysical structures in smooth areas. To address these issues, we employ the same SI to distinguish contact discontinuities, treated with the overcompressive limiter, from other “rough” regions, where we switch to the dissipative Minmod2 limiter. Advantage of the resulting adaptive schemes are clearly demonstrated on a number of challenging numerical examples. Key words: Adaptive schemes; low-dissipation central-upwind numerical fluxes; smoothness indicator; overcompressive and dissipative limiters; quasi-linear fifth-order finite-difference scheme; Euler equations of gas dynamics. AMS subject classification: 65M08, 65M06, 76M12, 76M20, 76L05, 35L65. 1 Introduction We consider hyperbolic systems of conservation laws, which in the one-dimensional (1-D) and two-dimensional (2-D) cases, read as \bm{U}_{t}+\bm{F}(\bm{U})_{x}=\bm{0}, (1.1) and \bm{U}_{t}+\bm{F}(\bm{U})_{x}+\bm{G}(\bm{U})_{y}=\bm{0}, (1.2) respectively. Here, x and y are spatial variables, t is the time, \bm{U}\in\mathbb{R}^{d} is a vector of unknown functions, and \bm{F}:\mathbb{R}^{d}\to\mathbb{R}^{d} and \bm{G}:\mathbb{R}^{d}\to\mathbb{R}^{d} are fluxes. It is well-known that solutions of ( 1.1 ) and ( 1.2 ) can produce extremely complex wave structures including shocks, rarefactions, and contact discontinuities even when the initial data are smooth. This makes it quite challenging to develop accurate and reliable numerical methods for ( 1.1 ) and ( 1.2 ). Such methods are typically based on equivalent integral forms of ( 1.1 ) and ( 1.2 ) for the time evolution of the computed solution and on nonlinear limiters, designed to prevent spurious oscillations. For a variety of existing numerical methods for hyperbolic systems of conservation laws, we refer the reader to the monographs and review papers [ 17 , 42 , 27 , 21 , 2 , 39 ] and references therein. In this paper, we focus on scheme adaption methods, which are based on automatic detection of smooth and nonsmooth (“rough”) parts of the computed solutions and on applying different numerical methods in the corresponding parts of the computational domain. The main advantage of such an approach is related to the fact that special nonlinear limiting techniques can be used to accurately capture shock waves and contact discontinuities only, while smooth parts of the solution may be accurately and efficiently computed by much simpler, possibly higher-order, (quasi-linear) unlimited methods. A success of adaptive schemes hinges on an accuracy and robustness of the smoothness indicator (SI) used to detect “rough” parts of the computed solution. A wide variety of different SIs has been developed; see, e.g., [ 12 , 11 , 1 , 14 , 16 , 15 , 20 , 31 , 33 , 43 , 13 , 44 , 22 , 30 , 45 , 6 ] . Another critical component of a successful adaptive scheme is a very sharp method used to capture shock and contact discontinuities inside the identified “rough” parts of the solution. In this paper, we follow [ 8 ] and implement the SI proposed in [ 30 ] along with the second-order semi-discrete finite-volume low-dissipation central-upwind (LDCU) schemes from [ 10 ] combined with the overcompressive and dissipative limiters introduced in [ 28 ] . The novelty of the proposed adaptive schemes is twofold. First, we treat the smooth parts of the computed solutions in a different way. Instead of using the second-order LDCU scheme with the dissipative Minmod2 limiter (see [ 28 ] for the classification of different limiters) as it was done in [ 8 ] , we increase the resolution by switching to the recently proposed quasi-linear fifth-order (in space and third-order in time) finite-difference scheme recently introduced in [ 24 ] . Following [ 26 ] , we call a numerical scheme quasi-linear if it becomes linear when applied to linear hyperbolic systems. As it was shown in [ 24 ] , the fifth-order quasi-linear scheme is very accurate and like the third-order Rusanov-Burstein-Mirin scheme [ 4 , 34 ] , it achieves increased accuracy in shock influence regions. This scheme, however, does not contain any nonlinear stabilization mechanism and thus cannot be used in the “rough” areas. Moreover, as we show in § 2.3 , it can produce spurious oscillations near the interfaces between the “rough” and “smooth” regions. To suppress these oscillations, one can “enlarge” the identified “rough” areas, but in this case, the overcompressive limiter would affect the smooth parts of the solution causing development of artificial kinks and staircase-like structures or even nonphysical jump discontinuities there. In order to address these issues, we introduce an alternative approach, which is the second novel component of the proposed adaptive schemes. To this end, we use the same SI, but applied to different fields to automatically distinguish between the contact waves and other “rough” parts of the solution. For example, in the case of Euler equations of gas dynamics, we take advantage of the fact that while the density jumps at the contact surfaces, the pressure remains continuous there. After identifying the locations of contact surfaces, we apply the second-order LDCU scheme with an overcompressive limiter in their vicinities only, while utilizing the dissipative Minmod2 limiter in the rest of the “rough” areas. In addition, the overcompressive and Minmod2 limiters are applied to the local characteristic variables using the local characteristic decomposition (LCD), which is often used in the context of high-order schemes [ 32 ] to suppress spurious oscillations, but can also be implemented to enhance the resolution of second-order schemes; see, e.g., [ 9 , 7 ] . This way, we construct efficient and robust adaptive schemes, which achieve a very high resolution of both jump discontinuities and smooth parts of the computed solutions. The paper is organized as follows. In § 2 , we review the 1-D LDCU scheme from [ 10 ] , the fifth-order finite-difference scheme from [ 24 ] , and then introduce a 1-D adaptive scheme based on the scheme adaption strategy from [ 8 ] . We also demonstrate that the obtained adaptive scheme is not robust and thus it has to be modified. In § 3 , we present a new 1-D adaptive scheme, which is based on the dynamical splitting of the computational domains into three parts: vicinities of the contact discontinuities, other “rough” parts of the computed solutions, and smooth parts of the computed solutions. In § 4 , we extend the new adaptive scheme to the 2-D case. In §