Modified Ghost Fluid Method with Acceleration Correction (MGFM/AC)

Tiegang Liu; Chengliang Feng; Liang Xu

doi:10.1007/s10915-019-01079-x

Journal of Scientific Computing

December 2019, Volume 81, Issue 3, pp 1906–1944 | Cite as

Modified Ghost Fluid Method with Acceleration Correction (MGFM/AC)

Authors
Authors and affiliations

Tiegang Liu
Chengliang Feng
Liang Xu

Article

First Online: 01 November 2019

38 Downloads

Abstract

In this work, we show that the modified ghost fluid method might suffer overheating and leads to inaccurate numerical results when directly applied to a moving rigid boundary with acceleration. We discover the insightful reasons and then develop a new technique to take into account the effect of boundary acceleration on the definition of ghost fluid states based on a generalized Piston–Riemann problem. Theoretical analysis and numerical results show that the modified ghost fluid method with acceleration correction can overcome such difficulty effectively.

Keywords

General piston problem Moving boundary Ghost fluid method Modified ghost fluid method Generalized Piston–Riemann problem

This is a preview of subscription content, to check access.

Notes

Acknowledgements

The research was supported in part by the Science Challenge Project (No. JCKY2016212A502) and National Natural Science Foundation of China (Nos. 11601013 and 91530325).

Appendix A: Basic Differential Relations and the Rankine–Hugoniot Relations for Fluid Flow

To solve the Riemann problem and the generalized Riemann problem, we need to use the concept of Riemann invariants and provide some differential relations among different variables. For more details, we can refer to [35, 39, 40].

For this purpose, we write the system of Euler equations (2.1) in the following form:

$$\begin{aligned} \left\{ {\begin{array}{*{20}{l}} {\frac{{\mathrm{{D}}\rho }}{{\mathrm{{D}}t}} = - \rho \frac{{\partial u}}{{\partial x}}}\\ {\frac{{\mathrm{{D}}u}}{{\mathrm{{D}}t}} = - \frac{1}{\rho }\frac{{\partial P}}{{\partial x}}}\\ {\frac{{\mathrm{{D}}S}}{{\mathrm{{D}}t}} = 0}, \end{array}} \right. \end{aligned}$$

(A.1)

where $ D/Dt=\partial /\partial t+u\cdot \partial /\partial x $ is the material derivative, and the entropy S is related to the other variables through the second law of thermodynamics

$$\begin{aligned} \mathrm{{d}}e = T\mathrm{{d}}S + \frac{P}{{{\rho ^2}}}\mathrm{{d}}\rho , \end{aligned}$$

(A.2)

and T is the temperature. Regarding P as a function of $ \rho $ and S, $ P=P(\rho ,S)$, then the local sound speed c is defined as

$$\begin{aligned} {c^2} = \frac{{\partial P(\rho ,S)}}{{\partial \rho }}. \end{aligned}$$

(A.3)

Thus, the third equation of (A.1) can be replaced equivalently by

$$\begin{aligned} \frac{{\mathrm{{D}}P}}{{\mathrm{{D}}t}} = - \rho {c^2}\frac{{\partial u}}{{\partial x}}. \end{aligned}$$

(A.4)

We observe that the entropy S is constant along a streamline. By the hyperbolic quasilinear analysis for equations (A.1), three eigenvalues of the system are obtained.

$$\begin{aligned} \begin{array}{*{20}{l}} {{\lambda _ - } = u - c},&\quad {{\lambda _0} = u},&\quad {{\lambda _ + } = u + c}. \end{array} \end{aligned}$$

(A.5)

We recall [35], and here introduce the important Riemann invariants $ \phi $ and $\psi $ for $ \lambda _{-},\lambda _{+} $

$$\begin{aligned} \begin{array}{*{20}{l}} {\phi = u - \mathop \int \limits ^\rho \frac{{c(\omega ,S)}}{\omega }d\omega },\\ {\psi = u + \mathop \int \limits ^\rho \frac{{c(\omega ,S)}}{\omega }d\omega }. \end{array} \end{aligned}$$

(A.6)

We carry out the total differential for (A.6)

$$\begin{aligned} d\psi= & {} du + \frac{c}{\rho }d\rho + \frac{{\partial \psi }}{{\partial S}}dS = du + \frac{1}{{\rho c}}dp + K(\rho ,S)dS,\end{aligned}$$

(A.7)

$$\begin{aligned} d\phi= & {} du - \frac{c}{\rho }d\rho - \frac{{\partial \phi }}{{\partial S}}dS = du - \frac{1}{{\rho c}}dp - K(\rho ,S)dS, \end{aligned}$$

(A.8)

where

$$\begin{aligned} K(\rho ,S) = - \frac{1}{{\rho c}}\frac{{\partial p}}{{\partial S}} + \mathop \int \limits ^\rho \frac{1}{\omega }\frac{{\partial c(\omega ,S)}}{{\partial S}}d\omega , \end{aligned}$$

(A.9)

Along the characteristic $ {{C}_{+}}{:}\,dx/dt=u+c$, we have

$$\begin{aligned} \begin{array}{*{20}{l}} {d\psi = K(\rho ,S)dS,}&{dS = c\frac{{\partial S}}{{\partial x}}dt}, \end{array} \end{aligned}$$

(A.10)

and along the characteristic $ {{C}_{-}}:dx/dt=u-c$, we have

$$\begin{aligned} \begin{array}{*{20}{l}} {d\phi = - K(\rho ,S)dS,}&\quad {dS = - c\frac{{\partial S}}{{\partial x}}dt}. \end{array} \end{aligned}$$

(A.11)

In particular, when the fluid is polytropic gases, we have

$$\begin{aligned} p = (\gamma - 1)\rho e, \end{aligned}$$

(A.12)

and the sound speed c,

$$\begin{aligned} {c^2} = \frac{{\gamma P}}{\rho }. \end{aligned}$$

(A.13)

The isentropic relation can be written as

$$\begin{aligned} P = {P_0}{\left( {\frac{\rho }{{{\rho _0}}}} \right) ^\gamma }. \end{aligned}$$

(A.14)

In this case, we can get the Riemann invariants as

$$\begin{aligned} \begin{array}{*{20}{l}} {\phi = u - \frac{{2c}}{{\gamma - 1}},}&\quad {\psi = u + \frac{{2c}}{{\gamma - 1}}}. \end{array} \end{aligned}$$

(A.15)

Taking the partial derivative of S for (A.13),

$$\begin{aligned} 2c\frac{{\partial c}}{{\partial S}}= & {} \frac{\gamma }{\rho }\frac{{\partial p}}{{\partial S}},\end{aligned}$$

(A.16)

$$\begin{aligned} \frac{{\partial \psi }}{{\partial S}}= & {} \frac{2}{{(\gamma - 1)}}\frac{{\partial c}}{{\partial S}} = \frac{\gamma }{{(\gamma - 1)\rho c}}\frac{{\partial p}}{{\partial S}}. \end{aligned}$$

(A.17)

Then (A.9) turns to

$$\begin{aligned} K(\rho ,S) = \frac{1}{{(\gamma - 1)\rho c}}\frac{{\partial p}}{{\partial S}} = \frac{T}{c}. \end{aligned}$$

(A.18)

Bringing (A.18) into (A.7) and (A.8)

$$\begin{aligned} d\psi= & {} du + \frac{1}{{\rho c}}dp + \frac{1}{c}TdS,\end{aligned}$$

(A.19)

$$\begin{aligned} d\phi= & {} du - \frac{1}{{\rho c}}dp - \frac{1}{c}TdS, \end{aligned}$$

(A.20)

and the second thermodynamics law can be written as

$$\begin{aligned} TdS = \frac{1}{{(\gamma - 1)\rho }}dp - \frac{{{c^2}}}{{(\gamma - 1)\rho }}d\rho . \end{aligned}$$

(A.21)

Then we have

$$\begin{aligned} TS'= & {} \frac{1}{{(\gamma - 1)\rho }}p' - \frac{{{c^2}}}{{(\gamma - 1)\rho }}\rho ',\end{aligned}$$

(A.22)

$$\begin{aligned} \psi '= & {} u' + \frac{1}{{\rho c}}p' + \frac{1}{c}TS',\end{aligned}$$

(A.23)

$$\begin{aligned} \phi '= & {} u' - \frac{1}{{\rho c}}p' - \frac{1}{c}TS'. \end{aligned}$$

(A.24)

Taking (A.1) to the Riemann invariants and entropy form

$$\begin{aligned} \left\{ {\begin{array}{*{20}{l}} {\frac{{\partial \phi }}{{\partial t}} + (u - c)\frac{{\partial \phi }}{{\partial x}} = T\frac{{\partial S}}{{\partial x}}}\\ {\frac{{\partial \psi }}{{\partial t}} + (u + c)\frac{{\partial \psi }}{{\partial x}} = T\frac{{\partial S}}{{\partial x}}}\\ {\frac{{\partial S}}{{\partial t}} + u\frac{{\partial S}}{{\partial x}} = 0}. \end{array}} \right. \end{aligned}$$

(A.25)

(A.25) still works across the rarefaction wave which is a smooth wave.

If there is a shock (or contact discontinuity) trajectory $ r=r(t) $ with speed $ \sigma =dr/dt$, assuming $ {{U}_{a}},{{U}_{b}} $ are the states on the wave ahead and behind the wave back. Then across the shock, $ {{U}_{a}},{{U}_{b}} $ satisfy the Rankine–Hugoniot relations,

$$\begin{aligned} \sigma ({U_a} - {U_b}) = F({U_a}) - F({U_b}). \end{aligned}$$

(A.26)

(a)
When $ {{u}_{a}}={{u}_{b}} $ and $ {{\rho }_{a}}\ne {{\rho }_{b}}$, the simple wave is a contact discontinuity, $ {{U}_{a}},{{U}_{b}} $ satisfy $ {{P}_{a}}={{P}_{b}}$, and $ \sigma ={{u}_{a}}\pm {{c}_{a}}={{u}_{b}}\pm {{c}_{a}} $
(b)
When $ {{u}_{a}}\ne {{u}_{b}}$, the simple wave is a shock wave, from (A.26), we can get
$$\begin{aligned} \left\{ {\begin{array}{*{20}{l}} {\sigma = \frac{{{u_a}{\rho _a} - {u_b}{\rho _b}}}{{{\rho _a} - {\rho _b}}}}\\ {{{({u_a} - {u_b})}^2} = ({P_a} - {P_b})\left( \frac{1}{{{\rho _b}}} - \frac{1}{{{\rho _a}}}\right) }\\ {{e_a} - {e_b} = \frac{{{\rho _a} - {\rho _b}}}{{2{\rho _a}{\rho _b}}}({P_a} + {P_b})}. \end{array}} \right. \end{aligned}$$

(A.27)

By the combination of EOS function $ e=e(\rho ,P) $ and (A.27), we can get the following relation form

$$\begin{aligned}&{u_b} = {u_a} \pm H({p_b};{p_a},{\rho _a}),\end{aligned}$$

(A.28)

$$\begin{aligned}&{\rho _b} = G({P_b};{P_a},{\rho _a}), \end{aligned}$$

(A.29)

here, when the shock is related to $ {{\lambda }_{+}}=u+c$, the sign is (+), else should be (−) which is related to $ {{\lambda }_{-}}=u-c $.

The partial derivatives are noted as following forms.

$$\begin{aligned}&\begin{array}{*{20}{l}} {{H_1}({p_b};{p_a},{\rho _a}) = \frac{{\partial H}}{{\partial {p_b}}},}&\quad {{H_2}({p_b};{p_a},{\rho _a}) = \frac{{\partial H}}{{\partial {p_a}}}}&\quad {{H_3}({p_b};{p_a},{\rho _a}) = \frac{{\partial H}}{{\partial {\rho _a}}}}, \end{array}\end{aligned}$$

(A.30)

$$\begin{aligned}&\begin{array}{*{20}{l}} {{G_1}({p_b};{p_a},{\rho _a}) = \frac{{\partial G}}{{\partial {p_b}}},}&\quad {{G_2}({p_b};{p_a},{\rho _a}) = \frac{{\partial G}}{{\partial {p_a}}}}&\quad {{G_3}({p_b};{p_a},{\rho _a}) = \frac{{\partial G}}{{\partial {\rho _a}}}}. \end{array} \end{aligned}$$

(A.31)

In particular, when the EOS function is (A.12), taking $ {{\mu }^{2}}=(\gamma -1)/(\gamma +1)$, we have

$$\begin{aligned}&H({P_b};{P_a},{\rho _a}) = ({P_b} - {P_a})\sqrt{\frac{{1 - {\mu ^2}}}{{{\rho _a}({P_b} + {\mu ^2}{P_a})}}},\end{aligned}$$

(A.32)

$$\begin{aligned}&\begin{array}{*{20}{l}} {{H_1} = \frac{1}{2}\sqrt{\frac{{1 - {\mu ^2}}}{{{\rho _a}({P_b} + {\mu ^2}{P_a})}}} \frac{{{P_b} + (1 + 2{\mu ^2}){P_a}}}{{{P_b} + {\mu ^2}{P_a}}}}\\ {{H_2} = - \frac{1}{2}\sqrt{\frac{{1 - {\mu ^2}}}{{{\rho _a}({P_b} + {\mu ^2}{P_a})}}} \frac{{(2 + {\mu ^2}){P_b} + {\mu ^2}{P_a}}}{{{P_b} + {\mu ^2}{P_a}}}}\\ {{H_3} = - \frac{{{P_b} - {P_a}}}{{2{\rho _a}}}\sqrt{\frac{{1 - {\mu ^2}}}{{{\rho _a}({P_b} + {\mu ^2}{P_a})}}} }, \end{array}\end{aligned}$$

(A.33)

$$\begin{aligned}&\begin{array}{*{20}{l}} {}&{G({P_b};{P_a},{\rho _a}) = {\rho _a}\frac{{{P_b} + {\mu ^2}{P_a}}}{{{P_a} + {\mu ^2}{P_b}}}}, \end{array}\end{aligned}$$

(A.34)

$$\begin{aligned}&\begin{array}{*{20}{l}} {{G_1} = \frac{{{\rho _a}(1 - {\mu ^4}){P_a}}}{{{{({P_a} + {\mu ^2}{P_b})}^2}}},}&{{G_2} = \frac{{{\rho _a}({\mu ^4} - 1){P_b}}}{{{{({P_a} + {\mu ^2}{P_b})}^2}}},}&{{G_3} = \frac{{{P_b} + {\mu ^2}{P_a}}}{{{P_a} + {\mu ^2}{P_b}}}}. \end{array} \end{aligned}$$

(A.35)

Appendix B: Wave Pattern and Interface Solution of Riemann Problem

Considering the following Riemann problem with the EOS of gas (A.12),

$$\begin{aligned} \begin{array}{*{20}{l}} {\frac{{\partial U}}{{\partial t}} + \frac{{\partial F(U)}}{{\partial x}} = 0}\\ {\begin{array}{*{20}{l}} {U(x,0) = \left\{ {\begin{array}{*{20}{l}} {{U_L}}&{}\quad {x < x_I^0}\\ {{U_R}}&{}\quad {x > x_I^0}. \end{array}} \right. }&{} \end{array}} \end{array} \end{aligned}$$

(B.1)

As shown in Fig. 15, it is known that the wave pattern is three-waves structure [39], which are a left simple wave $ W_L $ related to $ {{\lambda }_{-}}=u-c$, a contact discontinuity related to $ {{\lambda }_{0}}=u$, and a right simple wave $ W_R $ related to $ {{\lambda }_{+}}=u+c $. The regions between these waves are two constant fluid states $ U_{1*}$, $ U_{2*} $.

Across the three waves, as shown in Table 1, these states satisfy conditions for Riemann invariants to be equal or satisfy the Rankine–Hugoniot relations.

Table 1

Relations across the waves of Riemann problem

	Left wave	Contact discontinuity	Right wave
Rarefaction wave	$ \int \limits _{{U_L}}^{{U_{1*}}} {d\psi } = 0 $		$ \int \limits _{{U_R}}^{{U_{2*}}} {d\phi } = 0 $
	$ S({\rho _{1}},{P_{1}}) = S({\rho _L},{P_L}) $		$ S({\rho _{2}},{P_{2}}) = S({\rho _R},{P_R}) $
Shock wave	$ {u_{1}} = {u_L} - H({P_{1}};{P_L},{\rho _L}) $	$ {u_{1}} = {u_{2}} $	${u_{2}} = {u_R} + H({P_{2}};{P_R},{\rho _R}) $
	$ {\rho _{1}} = G({P_{1}};{P_L},{\rho _L})$	${P_{1}} = {P_{2}} $	$ {\rho _{2}} = G({P_{2}};{P_R},{\rho _R}) $

Wave Pattern and Interface Solution of Riemann Problem Obtained by RBC

From the right side state definition (2.6) by RBC, the initial state distributions of Riemann problem (B.1) satisfy

$$\begin{aligned} \begin{array}{*{20}{l}} {{U_L}{:}\,\left\{ {\begin{array}{*{20}{l}} {{u_L}}\\ {{P_L}}\\ {{\rho _L}} \end{array}} \right. },&{}{{U_R}{:}\,\left\{ {\begin{array}{*{20}{l}} {{u_R} = 2u_I^0 - {u_L}}\\ {{P_R} = {P_L}}\\ {{\rho _R} = {\rho _L}} \end{array}} \right. }. \end{array} \end{aligned}$$

(B.2)

Then, we can get the solution of this problem as following form.

Fig. 16
Wave pattern of Riemann problem obtained by RBC when $ u_{I}^{0}<{{u}_{L}} $

a.
When $ u_{I}^{0}<{{u}_{L}}$, as shown in Fig. 16, the three-waves structure is composed of two shock wave $ W_L, W_R $ and one contact discontinuity, and the states $ U_{1*},U_{2*} $ satisfy
$$\begin{aligned}&\begin{array}{*{20}{l}} {{u_{1*}} = {u_L} - H({P_{1*}};{P_L},{\rho _L})}\\ {{\rho _{1*}} = G({P_{1*}};{P_L},{\rho _L})}, \end{array}\end{aligned}$$

(B.3)
$$\begin{aligned}&\begin{array}{*{20}{l}} {{u_{2*}} = {u_R} + H({P_{2*}};{P_R},{\rho _R})}\\ {{\rho _{2*}} = G({P_{2*}};{P_R},{\rho _R})}, \end{array}\end{aligned}$$

(B.4)
$$\begin{aligned}&\begin{array}{*{20}{l}} {{u_{1*}} = {u_{2*}}}\\ {{P_{1*}} = {P_{2*}}}. \end{array} \end{aligned}$$

(B.5)

Bringing (B.2) into (B.4)

$$\begin{aligned} \begin{array}{*{20}{l}} {{u_{2*}} = 2u_I^0 - {u_L} + H({P_{2*}};{P_L},{\rho _L})}\\ {{\rho _{2*}} = G({P_{2*}};{P_L},{\rho _L})}. \end{array} \end{aligned}$$

(B.6)

By the combination of (B.5), (B.3) and (B.6), we can get the following relation form

$$\begin{aligned} \begin{array}{*{20}{l}} {{u_{1*}} = {u_{2*}} = u_I^0}\\ {{\rho _{1*}} = {\rho _{2*}} = G({P_{2*}};{P_L},{\rho _L})}. \end{array} \end{aligned}$$

(B.7)

Here, we note the state on the piston interface as $ (u_{ shock }^{*},P_{ shock }^{*},\rho _{ shock }^{*}):=(u_{1*},P_{1*},\rho _{1*}) $, then they should satisfy

$$\begin{aligned} \begin{array}{*{20}{l}} {u_{ shock }^* = u_I^0}\\ {H(P_{ shock }^*;{P_L},{\rho _L}) = {u_L} - u_{ shock }^*}\\ {\rho _{ shock }^* = G(P_{ shock }^*;{P_L},{\rho _L})}. \end{array} \end{aligned}$$

(B.8)

Fig. 17
Wave pattern of Riemann problem obtained by RBC when $ u_{I}^{0}\ge {{u}_{L}} $

b.
When $ u_{I}^{0}\ge {{u}_{L}}$, as shown in Fig. 17, the three-waves structure is composed of two centered rarefaction wave $ W_L, W_R $ and one contact discontinuity, and the states $ U_{1*},U_{2*} $ satisfy
$$\begin{aligned}&\begin{array}{*{20}{l}} {\int \limits _{{U_L}}^{{U_{1*}}} {d\psi } = {u_{1*}} - {u_L} + \int \limits _{{P_L}}^{{P_{1*}}} {\frac{1}{{\rho c}}dP} = 0}\\ {S({\rho _{1*}},{P_{1*}}) = S({\rho _L},{P_L})}, \end{array}\end{aligned}$$

(B.9)
$$\begin{aligned}&\begin{array}{*{20}{l}} {\int \limits _{{U_R}}^{{U_{2*}}} {d\phi } = {u_{2*}} - {u_R} - \int \limits _{{P_R}}^{{P_{2*}}} {\frac{1}{{\rho c}}dP} = 0}\\ {S({\rho _{2*}},{P_{2*}}) = S({\rho _R},{P_R})}, \end{array}\end{aligned}$$

(B.10)
$$\begin{aligned}&\begin{array}{*{20}{l}} {{u_{1*}} = {u_{2*}}}\\ {{P_{1*}} = {P_{2*}}}. \end{array} \end{aligned}$$

(B.11)

Bringing (B.2) into (B.10)

$$\begin{aligned} \begin{array}{*{20}{l}} {{u_{2*}} = 2u_I^0 - {u_L} + \int \limits _{{P_L}}^{{P_{2*}}} {\frac{1}{{\rho c}}dP} = 0}\\ {S({\rho _{2*}},{P_{2*}}) = S({\rho _L},{P_L})}. \end{array} \end{aligned}$$

(B.12)

By the combination of (B.9), (B.11) and (B.12), we can get the following relation form

$$\begin{aligned} \begin{array}{*{20}{l}} {{u_{1*}} = {u_{2*}} = u_I^0}\\ {{\rho _{1*}} = {\rho _{2*}}}. \end{array} \end{aligned}$$

(B.13)

Here, we note the state on the piston interface as $ (u_{ rare }^{*},P_{ rare }^{*},\rho _{ rare }^{*}):=({{u}_{1*}},{{P}_{1*}},{{\rho }_{1*}})$, then they should satisfy

$$\begin{aligned} \begin{array}{*{20}{l}} {u_{ rare }^* = u_I^0},\\ {\int \limits _{{P_L}}^{P_{ rare }^*} {\frac{1}{{\rho c}}dP} = {u_L} - u_{ rare }^*},\\ {S(\rho _{ rare }^*,P_{ rare }^*) = S({\rho _L},{P_L})}. \end{array} \end{aligned}$$

(B.14)

Appendix C: Wave Pattern and Interface Solution of Generalized Riemann Problem

Considering the following generalized Riemann problem,

$$\begin{aligned} \begin{array}{*{20}{l}} {\frac{{\partial U}}{{\partial t}} + \frac{{\partial F(U)}}{{\partial x}} = 0}\\ {\begin{array}{*{20}{l}} {U(x,0) = \left\{ {\begin{array}{*{20}{l}} {{U_L} + (x - x_I^0){{U'}_L}}&{}\quad {x < x_I^0}\\ {{U_R} + (x - x_I^0){{U'}_R}}&{}\quad {x > x_I^0}. \end{array}} \right. }&{} \end{array}} \end{array} \end{aligned}$$

(C.1)

As shown in Fig. 18, The solution is piecewise smooth in a small-time range.

Note $ {{U}_{1*}}={\mathop {\lim }\nolimits _{t\rightarrow 0+}}\,{{U}_{1}}({{x}_{I}}-,t) $ and $ {{U}_{2*}}={\mathop {\lim }\nolimits _{t\rightarrow 0+}}\,{{U}_{2}}({{x}_{I}}+,t)$, referring to [35], it is known that $ (U_{1*}^{{}},U_{2*}^{{}}) $ are the solutions on both side of the contact discontinuity for Riemann problem (B.1). With time $ t\rightarrow 0+ $ and $ x\rightarrow {{x}_{I}}-$, the limiting material derivatives $ (Du/Dt)_{1*}$, $ {{(DP/Dt)}_{1*}} $ behind the left wave $ W_L $ satisfy

$$\begin{aligned} {A_L}\left( {\frac{{Du}}{{Dt}}} \right) _{1*}^ - + {B_L}\left( {\frac{{DP}}{{Dt}}} \right) _{1*}^ - = {D_L}, \end{aligned}$$

(C.2)

Fig. 18
Wave pattern of generalized Riemann problem

with

$$\begin{aligned}&({A_L},{B_L},{D_L}) = \left\{ {\begin{array}{*{20}{ll}} {(A_L^{ shock },B_L^{ shock },D_L^{ shock })}&{}\quad {if}{{u_L} > {u_{1*}}}\\ {(A_L^{ Rare },B_L^{ Rare },D_L^{ Rare })}&{}\quad {if}{{u_L} \le {u_{1*}}}, \end{array}} \right. \end{aligned}$$

(C.3)

$$\begin{aligned}&\begin{array}{*{20}{l}} {A_L^{ shock } = 1 - {\rho _{1*}}({\sigma _L} - {u_{1*}})H_1^L},&{}{B_L^{ shock } = - \left[ {\frac{1}{{{\rho _{1*}}c_{1*}^2}}({\sigma _L} - {u_{1*}}) - H_1^L} \right] },\\ {D_L^{ shock } = L_P^L{{P'}_L} + L_u^L{{u'}_L} + L_\rho ^L{{\rho '}_L}}, &{} {{\sigma _L} = \frac{{{\rho _L}{u_L} - {\rho _{1*}}{u_{1*}}}}{{{\rho _L} - {\rho _{1*}}}}}, \end{array}\end{aligned}$$

(C.4)

$$\begin{aligned}&\begin{array}{*{20}{l}} {L_P^L = - \frac{1}{{{\rho _L}}} - ({\sigma _L} - {u_L})H_2^L,}&{}{}&{}{L_u^L = {\sigma _L} - {u_L} + ({\rho _L}c_L^2H_2^L + {\rho _L}H_3^L),}\\ {L_\rho ^L = - ({\sigma _L} - {u_L})H_3^L.}&{}{}&{}{} \end{array} \end{aligned}$$

(C.5)

Here, $ H_{i}^{L}={{H}_{i}}({{P}_{*}};{{P}_{L}},{{\rho }_{L}}),i=1,2,3 $ are given in (A.30) and (A.33).

$$\begin{aligned}&\begin{array}{*{20}{l}} {A_L^{ rare } = 1,}&{B_L^{ rare } = \frac{1}{{{\rho _{1*}}{c_{1*}}}}}, \end{array}\end{aligned}$$

(C.6)

$$\begin{aligned}&D_L^{ rare } \!= \! \left[ {\frac{{1 \!+\! {\mu ^2}}}{{1 \!+\! 2{\mu ^2}}}{{\left( {\frac{{{c_{1*}}}}{{{c_L}}}} \right) }^{1/(2{\mu ^2})}} \!+\! \frac{{{\mu ^2}}}{{1 \!+\! 2{\mu ^2}}}{{\left( {\frac{{{c_{1*}}}}{{{c_L}}}} \right) }^{(1 + {\mu ^2})/{\mu ^2}}}} \right] {T_L}{S'_L} - {c_L}{\left( {\frac{{{c_{1*}}}}{{{c_L}}}} \right) ^{1/(2{\mu ^2})}}{\psi '_L}.\nonumber \\ \end{aligned}$$

(C.7)

Here, $ {{T}_{L}}{{{S}'}_{L}}$, $ {{{\psi }'}_{L}} $ are given in (A.22) and (A.23).

Particularly, when $ {{u}_{*}}={{u}_{L}}$, (C.7) becomes

$$\begin{aligned} D_L^{ rare } = - \frac{1}{{{\rho _L}}}{P'_L} - {c_L}{u'_L}. \end{aligned}$$

(C.8)

The limiting material derivatives $ {{(Du/Dt)}_{2*}}$, $ {{(DP/Dt)}_{2*}} $ behind the right wave $ {{W}_{R}} $ satisfy

$$\begin{aligned} {A_R}\left( {\frac{{Du}}{{Dt}}} \right) _{2*}^ + + {B_R}\left( {\frac{{DP}}{{Dt}}} \right) _{2*}^ + = {D_R}, \end{aligned}$$

(C.9)

with

$$\begin{aligned}&({A_R},{B_R},{D_R}) = \left\{ {\begin{array}{*{20}{ll}} {(A_R^{ shock },B_R^{ shock },D_R^{ shock })}&{}\quad {if}{{u_{2*}} > {u_R}}\\ {(A_R^{ Rare },B_R^{ Rare },D_R^{ Rare })}&{}\quad {if}{{u_{2*}} \le {u_R}}, \end{array}} \right. \end{aligned}$$

(C.10)

$$\begin{aligned}&\begin{array}{*{20}{l}} {A_R^{ shock } = 1 + {\rho _{2*}}({\sigma _R} - {u_*})H_1^R},&{}{B_R^{ shock } = - \left[ {\frac{1}{{{\rho _{2*}}c_{2*}^2}}({\sigma _R} - {u_*}) + H_1^R} \right] },\\ {D_R^{ shock } = L_P^R{{P'}_R} + L_u^R{{u'}_R} + L_\rho ^R{{\rho '}_R}},&{}{{\sigma _R} = \frac{{{\rho _R}{u_R} - {\rho _{2*}}{u_{2*}}}}{{{\rho _R} - {\rho _{2*}}}}}, \end{array}\nonumber \\ \end{aligned}$$

(C.11)

$$\begin{aligned}&\begin{array}{*{20}{l}} {L_P^R = - \frac{1}{{{\rho _R}}} + ({\sigma _R} - {u_R})H_2^R,}&{}{}&{}{L_u^R = {\sigma _R} - {u_R} - ({\rho _R}c_R^2H_2^R + {\rho _R}H_3^R),}\\ {L_\rho ^R = ({\sigma _R} - {u_R})H_3^R.}&{}{}&{}{} \end{array} \end{aligned}$$

(C.12)

Here, $ H_{i}^{R}={{H}_{i}}({{P}_{*}};{{P}_{R}},{{\rho }_{R}}),i=1,2,3$, are given in (A.30) and (A.33).

$$\begin{aligned}&\left( {A_R^{ rare },B_R^{ rare }} \right) = \left( {1, - \frac{1}{{{\rho _{2*}}{c_{2*}}}}} \right) ,\end{aligned}$$

(C.13)

$$\begin{aligned}&{D_{ rare }} \!= \!\left[ {\frac{{1 \!+\! {\mu ^2}}}{{1 \!+\! 2{\mu ^2}}}{{\left( {\frac{{{c_{2*}}}}{{{c_R}}}} \right) }^{1/(2{\mu ^2})}} \!+\! \frac{{{\mu ^2}}}{{1 + 2{\mu ^2}}}{{\left( {\frac{{{c_{2*}}}}{{{c_R}}}} \right) }^{(1 \!+\! {\mu ^2})/{\mu ^2}}}} \right] {T_R}{S'_R} + {c_R}{\left( {\frac{{{c_{2*}}}}{{{c_R}}}} \right) ^{1/(2{\mu ^2})}}{\phi '_R}.\nonumber \\ \end{aligned}$$

(C.14)

Here, $ {{T}_{R}}{{{S}'}_{R}}$, $ {{{\phi }'}_{R}} $ are given in (A.22) and (A.24).

Particularly, when $ {{u}_{*}}={{u}_{R}}$, (C.14) turns to

$$\begin{aligned} D_R^{ rare } = - \frac{1}{{{\rho _R}}}{P'_R} + {c_R}{u'_R}. \end{aligned}$$

(C.15)

Thus, turning the material derivatives to spatial derivatives, we have the limiting spatial derivatives $ {{(\partial u/\partial x)}_{1*}}$, $ {{(\partial \rho /\partial x)}_{1*}}$, $ {{(\partial P/\partial x)}_{1*}} $ behind the left wave $ W_L $ satisfy

$$\begin{aligned} \begin{array}{*{20}{l}} {\frac{{\partial {P_{1*}}}}{{\partial x}} = - \rho _{1*}^{}\frac{{D{u_{1*}}}}{{Dt}}},\\ {\frac{{\partial {u_{1*}}}}{{\partial x}} = - \frac{1}{{\rho _{1*}^{}c_{1*}^2}}\frac{{D{P_{1*}}}}{{Dt}}},\\ {\frac{{\partial {\rho _{1*}}}}{{\partial x}} = \left\{ {\begin{array}{*{20}{l}} {\frac{{\partial \rho _{ shock }^{1*}}}{{\partial x}}}&{}\quad {u_*^{} < {u_L}}\\ {\frac{{\partial \rho _{ rare }^{1*}}}{{\partial x}}}&{}\quad {u_*^{} \ge {u_L}} \end{array}} \right. }. \end{array} \end{aligned}$$

(C.16)

Here, $ \frac{\partial \rho _{ shock }^{1*}}{\partial x} $ and $ \frac{\partial \rho _{ rare }^{1*}}{\partial x} $ satisfy

$$\begin{aligned}&\frac{{\partial {P_{1*}}}}{{\partial x}} - c_{1*}^2\frac{{\partial \rho _{ rare }^{1*}}}{{\partial x}} = (\gamma - 1){\rho _{1*}}{\left( {\frac{{{c_{1*}}}}{{{c_L}}}} \right) ^{\frac{{1 + {\mu ^2}}}{{{\mu ^2}}}}} \cdot {T_L}S{'_L},\end{aligned}$$

(C.17)

$$\begin{aligned}&g_{\rho ,L}^{ shock }\frac{{\partial {\rho _{1*}}}}{{\partial x}} + g_{P,L}^{ shock }\frac{{\partial {P_{1*}}}}{{\partial x}} + g_{u,L}^{ shock }\frac{{\partial {u_{1*}}}}{{\partial x}} = f_L^{ shock }, \end{aligned}$$

(C.18)

with

$$\begin{aligned}&{g_{\rho ,L}^{ shock } = {\sigma _L} - {u_{1*}},}\quad {g_{P,L}^{ shock } = - ({\sigma _L} - {u_{1*}}){G_1},}\quad {g_{u,L}^{ shock } = - {\rho _{1*}} + {\rho _{1*}}c_{1*}^2{G_1}},\nonumber \\&f_L^{ shock } = ({\sigma _L} - {u_L}){G_2}{P'_L} + ({\sigma _L} - {u_L}){G_3}{\rho '_L} - {\rho _L}({G_2}c_L^2 + {G_3}){u'_L}. \end{aligned}$$

(C.19)

where $ {{G}_{i}}$, $ i = 1,2,3 $ are given in (A.35);

The limiting spatial derivatives $ {{(\partial u/\partial x)}_{2*}} $, $ {{(\partial \rho /\partial x)}_{2*}}$, $ {{(\partial P/\partial x)}_{2*}} $ behind the right wave $ {{W}_{R}} $ satisfy

$$\begin{aligned} \begin{array}{*{20}{l}} {\frac{{\partial {P_{2*}}}}{{\partial x}} = - \rho _{2*}^{}\frac{{D{u_{2*}}}}{{Dt}}},\\ {\frac{{\partial {u_{2*}}}}{{\partial x}} = - \frac{1}{{\rho _{2*}^{}c_{2*}^2}}\frac{{D{P_{2*}}}}{{Dt}}},\\ {\frac{{\partial {\rho _{2*}}}}{{\partial x}} = \left\{ {\begin{array}{*{20}{l}} {\frac{{\partial \rho _{ shock }^{2*}}}{{\partial x}}}&{}\quad {u_*^{} > {u_R}}\\ {\frac{{\partial \rho _{ rare }^{2*}}}{{\partial x}}}&{}\quad {u_*^{} \le {u_R}} \end{array}} \right. }. \end{array} \end{aligned}$$

(C.20)

Here, $ \frac{\partial \rho _{ shock }^{2*}}{\partial x} $ and $ \frac{\partial \rho _{ rare }^{2*}}{\partial x} $ satisfy

$$\begin{aligned}&\frac{{\partial {P_{2*}}}}{{\partial x}} - c_{2*}^2\frac{{\partial \rho _{ rare }^{2*}}}{{\partial x}} = (\gamma - 1){\rho _{2*}}{\left( {\frac{{{c_{2*}}}}{{{c_R}}}} \right) ^{\frac{{1 + {\mu ^2}}}{{{\mu ^2}}}}} \cdot {T_R}S{'_R},\end{aligned}$$

(C.21)

$$\begin{aligned}&g_{\rho ,R}^{ shock }\frac{{\partial {\rho _{2*}}}}{{\partial x}} + g_{P,R}^{ shock }\frac{{\partial {P_{2*}}}}{{\partial x}} + g_{u,R}^{ shock }\frac{{\partial {u_{2*}}}}{{\partial x}} = f_R^{ shock }, \end{aligned}$$

(C.22)

with

$$\begin{aligned}&{g_{\rho ,R}^{ shock } = {\sigma _R} - {u_{2*}},}\quad {g_{P,R}^{ shock } = - ({\sigma _R} - {u_{2*}}){G_1},}\quad {g_{u,R}^{ shock } = - {\rho _{2*}} + {\rho _{2*}}c_{2*}^2{G_1},}\nonumber \\&f_R^{ shock } = ({\sigma _R} - {u_R}){G_2}{P'_R} + ({\sigma _R} - {u_R}){G_3}{\rho '_R} - {\rho _R}({G_2}c_R^2 + {G_3}){u'_R}. \end{aligned}$$

(C.23)

Here $ {{G}_{i}}$, $ i = 1,2,3 $ are given in (A.35).

Wave Pattern and Interface Solution of Generalized Riemann Problem Obtained by RBC

From the right side state definition (2.6) by RBC, the initial state distributions of generalized Riemann problem (C.1) satisfy

$$\begin{aligned} \begin{array}{*{20}{l}} {{U_L}{:}\,\left\{ {\begin{array}{*{20}{l}} {{u_L}}\\ {{P_L}}\\ {{\rho _L}} \end{array}} \right. },&{}\quad {{U_R}{:}\,\left\{ {\begin{array}{*{20}{l}} {{u_R} = 2u_I^0 - {u_L}}\\ {{P_R} = {P_L}}\\ {{\rho _R} = {\rho _L}} \end{array}} \right. },\\ {{{U'}_L}{:}\,\left\{ {\begin{array}{*{20}{l}} {{{u'}_L}}\\ {{{P'}_L}}\\ {{{\rho '}_L}} \end{array}} \right. },&{}\quad {{{U'}_R}{:}\,\left\{ {\begin{array}{*{20}{l}} {{{u'}_R} = {{u'}_L}}\\ {{{P'}_R} = - {{P'}_L}}\\ {{{\rho '}_R} = - {{\rho '}_L}} \end{array}} \right. }. \end{array} \end{aligned}$$

(C.24)

Then, as shown in Fig. 19, we can get the solution of this problem as following form.

a.
When $ u_{I}^{0}<{{u}_{L}}$, the three-waves structure is composed of two shock wave $ W_L, W_R $ and one contact discontinuity, the limiting states $ {{U}_{1*}}$, $ {{U}_{2*}} $ and their material derivatives $ {{(Du/Dt)}_{1*}}$, $ {{(DP/Dt)}_{1*}}$, $ {{(Du/Dt)}_{2*}}$, $ {{(DP/Dt)}_{2*}} $ behind the wave $ W_L, W_R $ satisfy
$$\begin{aligned}&{U_{1*}} = {U_{2*}},\end{aligned}$$

(C.25)
$$\begin{aligned}&A_L^{ shock }\left( {\frac{{Du}}{{Dt}}} \right) _{1*}^ - + B_L^{ shock }\left( {\frac{{DP}}{{Dt}}} \right) _{1*}^ - = D_L^{ shock },\end{aligned}$$

(C.26)
$$\begin{aligned}&A_R^{ shock }\left( {\frac{{Du}}{{Dt}}} \right) _{2*}^ + + B_R^{ shock }\left( {\frac{{DP}}{{Dt}}} \right) _{2*}^ + = D_R^{ shock },\end{aligned}$$

(C.27)
$$\begin{aligned}&\begin{array}{*{20}{l}} {\left( {\frac{{Du}}{{Dt}}} \right) _{1*}^ - = \left( {\frac{{Du}}{{Dt}}} \right) _{2*}^ + },\\ {\left( {\frac{{DP}}{{Dt}}} \right) _{1*}^ - = \left( {\frac{{DP}}{{Dt}}} \right) _{2*}^ + }. \end{array} \end{aligned}$$

(C.28)

Here, $ {{U}_{1*}}$, $ {{U}_{2*}} $ are given in (B.8).

Bringing (C.24), (C.25) into (C.4) and (C.11), we can get

$$\begin{aligned} \begin{array}{*{20}{l}} {{\sigma _R} = 2u_I^0 - {\sigma _L}},\\ {A_R^{ shock } = A_L^{ shock }},\\ {B_R^{ shock } = - B_L^{ shock }},\\ {D_R^{ shock } = - D_L^{ shock }}. \end{array} \end{aligned}$$

(C.29)

Then, (C.27) becomes

$$\begin{aligned} A_L^{ shock }\left( {\frac{{Du}}{{Dt}}} \right) _{2*}^ + - B_L^{ shock }\left( {\frac{{DP}}{{Dt}}} \right) _{2*}^ + = - D_L^{ shock }. \end{aligned}$$

(C.30)

From (C.28), (C.26) and (C.30), we have

$$\begin{aligned} \begin{array}{*{20}{l}} {\left( {\frac{{Du}}{{Dt}}} \right) _*^{}: = \left( {\frac{{Du}}{{Dt}}} \right) _{1*}^ - = 0},\\ {\left( {\frac{{DP}}{{Dt}}} \right) _*^{}: = \left( {\frac{{DP}}{{Dt}}} \right) _{1*}^ - = \frac{{D_L^{ shock }}}{{B_L^{ shock }}}}. \end{array} \end{aligned}$$

(C.31)

b.
When $ u_{I}^{0}\ge {{u}_{L}}$, the three-waves structure is composed of two rarefaction wave $ W_L, W_R $ and one contact discontinuity, the limiting states $ {{U}_{1*}}$, $ {{U}_{2*}} $ and their material derivatives $ {{(Du/Dt)}_{1*}}$, $ {{(DP/Dt)}_{1*}}$, $ {{(Du/Dt)}_{2*}}$, $ {{(DP/Dt)}_{2*}} $ behind the wave $ W_L, W_R $ satisfy
$$\begin{aligned}&{U_{1*}} = {U_{2*}},\end{aligned}$$

(C.32)
$$\begin{aligned}&A_L^{ rare }\left( {\frac{{Du}}{{Dt}}} \right) _{1*}^ - + B_L^{ rare }\left( {\frac{{DP}}{{Dt}}} \right) _{1*}^ - = D_L^{ rare },\end{aligned}$$

(C.33)
$$\begin{aligned}&A_R^{ rare }\left( {\frac{{Du}}{{Dt}}} \right) _{2*}^ + + B_R^{ rare }\left( {\frac{{DP}}{{Dt}}} \right) _{2*}^ + = D_R^{ rare },\end{aligned}$$

(C.34)
$$\begin{aligned}&\begin{array}{*{20}{l}} {\left( {\frac{{Du}}{{Dt}}} \right) _{1*}^ - = \left( {\frac{{Du}}{{Dt}}} \right) _{2*}^ + },\\ {\left( {\frac{{DP}}{{Dt}}} \right) _{1*}^ - = \left( {\frac{{DP}}{{Dt}}} \right) _{2*}^ + }. \end{array} \end{aligned}$$

(C.35)

Here, $ {{U}_{1*}}$, $ {{U}_{2*}} $ are given in (B.14).

Bringing (C.24), (C.32) into (C.6) and (C.13), we can get

$$\begin{aligned} \begin{array}{*{20}{l}} {A_R^{ rare } = A_L^{ rare }},\\ {B_R^{ rare } = - B_L^{ rare }},\\ {D_R^{ rare } = - D_L^{ rare }}. \end{array} \end{aligned}$$

(C.36)

Then, (C.34) becomes

$$\begin{aligned} A_L^{ rare }\left( {\frac{{Du}}{{Dt}}} \right) _{2*}^ + - B_L^{ rare }\left( {\frac{{DP}}{{Dt}}} \right) _{2*}^ + = - D_L^{ rare }. \end{aligned}$$

(C.37)

From (C.33), (C.37) and (C.35), we have

$$\begin{aligned}&\left( {\frac{{Du}}{{Dt}}} \right) _*^{}: = \left( {\frac{{Du}}{{Dt}}} \right) _{1*}^ - = 0,\end{aligned}$$

(C.38)

$$\begin{aligned}&\left( {\frac{{DP}}{{Dt}}} \right) _*^{}: = \left( {\frac{{DP}}{{Dt}}} \right) _{1*}^ - = \frac{{D_L^{ rare }}}{{B_L^{ rare }}}. \end{aligned}$$

(C.39)

References

1.
Larrouturou, B.: How to preserve the mass fractions positivity when computing compressible multicomponent flows. J. Comput. Phys. 95, 59–84 (1991)MathSciNetCrossRefGoogle Scholar
2.
Karni, S.: Multicomponent flow calculations by a consistent primitive algorithm. J. Comput. Phys. 112, 31–43 (1994)MathSciNetCrossRefGoogle Scholar
3.
Karni, S.: Hybrid multifluid algorithms. SIAM J. Sci. Comput. 17, 1019–1039 (1996)MathSciNetCrossRefGoogle Scholar
4.
Abgrall, R.: How to prevent oscillations in multicomponent flow calculations: a quasi conservative approach. J. Comput. Phys. 125, 150–160 (1996)MathSciNetCrossRefGoogle Scholar
5.
Jenny, P., Muller, B., Thomann, H.: Correction of conservative Euler solvers for gas mixtures. J. Comput. Phys. 132, 91–107 (1997)MathSciNetCrossRefGoogle Scholar
6.
Abgrall, R., Karni, S.: Computations of compressible multifluids. J. Comput. Phys. 169, 594–623 (2001)MathSciNetCrossRefGoogle Scholar
7.
van Brummelen, E.H., Koren, B.: A pressure-invariant conservative Godunov-type method for barotropic two-fluid flows. J. Comput. Phys. 185, 289–308 (2003)MathSciNetCrossRefGoogle Scholar
8.
Nourgaliev, R.R., Dinh, T.N., Theofanous, T.G.: Adaptive characteristics-based matching for compressible multifluid dynamics. J. Comput. Phys. 213, 500–529 (2006)CrossRefGoogle Scholar
9.
Johnsen, E., Colonius, T.: Implementation of WENO schemes in compressible multicomponent flow problems. J. Comput. Phys. 219, 715–732 (2006)MathSciNetCrossRefGoogle Scholar
10.
Hirt, C., Nichols, B.: Volume of fluid (VOF) method for the dynamics of free boundaries. J. Comput. Phys. 39, 201–225 (1981)CrossRefGoogle Scholar
11.
Sussman, M., Smereka, P., Osher, S.: A level set approach for computing solutions to incompressible two-phase flow. J. Comput. Phys. 114, 146–159 (1994)CrossRefGoogle Scholar
12.
Unverdi, S.O., Tryggvason, G.: A front-tracking method for viscous incompressible multi-fluid flows. J. Comput. Phys. 100, 25–37 (1992)CrossRefGoogle Scholar
13.
Glimm, J., Marchesin, D., McBryan, O.: Subgrid resolution of fluid discontinuities, II. J. Comput. Phys. 37, 336–354 (1980)MathSciNetCrossRefGoogle Scholar
14.
Glimm, J., Marchesin, D., McBryan, O.: A numerical method for two phase flow with an unstable interface. J. Comput. Phys. 39, 179–200 (1981)MathSciNetCrossRefGoogle Scholar
15.
Fedkiw, R.P., Aslam, T., Merriman, B., Osher, S.: A non-oscillatory Eulerian approach to interfaces in multimaterial flows (the ghost fluid method). J. Comput. Phys. 152, 457–492 (1999)MathSciNetCrossRefGoogle Scholar
16.
Fedkiw, R.P.: Coupling an Eulerian fluid calculation to a Lagrangian solid calculation with the ghost fluid method. J. Comput. Phys. 175, 200–224 (2002)CrossRefGoogle Scholar
17.
Liu, T.G., Khoo, B.C., Yeo, K.S.: Ghost fluid method for strong shock impacting on material interface. J. Comput. Phys. 190, 651–681 (2003)CrossRefGoogle Scholar
18.
Hu, X.Y., Khoo, B.C.: An interface interaction method for compressible multifluids. J. Comput. Phys. 198, 35–64 (2004)CrossRefGoogle Scholar
19.
Wang, C.W., Liu, T.G., Khoo, B.C.: A real-ghost fluid method for the simulation of multimedium compressible flow. SIAM J. Sci. Comput. 28, 278–302 (2006)MathSciNetCrossRefGoogle Scholar
20.
Liu, T.G., Khoo, B.C., Wang, C.W.: The ghost fluid method for compressible gas–water simulation. J. Comput. Phys. 204, 193–221 (2005)MathSciNetCrossRefGoogle Scholar
21.
Hao, Y., Prosperetti, A.: A numerical method for three-dimensional gas–liquid flow computations. J. Comput. Phys. 196, 126–144 (2004)MathSciNetCrossRefGoogle Scholar
22.
Farhat, C., Rallu, A., Shankaran, S.: A higher-order generalized ghost fluid method for the poor for the three-dimensional two-phase flow computation of underwater implosions. J. Comput. Phys. 227, 7674–7700 (2008)CrossRefGoogle Scholar
23.
Terashima, H., Tryggvason, G.: A front-tracking/ghost-fluid method for fluid interfaces in compressible flows. J. Comput. Phys. 228, 4012–4037 (2009)CrossRefGoogle Scholar
24.
Xu, L., Feng, C.L., Liu, T.G.: Practical techniques in ghost fluid method for compressible multi-medium flows. Commun. Comput. Phys. 20, 619–659 (2016)MathSciNetCrossRefGoogle Scholar
25.
Qiu, J.X., Liu, T.G., Khoo, B.C.: Simulations of compressible two-medium flow by Runge–Kutta discontinuous Galerkin methods with the ghost fluid method. Commun. Comput. Phys. 3, 479–504 (2008)MathSciNetzbMATHGoogle Scholar
26.
Liu, T.G., Xie, W.F., Khoo, B.C.: The modified ghost fluid method for coupling of fluid and structure constituted with hydro-elasto-plastic equation of state. SIAM J. Sci. Comput. 30, 1105–1130 (2008)MathSciNetCrossRefGoogle Scholar
27.
Liu, T.G., Khoo, B.C., Xie, W.F.: The modified ghost fluid method as applied to extreme fluid–structure interaction in the presence of cavitation. Commun. Comput. Phys. 1, 898–919 (2006)zbMATHGoogle Scholar
28.
Sambasivan, S., UdayKumar, H.S.: Ghost fluid method for strong shock interactions. Part 1: fluid–fluid interfaces. AIAA J. 47, 2907–2922 (2009)CrossRefGoogle Scholar
29.
Barton, P.T., Drikakis, D.: An Eulerian method for multi-component problems in non-linear elasticity with sliding interfaces. J. Comput. Phys. 229, 5518–5540 (2010)MathSciNetCrossRefGoogle Scholar
30.
Xu, L., Liu, T.G.: Modified ghost fluid method as applied to fluid–plate interaction. Adv. Appl. Math. Mech. 6(1), 24–48 (2014)MathSciNetCrossRefGoogle Scholar
31.
Gao, S., Liu, T.G.: 1D Exact elastic-perfectly plastic solid Riemann solver and its multi-material application. Adv. Appl. Math. Mech. 9(3), 621–650 (2017)MathSciNetCrossRefGoogle Scholar
32.
Gao, S., Liu, T.G., Yao, C.B.: A complete list of exact solution for one-dimensional elastic-perfectly plastic solid Riemann problem without vacuum. Commun. Nonlinear Sci. Numer. Simul. 63, 205–227 (2018)MathSciNetCrossRefGoogle Scholar
33.
Xu, L., Liu, T.G.: Optimal error estimation of the modified ghost fluid method. Commun. Comput. Phys. 8, 403–426 (2010)MathSciNetzbMATHGoogle Scholar
34.
Xu, L., Liu, T.G.: Accuracies and conservation errors of various ghost fluid methods for multimedium Riemann problem. J. Comput. Phys. 230, 4975–4990 (2011)MathSciNetCrossRefGoogle Scholar
35.
Ben-Artzi, M., Li, J.Q., Warnecke, G.: A direct Eulerian GRP scheme for compressible fluid flows. J. Comput. Phys. 218(1), 19–43 (2006)MathSciNetCrossRefGoogle Scholar
36.
Liu, T.G., Khoo, B.C., Yeo, K.S.: The simulation of compressible multi-medium flow. Part I: a new methodology with test applications to 1D gas–gas and gas–water cases. Comput. Fluids 30, 291–314 (2001)CrossRefGoogle Scholar
37.
Leer, B.V.: Towards the ultimate conservative difference scheme. V—a second-order sequel to Godunov’s method. J. Comput. Phys. 32(1), 101–136 (1979)CrossRefGoogle Scholar
38.
Courant, R., Friedrichs, K.O.: Supersonic Flows and Shock Waves, p. 424. Interscience Publishers, New York (1948)zbMATHGoogle Scholar
39.
Toro, E.F.: Riemann Solvers and Numerical Methods for Fluid Dynamics. Springer, Berlin (1999)CrossRefGoogle Scholar
40.
Li, J.Q., Liu, T.G., Sun, Z.F.: Implementation of the GRP scheme for computing radially symmetric compressible fluid flows. J. Comput. Phys. 228(16), 5867–5887 (2009)MathSciNetCrossRefGoogle Scholar

Copyright information

Authors and Affiliations

Tiegang Liu
- 1
Email author
Chengliang Feng
- 1
Liang Xu
- 2

1.LMIB and School of Mathematics and Systems ScienceBeijing University of Aeronautics and AstronauticsBeijingPeople’s Republic of China
2.China Academy of Aerospace AerodynamicsBeijingPeople’s Republic of China

	Left wave	Contact discontinuity	Right wave
Rarefaction wave	\( \int \limits _{{U_L}}^{{U_{1*}}} {d\psi } = 0 \)		\( \int \limits _{{U_R}}^{{U_{2*}}} {d\phi } = 0 \)
	\( S({\rho _{1}},{P_{1}}) = S({\rho _L},{P_L}) \)		\( S({\rho _{2}},{P_{2}}) = S({\rho _R},{P_R}) \)
Shock wave	\( {u_{1}} = {u_L} - H({P_{1}};{P_L},{\rho _L}) \)	\( {u_{1}} = {u_{2}} \)	\({u_{2}} = {u_R} + H({P_{2}};{P_R},{\rho _R}) \)
	\( {\rho _{1}} = G({P_{1}};{P_L},{\rho _L})\)	\({P_{1}} = {P_{2}} \)	\( {\rho _{2}} = G({P_{2}};{P_R},{\rho _R}) \)

Modified Ghost Fluid Method with Acceleration Correction (MGFM/AC)

Abstract

Keywords

Notes

Acknowledgements

References

Copyright information

Authors and Affiliations

Personalised recommendations

Cite article