Skip to main content
Log in

Entropy Stable Schemes For Ten-Moment Gaussian Closure Equations

  • Published:
Journal of Scientific Computing Aims and scope Submit manuscript

Abstract

In this article, we propose high order, semi-discrete, entropy stable, finite difference schemes for Ten-Moment Gaussian Closure equations. The crucial components of these schemes are an entropy conservative flux and suitable high order entropy dissipative operators to ensure entropy stability. We design two numerical fluxes, one is approximately entropy conservative, and another is entropy conservative flux. For the construction of appropriate entropy dissipative operators, we also derive entropy scaled right eigenvectors. This is used for sign preserving reconstruction of scaled entropy variables, which results in second and third order entropy stable schemes. We also extend these schemes to a plasma flow model with source term. Several numerical results are presented for homogeneous and non-homogeneous cases to demonstrate stability and performance of these schemes.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

References

  1. Barth, T.J.: Numerical methods for gas-dynamics systems on unstructured meshes. In: Kroner, D., Ohlberger, M., Rohde, C. (eds.) An Introduction to Recent Developments in Theory and Numerics of Conservation Laws. Lecture Notes in Computational Science and Engineering, vol. 5, pp. 195–285. Springer, Berlin (1999)

    Chapter  Google Scholar 

  2. Berthon, C.: Numerical approximations of the 10-moment Gaussian-closure. Math. Comput. 75(256), 1809–1831 (2006)

    Article  MathSciNet  MATH  Google Scholar 

  3. Berthon, C., Desveaux, V.: An entropy preserving MOOD scheme for the Euler equations. Int. J. Finite 11, 1–39 (2014)

    MathSciNet  Google Scholar 

  4. Berthon, C., Dubroca, B., Sangam, A.: An entropy preserving relaxation scheme for the Ten-Moments equations with source terms. Commun. Math. Sci. 13(8), 2119–2154 (2015)

    Article  MathSciNet  MATH  Google Scholar 

  5. Chandrashekar, P.: Kinetic energy preserving and entropy stable finite volume schemes for compressible euler and Navier–Stokes equations. Commun. Comput. Phys. 14, 1252–1286 (2013)

    Article  MathSciNet  MATH  Google Scholar 

  6. Brown, S.L.,Roe, P.L., Groth, C.P.: Numerical solution of a 10-moment model for nonequilibrium gasdynamics. In: 12th AIAA Computational Fluid Dynamics Conference (1995)

  7. Dubroca, B., Tchong, M., Charrier, P., Tikhonchuk, V.T., Morreeuw, J.P.: Magnetic field generation in plasmas due to anisotropic laser heating. Phys. Plasm. 11, 3830–3839 (2004)

    Article  Google Scholar 

  8. Farzad Ismail, R., Philip, L.: Affordable, entropy-consistent euler flux functions II: entropy production at shocks. J. Comput. Phys. 228(15), 5410–5436 (2009)

    Article  MathSciNet  MATH  Google Scholar 

  9. Fjordholm, U.S., Mishra, S., Tadmor, E.: Arbitrarily high-order accurate entropy stable essentially nonoscillatory schemes for systems of conservation laws. SIAM J. Numer. Anal. 50(2), 544–573 (2012)

    Article  MathSciNet  MATH  Google Scholar 

  10. Fjordholm, U.S., Mishra, S., Tadmor, E.: ENO reconstruction and ENO interpolation are stable. Found. Comput. Math. 13(2), 139–159 (2012)

    Article  MathSciNet  MATH  Google Scholar 

  11. Fjordholm, U.S.: High-order accurate entropy stable numerical schemes for hyperbolic conservation laws. Diss. No. 21025, Ph.D. Thesis, ETH Zurich (2013)

  12. Gottlieb, S., Shu, C.W., Tadmor, E.: Strong stability-preserving high-order time discretization methods. SIAM Rev. 43, 89–112 (2001)

    Article  MathSciNet  MATH  Google Scholar 

  13. Hakim, A.: Extended MHD modelling with ten-moment equations. J. Fusion Energy 27, 36–43 (2008)

    Article  Google Scholar 

  14. Johnson, E.A.: Gaussian-Moment relaxation closures for verifiable numerical simulation of fast magnetic reconnection in plasma. Ph.D. Thesis, UW-Madison (2011)

  15. LeFloch, P.G., Mercier, J.M., Rohde, C.: Fully discrete entropy conservative schemes of arbitrary order. SIAM J. Numer. Anal. 40(5), 1968–1992 (2002)

    Article  MathSciNet  MATH  Google Scholar 

  16. Levermore, C.D.: Moment closure hierarchies for kinetic theories. J. Stat. Phys. 83, 1021–1065 (1996)

    Article  MathSciNet  MATH  Google Scholar 

  17. Levermore, C.D., Morokoff, W.J.: The Gaussian moment closure for gas dynamics. SIAM J. Appl. Math. 59(1), 72–96 (1996)

    Article  MathSciNet  MATH  Google Scholar 

  18. Meena, A.K., Kumar, H.: A robust MUSCL scheme for ten-moment Gaussian closure equations with source terms. Submitted

  19. Meena, A.K., Kumar, H., Chnadrashekar, P.: Positivity-preserving high-order discontinuous Galerkin schemes for ten-moment Gaussian closure equations. J. Comput. Phys. 339, 370–395 (2017)

    Article  MathSciNet  MATH  Google Scholar 

  20. Ray, D., Chandrashekar, P., Fjordholm, U.S., Mishra, S.: Entropy stable scheme on two-dimensional unstructured grids for Euler equations. Commun. Comput. Phys. 19, 1111–1140 (2016)

    Article  MathSciNet  MATH  Google Scholar 

  21. Sangam, A., Morreeuw, J.P., Tikhonchuk, V.T.: Anisotropic instability in a laser heated plasma. Phys. Plasm. 14, 053111 (2007)

    Article  Google Scholar 

  22. Sangam, A.: An HLLC scheme for ten-moments approximation coupled with magnetic field. Int. J. Comput. Sci. Math. 2, 73–109 (2008)

    Article  MathSciNet  MATH  Google Scholar 

  23. Shu, C.W.: TVD time discretizations. SIAM J. Math. Anal. 14, 1073–1084 (1988)

    MATH  Google Scholar 

  24. Struchtrup, H., Torrilhon, M.: Regularization of grad’s 13-moment-equations: derivation and linear analysis. Phys. Fluids 15(9), 2668–2680 (2003)

    Article  MathSciNet  MATH  Google Scholar 

  25. Struchtrup, H.: Macroscopic Transport Equations for Rarefied Gas Flows. Springer, Berlin (2005)

    Book  MATH  Google Scholar 

  26. Tadmor, E.: The numerical viscosity of entropy stable schemes for systems of conservation laws. I. Math. Comp. 49, 91–103 (1987)

    Article  MathSciNet  MATH  Google Scholar 

  27. Tadmor, E.: Entropy stability theory for difference approximations of nonlinear conservation laws and related time-dependent problems. Act. Numer. 12, 451–512 (2004)

    Article  MathSciNet  MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Chhanda Sen.

Additional information

Ms. Chhanda Sen is supported by CSIR Fellowship at Dept. of Mathematics, IIT Delhi. Harish Kumar has been funded in part by SERB, DST Grant with file No. YSS/2015/001663.

Appendix: Symmetrization and Entropy Scaled Eigenvectors for Ten-Moment Equations

Appendix: Symmetrization and Entropy Scaled Eigenvectors for Ten-Moment Equations

In this section we will derive the entropy scaled right eigenvectors. For simplicity, let us consider the x-directional homogenous part of the system (1), i.e.

$$\begin{aligned} \frac{\partial \mathbf{u }}{\partial t}+\frac{\partial \,\mathbf f ^x(\mathbf{u })}{\partial x}=0. \end{aligned}$$
(26)

The entropy-entropy flux pair for this system is same as for the complete system. So, the results presented here can be easily extended to the complete system. Let us recall that,

$$\begin{aligned} \mathbf{v }\,=\,\frac{\partial e}{\partial \mathbf{u }}\,=\,\left( \begin{array}{c}4-s-\rho \frac{(p^{xx}v^{y2}\,+\,p^{yy}v^{x2}\,-\,2p^{xy}v^xv^y)}{p^{xx}p^{yy}\,-\,p^{xy2}}\\ \frac{2\rho (p^{yy}v^x\,-\,p^{xy}v^y)}{p^{xx}p^{yy}\,-\,p^{xy2}}\\ \frac{2\rho (p^{xx}v^y\,-\,p^{xy}v^x)}{p^{xx}p^{yy}\,-\,p^{xy2}}\\ \frac{-\rho p^{yy}}{p^{xx}p^{yy}\,-\,p^{xy2}}\\ \frac{2\rho p^{xy}}{p^{xx}p^{yy}\,-\,p^{xy2}}\\ \frac{-\rho p^{xx}}{p^{xx}p^{yy}\,-\,p^{xy2}} \end{array}\right) . \end{aligned}$$

and

$$\begin{aligned} {\psi }^x\,=\,\mathbf{v }^{\top }\cdot \mathbf f ^x\,-\,q^x\,=\,2\rho v^x. \end{aligned}$$

Theorem A.1

(Barth [1]) (Eigenvector scaling)

Let \(A\in \mathbb {R}^{n\times n}\) be an arbitrary diagonalizable matrix and S the set of all right symmetrizers:

$$\begin{aligned} S=\{B\in \mathbb {R}^{n\times n}| B\quad SPD,\quad AB\,\, \text{ symmetric }\}. \end{aligned}$$

Further, let \(\mathbf{R}\in \mathbb {R}^{n\times n}\) denote the right eigenvector matrix which diagonalizes A, \(A=\mathbf{R}\Lambda \mathbf{R}^{-1}\) with r distinct eigenvalues, \(\Lambda =Diag(\lambda _1I_{m_1\times m_1}, \lambda _2I_{m_2\times m_2}, \ldots , \lambda _rI_{m_r\times m_r}).\) Then for each \(B\in S\) there exists a symmetric block diagonal matrix \(T=Diag(T_{m_1\times m_1}, T_{m_2\times m_2},\ldots T_{m_r\times m_r} )\), such that the block scales columns of \(\mathbf{R}\), \(\tilde{\mathbf{R}}=\mathbf{R}T\), with

$$\begin{aligned} B=\tilde{\mathbf{R}}\tilde{\mathbf{R}}^{\top },\quad A=\tilde{\mathbf{R}}\Lambda \tilde{\mathbf{R}}^{-1} \end{aligned}$$

which imply

$$\begin{aligned} AB=\tilde{\mathbf{R}}\Lambda \tilde{\mathbf{R}}^{\top }. \end{aligned}$$

By change of variable from \(\mathbf{u}\) to \(\mathbf{v}\) we can rewrite (26) as,

$$\begin{aligned} \frac{\partial \mathbf{u }}{\partial \mathbf{v }}\frac{\partial \mathbf{v }}{\partial t}+\frac{\partial \mathbf f ^x(\mathbf{u })}{\partial \mathbf{u }}\frac{\partial \mathbf{u }}{\partial \mathbf{v }}\frac{\partial \mathbf{v }}{\partial x}=\mathbf 0 . \end{aligned}$$
(27)

As we have \((e,q_{x})\) as entropy pair for the system (26), the above change of variable should symmetrize the system (see [1]). Furthermore, the matrix \(\frac{\partial \mathbf{u }}{\partial \mathbf{v }}\) is symmetric positive definite. We now want to find scaled entropy right eigenvectors \(\tilde{\mathbf{R}}\) such that,

$$\begin{aligned} \frac{\partial \mathbf{u }}{\partial \mathbf{v }}=\tilde{\mathbf{R}}\tilde{\mathbf{R}}^{\top }. \end{aligned}$$
(28)

To simplify the calculation of the entropy scaling, let us introduce primitive variables \(\mathbf{w}=\{\rho , \vec {v},\mathbf{p}\}^{\top }.\) Rewriting the system (26) in primitive variables, we get eigenvalues as,

$$\begin{aligned} v^x-\sqrt{\frac{3p^{xx}}{\rho }},\,v^x-\sqrt{\frac{p^{xx}}{\rho }},\,v^x,\,v^x,\,v^x+\sqrt{\frac{p^{xx}}{\rho }},\,v^x+\sqrt{\frac{3p^{xx}}{\rho }}, \end{aligned}$$

and corresponding right eigenvector matrix as,

$$\begin{aligned} \mathbf{R}_{\mathbf{w}}^x= \left( \begin{array}{c c c c c c} \rho p^{xx} &{}\quad 0 &{}\quad 1 &{}\quad 0 &{}\quad 0 &{}\quad \rho p^{xx}\\ -\sqrt{\frac{3p^{xx}}{\rho }}p^{xx} &{}\quad 0 &{}\quad 0 &{}\quad 0 &{}\quad 0 &{}\quad \sqrt{\frac{3p^{xx}}{\rho }}p^{xx}\\ -\sqrt{\frac{3p^{xx}}{\rho }}p^{xy} &{}\quad -\,\sqrt{\frac{p^{xx}}{\rho }} &{}\quad 0 &{}\quad 0 &{}\quad \sqrt{\frac{p^{xx}}{\rho }} &{} \sqrt{\frac{3p^{xx}}{\rho }}p^{xy} \\ 3p^{xx2} &{}\quad 0 &{}\quad 0 &{}\quad 0 &{}\quad 0 &{}\quad 3p^{xx2}\\ 3p^{xx}p^{xy} &{}\quad p^{xx} &{}\quad 0 &{}\quad 0 &{}\quad p^{xx} &{}\quad 3p^{xx}p^{xy}\\ 2p^{xy2}+p^{xx}p^{yy} &{}\quad 2p^{xy} &{}\quad 0 &{}\quad 1 &{}\quad 2p^{xy} &{}\quad 2p^{xy2}+p^{xx}p^{yy} \end{array}\right) . \end{aligned}$$

Once the entropy scaled right eigenvector for primitive variable system, \(\tilde{\mathbf{R}}_{\mathbf{w}}^x\) is found, entropy scaled right eigenvectors for the conservative system can be easily calculated using,

$$\begin{aligned} \tilde{\mathbf{R}}^x=\frac{\partial \mathbf{u }}{\partial \mathbf{w }}\,\tilde{\mathbf{R}}_{\mathbf{w}}^x. \end{aligned}$$
(29)

Here, the jacobian matrix for the change of variable is,

$$\begin{aligned} \frac{\partial \mathbf{u }}{\partial \mathbf{w }}\,=\,\left( \begin{array}{c c c c c c} 1 &{}\quad 0 &{}\quad 0 &{}\quad 0 &{}\quad 0 &{}\quad 0\\ v^x &{}\quad \rho &{}\quad 0 &{}\quad 0 &{}\quad 0 &{}\quad 0\\ v^y &{}\quad 0 &{}\quad \rho &{}\quad 0 &{}\quad 0 &{}\quad 0\\ v^{x2} &{}\quad 2\rho v^x &{}\quad 0 &{}\quad 1 &{}\quad 0 &{}\quad 0\\ v^xv^y &{}\quad \rho v^y &{}\quad \rho v^x &{}\quad 0 &{}\quad 1 &{}\quad 0\\ v^{y2} &{}\quad 0 &{}\quad 2\rho v^y &{}\quad 0 &{}\quad 0 &{}\quad 1 \end{array}\right) . \end{aligned}$$

Using Eqs.(28) and (29), we note that,

$$\begin{aligned} \frac{\partial \mathbf{u}}{\partial \mathbf{v}}= \frac{\partial \mathbf{u}}{\partial \mathbf{w}}\frac{\partial \mathbf{w}}{\partial \mathbf{v}}=\tilde{\mathbf{R}}^x \tilde{\mathbf{R}}^{x,\top }= \frac{\partial \mathbf{u }}{\partial \mathbf{w }}\,\tilde{\mathbf{R}}_{\mathbf{w}}^x \tilde{\mathbf{R}}_{\mathbf{w}}^{x,\top }{\frac{\partial \mathbf{u }}{\partial \mathbf{w }}}^{\top } \end{aligned}$$

which implies,

$$\begin{aligned} \tilde{\mathbf{R}}_{\mathbf{w}}^x \tilde{\mathbf{R}}_{\mathbf{w}}^{x,\top }= \frac{\partial \mathbf{w }}{\partial \mathbf{v }}\frac{\partial \mathbf{u }}{\partial \mathbf{w }}^{(-\top )}. \end{aligned}$$
(30)

To calculate the matrix \(\frac{\partial \mathbf{w }}{\partial \mathbf{v }}\) we need to write primitive variables in terms of entropy variables \(\mathbf{v}\). Let us define the notations,

figure b

Plugging in the expression for \(v_{4},v_{5}\) and \(v_{6}\) into the equation for \(v_{2}\) and \(v_{3}\) we get,

$$\begin{aligned} v_2= & {} -2v_4v^x-v_5v^y,\\ v_3= & {} -2v_6v^y-v_5v^x. \end{aligned}$$

On solving for \(v^{x}\) and \(v^{y}\), we get,

$$\begin{aligned} v^x= & {} \frac{v_5v_3-2v_2v_6}{4v_4v_6-v^2_5},\end{aligned}$$
(31)
$$\begin{aligned} v^y= & {} \frac{v_2v_5-2v_3v_4}{4v_4v_6-v^2_5}. \end{aligned}$$
(32)

We now rewrite \(v_{1}\) as,

$$\begin{aligned} v_1= & {} 4-s+(v_6v^{y2}+v_4v^{x2}+v_5v^xv^y)\\= & {} 4-\log (4)+\log (4v_4v_6-v^2_5)+2\log (\rho )+(v_6v^{y2}+v_4v^{x2}+v_5v^xv^y), \end{aligned}$$

where we have used

$$\begin{aligned} \log (p^{xx}p^{yy}-p^{xy2})= & {} \log (\rho ^2)+\log (4)-\log (4v_4v_6-v^2_5). \end{aligned}$$

This results in,

$$\begin{aligned} \rho =e^{(v_1-4+\log (4)-\log (4v_4v_6-v^2_5)-(v_6v^{y2}+v_4v^{x2}+v_5v^xv^y))/2}. \end{aligned}$$
(33)

Now using expression for \(v_{4}\),

$$\begin{aligned} p^{yy}=-\frac{4v_4\rho }{4v_4v_6-v^2_5},\nonumber= & {} -\frac{4v_4e^{(v_1-4+\log (4)-\log (4v_4v_6-v^2_5)-(v_6v^{y2}+v_4v^{x2}+v_5v^xv^y))/2}}{4v_4v_6-v^2_5}.\\ \end{aligned}$$
(34)

Similarly,

$$\begin{aligned} p^{xy}= & {} \frac{2v_5e^{(v_1-4+\log (4)-\log (4v_4v_6-v^2_5)-(v_6v^{y2}+v_4v^{x2}+v_5v^xv^y))/2}}{4v_4v_6-v^2_5},\end{aligned}$$
(35)
$$\begin{aligned} p^{xx}= & {} -\frac{4v_6e^{(v_1-4+\log (4)-\log (4v_4v_6-v^2_5)-(v_6v^{y2}+v_4v^{x2}+v_5v^xv^y))/2}}{4v_4v_6-v^2_5}. \end{aligned}$$
(36)

The matrix \(\frac{\partial \mathbf{w }}{\partial \mathbf{v }}\) can now be easily calculated and is found to be,

$$\begin{aligned} \left( \begin{array}{c c c c c c} \frac{\rho }{2} &{}\quad \frac{\rho v^x}{2} &{}\quad \frac{\rho v^y}{2} &{}\quad \frac{e^{xx}}{2} &{}\quad \frac{e^{xy}}{2} &{} \frac{e^{yy}}{2}\\ &{}\quad &{}\quad &{} \quad &{}\quad &{}\quad \\ 0 &{}\quad \frac{p^{xx}}{2\rho } &{}\quad \frac{p^{xy}}{2\rho } &{}\quad \frac{p^{xx}v^x}{\rho } &{}\quad \frac{p^{xy}v^x+p^{xx}v^y}{2\rho } &{}\quad \frac{p^{xy}v^y}{\rho }\\ &{}\quad &{}\quad &{}\quad &{}\quad &{}\\ 0 &{}\quad \frac{p^{xy}}{2\rho } &{}\quad \frac{p^{yy}}{2\rho } &{}\quad \frac{p^{xy}v^x}{\rho } &{}\quad \frac{p^{yy}v^x+p^{xy}v^y}{2\rho } &{}\quad \frac{p^{yy}v^y}{\rho }\\ &{}\quad &{}\quad &{}\quad &{}\quad &{}\\ \frac{p^{xx}}{2} &{}\quad \frac{p^{xx}v^x}{2} &{}\quad \frac{p^{xx}v^y}{2} &{}\quad \frac{3 p^{xx2}}{2\rho } +\frac{p^{xx}v^{x2}}{2} &{}\quad \frac{3 p^{xx}p^{xy}}{2\rho } +\frac{p^{xx}v^{x}v^y}{2} &{}\quad \frac{p^{xy2}}{\rho }+\frac{p^{xx}p^{yy}}{2\rho }+\frac{p^{xx}v^{y2}}{2}\\ &{}\quad &{}\quad &{}\quad &{}\quad &{}\\ \frac{p^{xy}}{2} &{}\quad \frac{p^{xy}v^x}{2} &{}\quad \frac{p^{xy}v^y}{2} &{}\quad \frac{3 p^{xx}p^{xy}}{2\rho } +\frac{p^{xy}v^{x2}}{2} &{}\quad \frac{p^{xy2}}{\rho }+\frac{p^{xx}p^{yy}}{2\rho }+\frac{p^{xy}v^{x}v^y}{2} &{} \frac{3 p^{xy}p^{yy}}{2\rho } +\frac{p^{xy}v^{y2}}{2} \\ &{}\quad &{}\quad &{}\quad &{}\quad &{}\\ \frac{p^{yy}}{2} &{}\quad \frac{p^{yy}v^x}{2} &{}\quad \frac{p^{yy}v^y}{2} &{}\quad \frac{p^{xy2}}{\rho }+\frac{p^{xx}p^{yy}}{2\rho }+\frac{p^{yy}v^{x2}}{2} &{}\quad \frac{3 p^{xy}p^{yy}}{2\rho } +\frac{p^{yy}v^xv^y}{2} &{}\quad \frac{3 p^{yy2}}{2\rho } +\frac{p^{yy}v^{y2}}{2}\\ \end{array}\right) . \end{aligned}$$

Furthermore, the matrix \(\frac{\partial \mathbf{w }}{\partial \mathbf{v }}\frac{\partial \mathbf{u }}{\partial \mathbf{w }}^{-\top }\) is,

$$\begin{aligned} \left( \begin{array}{c c c c c c} \frac{\rho }{2} &{} 0 &{} 0 &{} \frac{p^{xx}}{2} &{} \frac{p^{xy}}{2} &{} \frac{p^{yy}}{2}\\ &{} &{} &{} &{} &{}\\ 0 &{} \frac{p^{xx}}{2\rho ^2} &{} \frac{p^{xy}}{2\rho ^2} &{} 0 &{} 0 &{} 0\\ &{} &{} &{} &{} &{}\\ 0 &{} \frac{p^{xy}}{2\rho ^2} &{} \frac{p^{yy}}{2\rho ^2} &{} 0 &{} 0 &{} 0\\ &{} &{} &{} &{} &{}\\ \frac{p^{xx}}{2} &{} 0 &{} 0 &{} \frac{3 p^{xx2}}{2\rho } &{} \frac{3 p^{xx}p^{xy}}{2\rho } &{} \frac{p^{xy2}}{\rho }+\frac{p^{xx}p^{yy}}{2\rho }\\ &{} &{} &{} &{} &{}\\ \frac{p^{xy}}{2} &{} 0 &{} 0 &{} \frac{3 p^{xx}p^{xy}}{2\rho } &{} \frac{p^{xy2}}{\rho }+\frac{p^{xx}p^{yy}}{2\rho } &{} \frac{3 p^{xy}p^{yy}}{2\rho } \\ &{} &{} &{} &{} &{}\\ \frac{p^{yy}}{2} &{} 0 &{} 0 &{} \frac{p^{xx}p^{yy}}{2\rho }+\frac{p^{xy2}}{\rho } &{} \frac{3 p^{xy}p^{yy}}{2\rho } &{} \frac{3 p^{yy2}}{2\rho } \\ \end{array}\right) . \end{aligned}$$

Following the Theorem (A.1) in [1], the scaling matrix \(\mathbf{T}^x\) is square root of matrix \(\mathbf{Y} =\mathbf{R}_{\mathbf{w}}^{-1x} \frac{\partial \mathbf{w }}{\partial \mathbf{v }}\frac{\partial \mathbf{u }}{\partial \mathbf{w }}^{-\top } \mathbf{R}_{\mathbf{w}}^{-\top x}\), which results in,

$$\begin{aligned} \mathbf{T}^x=\left( \begin{array}{c c c c c c} \frac{1}{\sqrt{12p^{xx2}}\rho } &{} 0 &{} 0 &{} 0 &{} 0 &{} 0\\ &{} &{} &{} &{} &{}\\ 0 &{} \sqrt{\frac{p^{xx}p^{yy}-p^{xy2}}{4\rho p^{xx2}}} &{} 0 &{} 0 &{} 0 &{} 0\\ &{} &{} &{} &{} &{}\\ 0 &{} 0 &{} \frac{\frac{\rho }{3}+s^x}{t^x} &{} \frac{p^{xx}p^{yy}-p^{xy2}}{3p^{xx}t^x} &{} 0 &{} 0\\ &{} &{} &{} &{} &{}\\ 0 &{} 0 &{} \frac{p^{xx}p^{yy}-p^{xy2}}{3p^{xx}t^x} &{} \frac{\frac{4(p^{xx}p^{yy}-p^{xy2})^2}{3p^{xx2}\rho }+s^x}{t^x} &{} 0 &{} 0\\ &{} &{} &{} &{} &{}\\ 0 &{} 0 &{} 0 &{} 0 &{} \sqrt{\frac{p^{xx}p^{yy}-p^{xy2}}{4\rho p^{xx2}}} &{} 0 \\ &{} &{} &{} &{} &{}\\ 0 &{} 0 &{} 0 &{} 0 &{} 0 &{} \sqrt{\frac{1}{12p^{xx2}\rho }} \\ \end{array}\right) , \end{aligned}$$

where \(s^x=\frac{p^{xx}p^{yy}-p^{xy2}}{\sqrt{3}p^{xx}}\) and \(t^{x2}=\frac{\rho }{3}+\frac{4}{3}\frac{(p^{xx}p^{yy}-p^{xy2})^2}{3p^{xx2}\rho }+\frac{2}{\sqrt{3}}\frac{p^{xx}p^{yy}-p^{xy2}}{p^{xx}}\). So, the entropy scaled right eigenvector matrix in primitive variables is,

$$\begin{aligned} \tilde{\mathbf{R}}_{\mathbf{w}}^x= & {} \,\mathbf{R}_{\mathbf{w}}^x\mathbf{T}^x\\= & {} \left( \begin{array}{c c c c c c} A^x\rho p^{xx} &{} 0 &{} \frac{\frac{\rho }{3}+s^x}{t^x} &{} \frac{1}{3t^x\beta _1} &{} 0 &{} A^x\rho p^{xx}\\ -A^x\sqrt{\frac{3p^{xx}}{\rho }}p^{xx} &{} 0 &{} 0 &{} 0 &{} 0 &{} A^x\sqrt{\frac{3p^{xx}}{\rho }}p^{xx}\\ -A^x\sqrt{\frac{3p^{xx}}{\rho }}p^{xy} &{} -B^x\sqrt{\frac{p^{xx}}{\rho }} &{} 0 &{} 0 &{} B^x\sqrt{\frac{p^{xx}}{\rho }} &{} A^x\sqrt{\frac{3p^{xx}}{\rho }}p^{xy} \\ A^x(3p^{xx2}) &{} 0 &{} 0 &{} 0 &{} 0 &{} A^x(3p^{xx2})\\ A^x(3p^{xx}p^{xy}) &{} B^xp^{xx} &{} 0 &{} 0 &{} B^xp^{xx} &{} A^x(3p^{xx}p^{xy})\\ A^x(2p^{xy2}+p^{xx}p^{yy}) &{} 2p^{xy}B^x &{} \frac{1}{3t^x\beta _1} &{} \frac{\frac{4}{3\rho \beta _1^2}+s^x}{t^x} &{} 2p^{xy}B^x &{} A^x(2p^{xy2}+p^{xx}p^{yy}) \end{array}\right) , \end{aligned}$$

where \(A^x=\frac{1}{2\sqrt{3}p^{xx}\sqrt{\rho }}\), \(B^x=\frac{\sqrt{p^{xx}p^{yy}-p^{xy2}}}{2p^{xx}\sqrt{\rho }}\), \(\beta _1=\frac{p^{xx}}{(p^{xx}p^{yy}-p^{xy2})}\). Using (29), we can now calculate entropy scaled right eigenvectors in conservative variables which satisfy (28).

Remark A.1

Proceeding similarly in y-direction, with right eigenvector matrix for system in primitive variable as,

$$\begin{aligned} \mathbf{R}_{\mathbf{w}}^y= \left( \begin{array}{c c c c c c} \rho p^{yy} &{} 0 &{} 1 &{} 0 &{} 0 &{} \rho p^{yy}\\ -\sqrt{\frac{3p^{yy}}{\rho }}p^{xy} &{} -\sqrt{\frac{p^{yy}}{\rho }} &{} 0 &{} 0 &{} \sqrt{\frac{p^{yy}}{\rho }} &{} \sqrt{\frac{3p^{yy}}{\rho }}p^{xy}\\ -\sqrt{\frac{3p^{yy}}{\rho }}p^{yy} &{} 0 &{} 0 &{} 0 &{} 0 &{} \sqrt{\frac{3p^{yy}}{\rho }}p^{yy}\\ 2p^{xy2}+p^{xx}p^{yy} &{} 2p^{xy} &{} 0 &{} 1 &{} 2p^{xy} &{} 2p^{xy2}+p^{xx}p^{yy} \\ 3p^{xy}p^{yy} &{} p^{yy} &{} 0 &{} 0 &{} p^{yy} &{} 3p^{xy}p^{yy}\\ 3p^{yy2} &{} 0 &{} 0 &{} 0 &{} 0 &{} 3p^{yy2} \end{array}\right) , \end{aligned}$$

we obtained scaling matrix \(\mathbf{T}^y\) as,

$$\begin{aligned} \mathbf{T}^{y}=\left( \begin{array}{c c c c c c} \frac{1}{\sqrt{12p^{yy2}}\rho } &{} 0 &{} 0 &{} 0 &{} 0 &{} 0\\ &{} &{} &{} &{} &{}\\ 0 &{} \sqrt{\frac{p^{xx}p^{yy}-p^{xy2}}{4\rho p^{yy2}}} &{} 0 &{} 0 &{} 0 &{} 0\\ &{} &{} &{} &{} &{}\\ 0 &{} 0 &{} \frac{\frac{\rho }{3}+s^y}{t^y} &{} \frac{p^{xx}p^{yy}-p^{xy2}}{3p^{yy}t^y} &{} 0 &{} 0\\ &{} &{} &{} &{} &{}\\ 0 &{} 0 &{} \frac{p^{xx}p^{yy}-p^{xy2}}{3p^{yy}t^y} &{} \frac{\frac{4(p^{xx}p^{yy}-p^{xy2})^2}{3p^{yy2}\rho }+s^y}{t^y} &{} 0 &{} 0\\ &{} &{} &{} &{} &{}\\ 0 &{} 0 &{} 0 &{} 0 &{} \sqrt{\frac{p^{xx}p^{yy}-p^{xy2}}{4\rho p^{yy2}}} &{} 0 \\ &{} &{} &{} &{} &{}\\ 0 &{} 0 &{} 0 &{} 0 &{} 0 &{} \sqrt{\frac{1}{12p^{yy2}\rho }} \\ \end{array}\right) , \end{aligned}$$

where \(s^y=\frac{p^{xx}p^{yy}-p^{xy2}}{\sqrt{3}p^{yy}}\) and \(t^{y2}=\frac{\rho }{3}+\frac{4}{3}\frac{(p^{xx}p^{yy}-p^{xy2})^2}{3p^{yy2}\rho }+\frac{2}{\sqrt{3}}\frac{p^{xx}p^{yy}-p^{xy2}}{p^{yy}}\). Using this, the entropy scaled right eigenvectors in primitive variable in y-direction are,

$$\begin{aligned} \tilde{\mathbf{R}}^y_{\mathbf{w}}= & {} \,\mathbf{R}^y_{\mathbf{w}}{} \mathbf{T}^y\\= & {} \, \left( \begin{array}{c c c c c c} A^y\rho p^{yy} &{} 0 &{} \frac{\frac{\rho }{3}+s^y}{t^y} &{} \frac{1}{3t^y\beta _2} &{} 0 &{} A^y\rho p^{yy}\\ -A^y\sqrt{\frac{3p^{yy}}{\rho }}p^{xy} &{} -B^y\sqrt{\frac{p^{yy}}{\rho }} &{} 0 &{} 0 &{} B^y\sqrt{\frac{p^{yy}}{\rho }} &{} A^y\sqrt{\frac{3p^{yy}}{\rho }}p^{xy}\\ -A^y\sqrt{\frac{3p^{yy}}{\rho }}p^{yy} &{} 0 &{} 0 &{} 0 &{} 0 &{} A^y\sqrt{\frac{3p^{yy}}{\rho }}p^{yy}\\ A^y(2p^{xy2}+p^{xx}p^{yy}) &{} 2p^{xy}B^y &{} \frac{1}{3t^y\beta _2} &{} \frac{\frac{4}{3\rho \beta _2^2}+s^y}{t^y} &{} 2p^{xy}B^y &{} A^y(2p^{xy2}+p^{xx}p^{yy}) \\ A^y(3p^{xy}p^{yy}) &{} B^yp^{yy} &{} 0 &{} 0 &{} B^yp^{yy} &{} A^y(3p^{xy}p^{yy})\\ A^y(3p^{yy2}) &{} 0 &{} 0 &{} 0 &{} 0 &{} A^y(3p^{yy2}) \end{array}\right) \end{aligned}$$

with \(A^y=\frac{1}{2\sqrt{3}p^{yy}\sqrt{\rho }}\), \(B^y=\frac{\sqrt{p^{xx}p^{yy}-p^{xy2}}}{2p^{yy}\sqrt{\rho }}\), \(\beta _2=\frac{p^{yy}}{(p^{xx}p^{yy}-p^{xy2})}\).

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sen, C., Kumar, H. Entropy Stable Schemes For Ten-Moment Gaussian Closure Equations. J Sci Comput 75, 1128–1155 (2018). https://doi.org/10.1007/s10915-017-0579-4

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10915-017-0579-4

Keywords

Mathematics Subject Classification

Navigation