Chaos control and numerical solution of time-varying fractional Newton-Leipnik system using fractional Atangana-Baleanu derivatives

Najat Almutairi; Sayed Saber; Najat Almutairi; Sayed Saber

doi:10.3934/math.20231319

AIMS Mathematics

2023, Volume 8, Issue 11: 25863-25887. doi: 10.3934/math.20231319

Previous Article Next Article

Research article

Chaos control and numerical solution of time-varying fractional Newton-Leipnik system using fractional Atangana-Baleanu derivatives

Najat Almutairi ^{1
,
,},
Sayed Saber ^2,3

1.
Department of Mathematics, College of Science, Qassim University, Buraidah, Saudi Arabia
2.
Department of Mathematics and Statistics, Faculty of Science, Beni-Suef University, Egypt
3.
Department of Mathematics, Faculty of Science and Arts in Baljurashi, Al-Baha University, Saudi Arabia

Received: 16 June 2023 Revised: 13 July 2023 Accepted: 24 July 2023 Published: 07 September 2023
MSC : 92D30, 92D25, 92C42, 34C60

Nonlinear fractional differential equations and chaotic systems can be modeled with variable-order differential operators. We propose a generalized numerical scheme to simulate variable-order fractional differential operators. Fractional calculus' fundamental theorem and Lagrange polynomial interpolation are used. Two methods, Atangana-Baleanu-Caputo and Atangana-Seda derivatives, were used to solve a chaotic Newton-Leipnik system problem with fractional operators. Our scheme examined the existence and uniqueness of the solution. We analyze the model qualitatively using its equivalent integral through an iterative convergence sequence. This novel method is illustrated with numerical examples. Simulated and analytical results agree. We contribute to real-world mathematical applications. Finally, we applied a numerical successive approximation method to solve the fractional model.

Keywords:

Citation: Najat Almutairi, Sayed Saber. Chaos control and numerical solution of time-varying fractional Newton-Leipnik system using fractional Atangana-Baleanu derivatives[J]. AIMS Mathematics, 2023, 8(11): 25863-25887. doi: 10.3934/math.20231319

Related Papers:

[1]	Hansaem Oh, Gwanghyun Jo . Physics-informed neural network for the heat equation under imperfect contact conditions and its error analysis. AIMS Mathematics, 2025, 10(4): 7920-7940. doi: 10.3934/math.2025364
[2]	Pengfei Zhu, Kai Liu . Numerical investigation of convergence in the $L^{\infty}$ norm for modified SGFEM applied to elliptic interface problems. AIMS Mathematics, 2024, 9(11): 31252-31273. doi: 10.3934/math.20241507
[3]	Hyun Geun Lee, Youngjin Hwang, Yunjae Nam, Sangkwon Kim, Junseok Kim . Benchmark problems for physics-informed neural networks: The Allen–Cahn equation. AIMS Mathematics, 2025, 10(3): 7319-7338. doi: 10.3934/math.2025335
[4]	Yu Xu, Youjun Deng, Dong Wei . Numerical solution of forward and inverse problems of heat conduction in multi-layered media. AIMS Mathematics, 2025, 10(3): 6144-6167. doi: 10.3934/math.2025280
[5]	Ruitao Liu, Wanshan Li . A new ADI-IIM scheme for solving two-dimensional wave equation with discontinuous coefficients. AIMS Mathematics, 2024, 9(11): 31180-31197. doi: 10.3934/math.20241503
[6]	Ahmed G. Salem, Turki D. Alharbi, Abdulaziz H. Alharbi, Anwar Ali Aldhafeeri . Impact of a spherical interface on a concentrical spherical droplet. AIMS Mathematics, 2024, 9(10): 28400-28420. doi: 10.3934/math.20241378
[7]	Dario Bambusi, Simone Paleari . A couple of BO equations as a normal form for the interface problem. AIMS Mathematics, 2024, 9(8): 23012-23026. doi: 10.3934/math.20241118
[8]	Min Wu, Jiali Zhou, Chaoyue Guan, Jing Niu . A numerical method using Legendre polynomials for solving two-point interface problems. AIMS Mathematics, 2025, 10(4): 7891-7905. doi: 10.3934/math.2025362
[9]	Xin Zhao, Xin Liu, Jian Li . Convergence analysis and error estimate of finite element method of a nonlinear fluid-structure interaction problem. AIMS Mathematics, 2020, 5(5): 5240-5260. doi: 10.3934/math.2020337
[10]	Maxime Krier, Julia Orlik . Solvability of a fluid-structure interaction problem with semigroup theory. AIMS Mathematics, 2023, 8(12): 29490-29516. doi: 10.3934/math.20231510

Abstract

1. Introduction

A class of problems that give rise of singular behavior are elliptic interface problems with imperfect contact ^[1], and it is featured by the implicit jump condition imposed on the imperfect contact interface, and the jumping quantity of the unknown is related to the flux across the interface. The elliptic interface problems with imperfect contact have been applied to model the Stefan problem of the solidification process and crystal growth, composite materials, multi-phase flows ^[2,3] and the problem of temperature discontinuity between the gas and cooling solid surface ^[4]. More examples include the heat conduction between materials of the different heat capacities and conductivities and interface diffusion processes ^[5,6], the temperature discontinuity between a gas and cooling solid surface ^[4], the conjugate heat transfer problem in thermodynamic processes between materials that are thermally coupled through non-adiabatic contacts ^[7], etc.

We consider the one-dimensional (1D) elliptic interface problem with imperfect contact

$\begin{equation} \begin{cases} & - ( \beta(x) u_{x})_x + c(x) u = f(x), \quad x\in (0, \alpha)\cup (\alpha, 1), \\ & u(0) = u_0, \quad u(1) = u_1, \end{cases} \end{equation}$

(1.1)

together with the following implicit jump conditions across the interface $\alpha$

$\begin{equation} \begin{cases} & [u] = u^{+} - u^{-} = \lambda \beta^{+} \nabla u^{+} \cdot \vec{n}, \\ & [\beta \frac{\partial u}{\partial \vec{n}}] = 0, \end{cases} \end{equation}$

(1.2)

where $0 < \alpha < 1$ , $\vec{n}$ is a unit normal to the interface pointing from $\Omega^{-}$ to $\Omega^{+}$ . $g^{\pm}$ denotes the right and left limits of the function $g$ at the point $\alpha$ , and $[g] = g^{+} - g^{-}$ . Without loss of generality, we assume that $\Omega$ is computational domain, and the interface $\alpha$ separates $\Omega$ into two sub-domains $\Omega^{-}$ and $\Omega^{+}$ .

The solution of the elliptic interface problems is often discontinuous due to discontinuous coefficients or singular sources across the interface. To recover the numerical accuracy near the interface, a variety of methods have been developed via enforcing the jump conditions (1.2) and (1.1) into numerical discretization such that accurate and robust numerical algorithms can be designed. Finite difference methods (FDMs) constitute a commonly used approach for elliptic interface problems. Since the publication of the original immersed interface method (IIM) ^[8], they have been applied to various problems, such as the Stokes flow with elastic boundaries or surface tension ^[9], incompressible flow based on the Navier-Stokes equations with singular source terms ^[10], and nonlinear problems in magneto-rheological fluids ^[11]. Many further developments and analysis in various aspects of the FDM for elliptic interface problems were carried out to improve the accuracy, stability or efficiency ^{[12,13,14,15,16,17,18,19]}. Many other elegant methods have been proposed in the past decade, including the ghost fluid method ^[20,21], finite volume method ^[7,22,23,24], the matched interface and boundary (MIB) method ^[25,26,27], etc. Finite element methods (FEMs) constitute another common practice to resolve the elliptic interface problems. With the help of body-fitted unstructured meshes, FEMs ^[28,29,30] have been developed for handling elliptic interfaces and irregular geometries. For interfaces with complex topologies, the construction of high quality body-fitted meshes could be difficult and time-consuming. This motivates the development of immersed FEMs based on non-body-fitted structured meshes ^{[31,32,33,34,35,36,37,38]}, etc. Nevertheless, most interface schemes in the literatures are designed to be of second-order accuracy.

The strategy for generating higher-order difference schemes can be roughly divided into two categories. (1) The first category is expanding the stencil and including more points in the schemes ^{[15,16,25,26,27,39,40,41,42]}. Advantages of this approaches is achieving high-order accuracy and using lower order jump conditions only. The obvious disadvantages of such approach is creating large matrix bandwidths and complicating the numerical treatment near the boundaries. Gibou and Fedkiw ^[16] introduced an $O(h^4)$ accurate finite difference discretization for the Laplace and heat equations on an irregular domain. Previous scholars ^[25,26,27] presented high-order MIB methods for solving elliptic equations with discontinuous coefficients and singular sources on Cartesian grids. This type of method is based on the use of fictitious points to achieve high-order accuracy. Zhong ^[40] presented a high-order IIM with general jump conditions by employing fictitious points to achieve high-order accuracy and using lower-order jump conditions only. Similar to the MIB approach, a wide stencil with more points are involved at irregular points. Feng and Zhao ^[42] introduced a new Cartesian grid FDM based on the fourth-order accurate MIB method. Colnago et al. ^[15] presented a high-order IIM for solving Poisson equations with discontinuous coefficients on Cartesian grids, it combines the FDM and ghost node strategy and requires only the ordinary jumps of the function. (2) The second category involves using the differential equation and the interface relations as additional identities ^[13,39,41]. The advantage of this approach is that the scheme is compact and minimal stencil is used. The disadvantage is that to maintain high-order accuracy, it requires the knowledge of jump conditions of high-order derivatives. Li and Ito ^[13] constructed a fourth-order IIM by computing high-order jump conditions involving mixed derivatives. Linnick and Fasel ^[39] presented a high-order IIM for simulating unsteady incompressible flow in an irregular domain. Instead of using analytical jump conditions, they compute the jump conditions for higher derivatives numerically. Angelova and Vulkov ^[41] presented high-order compact FDM for elliptic equations with intersecting interfaces by using the differential equation and the jump (interface) relations as additional identities, which can be differentiated to eliminate higher-order local truncation errors. For FEMs, the higher-order convergence crucially depends on how well the interface is resolved by the triangular mesh. In practice, subparametric, isoparametric or superparametric elements are usually employed to secure the optimal order of $p+1$ in the $L_2$ norm for a polynomial order of $p$ , for both continuous ^[33,43] and discontinuous Galerkin ^{[34,35,44,45,46]} FEMs. It is worth pointing out that most of the numerical methods in literatures are aimed at the elliptical interface problem with homogeneous jump conditions $([u] = 0, [\beta u_{\vec{n}}] = 0)$ or nonhomogeneous jump conditions where the jumps of the temperature as well as the conductive heat flux along the interface are known explicitly (say $[u] = g_1$ , $[\beta \partial u/ \partial \vec{n}] = g_2$ , with known $g_1$ and $g_2$ ). When the elliptic interface problem has an implicit jump condition as in (1.2), there are comparatively few numerical methods for solving such problems.

In this paper, we consider the 1D elliptic interface problems with imperfect contact, where the implicit jump condition is a setting for the interface and the jump of the primary variable is proportional to the normal flux across the interface. A class of high-order FDMs is constructed for the 1D elliptic interface for both the body-fitted and non-body-fitted mesh. For each case, the second-, third- and fourth-order approximations of the implicit jump condition are provided by using the original variables on both sides of the interface and a set of jump conditions and its high order derivatives. Numerical examples are presented to verify the performance of the scheme. The numerical results show that the presented schemes can reach the theoretical accuracy for solving the elliptic interface problems with imperfect contact.

The rest of this paper is organized as follows. In Section 2, we formulate the scheme for 1D elliptic equations with imperfect interfaces. Section 3 constructs the approximation of the normal derivative for the interface. Section 4 computes the primary variables $u^{+}$ and $u^{-}$ for the interface. Numerical examples are provided in Section 5 to demonstrate the accuracy and stability of the presented scheme. A concluding remark is given in Section 6.

2. High-order compact difference scheme

We assume that $\beta(x)$ in (1.1) is a piecewise smooth function with a jump at the interface $\alpha$ , and that $\beta(x)$ has upper and lower boundries:

$\begin{equation} 0 < \beta_{min} \leq \beta(x) \leq \beta_{max}, \end{equation}$

(2.1)

where $\beta_{min}$ and $\beta_{max}$ are two constants. The source term $f(x)$ is piecewise smooth.

Introduce a uniform grid $x_i = ih, i = 0, 1, \cdots, n$ with $h = \frac{1}{n}$ . The fourth-order compact difference scheme is constructed and three-point computational stencil for each point is adopted. For node $i$ , the three points in computational stencil are $\{x_{i-1}, x_i, x_{i+1}\}$ . Without loss of generality, assume the interface $\alpha$ is located in a grid interval, $x_k \leq \alpha \leq x_{k+1}$ . Because of the interface is located in the stencil, the points $k$ and $k+1$ are named as irregular points, and $i \ne k, k+1$ are name as regular points.

At a regular point $i$ , i.e., $i \ne k, k+1$ , for convenience, we briefly give the construction process of the fourth order compact difference scheme, for more detailed information one can refer to Ref.[47].

$\begin{equation} -\beta \delta^{2}_{x} u_i - \beta_{x} \delta_x u_i + c_{i} u_{i} + \tau = f_{i}, \end{equation}$

(2.2)

where

$\begin{equation*} \delta^{2}_x u_i = \frac{ u_{i+1} - 2 u_{i} + u_{i-1} }{h^2}, \quad \delta_{x} u_{i} = \frac{ u_{i+1} - u_{i-1} }{2h}, \end{equation*}$

and

$\begin{equation*} \tau = \beta_{x} \frac{h^2}{3!}\frac{\partial^3 u }{\partial x^3}\Big{|}_i + \beta \frac{h^2}{12} \frac{\partial^4 u}{x^4}\Big{|}_i + O(h^4). \end{equation*}$

Dropping the term $\tau$ in (2.2), it is the standard second-order central difference scheme. In order to get fourth-order scheme, we have to handle the high-order derivative terms $\frac{\partial^3 u }{\partial x^3}$ and $\frac{\partial^4 u}{\partial x^4}$ . According to the governing equation (1.1), there is

$\begin{equation} u_{xx} = \frac{1}{\beta} ( - \beta_x u_x + c u - f ). \end{equation}$

(2.3)

Taking the first and second derivatives to (2.3) respectively, we have

$\begin{equation} u_{xxx} = \phi_1 u_{xx} + \phi_2 u_{x} + \phi_3 u + \phi_4 f + \phi_5 f_x, \end{equation}$

(2.4)

and

$\begin{equation} u_{xxxx} = \psi_1 u_{xx} + \psi_2 u_{x} + \psi_3 u + \psi_4, \end{equation}$

(2.5)

where

$\begin{eqnarray*} & & \phi_1 = -\frac{\beta_x}{\beta}, \quad \phi_2 = \frac{\beta^2_x}{\beta^2} - \frac{\beta_{xx}}{\beta} + \frac{c}{\beta}, \quad \phi_3 = -\frac{c \beta_x}{\beta^2}, \quad \phi_4 = \frac{\beta_x}{\beta^2} , \quad \phi_5 = - \frac{1}{\beta}, \\ & & \psi_1 = \phi^2_1 + \phi_{1, x} + \phi_2, \quad \psi_2 = \phi_1 \phi_2 + \phi_{2, x} + \phi_3, \quad \psi_3 = \phi_1 \phi_3 + \phi_{3, x}, \\ & & \psi_4 = (\phi_1 \phi_4 + \phi_{4, x})f + (\phi_1 \phi_5 + \phi_4 + \phi_{5, x}) f_x + \phi_5 f_{xx}, \\ & & \phi_{1, x} = \frac{\beta^2_{x} - \beta \beta_{xx} }{\beta^2} , \quad \phi_{2, x} = 2 \frac{\beta_x}{\beta} \frac{\beta \beta_{xx} - \beta^2}{\beta^2} - \frac{ \beta \beta_{xxx} - \beta_x \beta_{xx} }{\beta^2} - \frac{c \beta_x}{\beta^2}, \\ & & \phi_{3, x} = \frac{c ( 2 \beta^2_{x} - \beta \beta_{xx} )}{\beta^3}, \quad \phi_{4, x} = \frac{ \beta \beta_{xx} - 2 \beta^2_{x} }{\beta^3}, \quad \phi_{5, x} = \frac{\beta_x}{\beta^2}. \end{eqnarray*}$

Substituting (2.4) and (2.5) into (2.2) and approximating the first- and second-derivatives with the central difference, we can get the fourth-order compact scheme:

$\begin{equation} A_i u_{i+1} + B_i u_{i} + C_{i} u_{i-1} = F_i + O(h^4), \end{equation}$

(2.6)

where

$\begin{eqnarray*} & & A_i = \frac{-\beta + \phi_1 \beta_x \frac{h^2}{6} + \psi_1 \beta \frac{h^2}{12} }{h^2} + \frac{ -\beta_x + \phi_2 \beta_x \frac{h^2}{6} + \psi_2 \beta \frac{h^2}{12} }{2h}, \\ & & B_i = 2 \frac{-\beta + \phi_1 \beta_x \frac{h^2}{6} + \psi_1 \beta \frac{h^2}{12} }{h^2} + ( c + \phi_3 \beta_x \frac{h^2}{6} + \psi_3 \beta \frac{h^2}{12} ), \\ & & C_i = \frac{-\beta + \phi_1 \beta_x \frac{h^2}{6} + \psi_1 \beta \frac{h^2}{12} }{h^2} - \frac{ -\beta_x + \phi_2 \beta_x \frac{h^2}{6} + \psi_2 \beta \frac{h^2}{12} }{2h}, \\ & & F_i = f_i - \beta_x \frac{h^2}{6} (\phi_4 f_i + \phi_5 f_{x, i}) - \beta \frac{h^2}{12} \psi_4. \end{eqnarray*}$

2.1. Interface fitted with mesh

When the interface is in contact with one of the mesh nodes, it is referred to as the body-fitted case. Without loss of generality, we assume that the interface is located the mesh node $k$ , as shown in Figure 1.

Figure 1. Interface fitted with body.

DownLoad: Full-Size Img PowerPoint

In this case, the irregular nodes are $k-1$ and $k+1$ , and the compact stencil of these two irregular nodes are shown in Figure {2}. Then, the high order compact scheme at these two irregular nodes can be given as

$\begin{equation} \bar{A}_{k-1} u_{k-2} + \bar{B}_{k-1} u_{k-1} + \bar{C}_{k-1} u^{-} = \bar{F}_{k-1}, \end{equation}$

(2.7)

and

$\begin{equation} \bar{A}_{k+1} u^{+} + \bar{B}_{k+1} u_{k+1} + \bar{C}_{k+1} u_{k+2} = \bar{F}_{k+1}. \end{equation}$

(2.8)

Apparently, there are two auxiliary qualities $u^{-}$ and $u^{+}$ in the schemes. To apply the above scheme, it is necessary to approximate the $u^{-}$ and $u^{+}$ with adequate accuracy. We prepare to use the interface jump connecting conditions and establish the linear systems with $u^{+}$ and $u^{-}$ as variables. The difficulty lies in the high-order discretization of the first-order derivatives $u^{+}_{x}$ and $u^{-}_{x}$ on both sides of the interface.

$\begin{equation} \begin{cases} [u] = \lambda \beta^{+} u^{+}_{x}, \\ [\beta u_{\vec{n}}] = 0. \end{cases} \end{equation}$

(2.9)

Figure 2. Irregular point stencil for interface-fitted mesh.

DownLoad: Full-Size Img PowerPoint

Expand the $u_{k+1}$ and $u_{k-1}$ at the interface $\alpha$ by using the Taylor series, it reads

$\begin{eqnarray} u_{k+1} & = & u^{+} + h \frac{\partial u}{\partial x}\Big{|}^{+} + \frac{ h^2}{2!} \frac{{\partial}^2 u}{\partial x^2}\Big{|}^{+} + \frac{ h^3}{3!} \frac{{\partial}^3 u}{\partial x^3}\Big{|}^{+} + \frac{ h^4}{4!} \frac{{\partial}^4 u}{\partial x^4}\Big{|}^{+} \\ &+& \frac{ h^5}{5!} \frac{{\partial}^5 u}{\partial x^5}\Big{|}^{+} + O( h^6), \end{eqnarray}$

(2.10)

and

$\begin{eqnarray} u_{k-1} & = & u^{-} - h \frac{\partial u}{\partial x}\Big|^{-} + \frac{ h^2}{2!} \frac{{\partial}^2 u}{\partial x^2}\Big|^{-} - \frac{ h^3}{3!} \frac{{\partial}^3 u}{\partial x^3}\Big|^{-} + \frac{ h^4}{4!} \frac{{\partial}^4 u}{\partial x^4}\Big|^{-} \\ &-& \frac{ h^5}{5!} \frac{{\partial}^5 u}{\partial x^5}\Big|^{-} + O( h^6). \end{eqnarray}$

(2.11)

Rewritting the above two formulas, we can get the representations of $\frac{\partial u}{\partial x}\Big{|}^{+}$ and $\frac{\partial u}{\partial x}\Big{|}^{-}$ as follows

$\begin{eqnarray} \frac{\partial u}{\partial x}\Big{|}^{+} & = & \frac{u_{k+1} - u^{+}}{h} - \frac{ h }{2!} \frac{{\partial}^2 u}{\partial x^2}\Big{|}^{+} - \frac{ h^2}{3!} \frac{{\partial}^3 u}{\partial x^3}\Big{|}^{+} - \frac{ h^3}{4!} \frac{{\partial}^4 u}{\partial x^4}\Big{|}^{+} \\ &-& \frac{ h^4}{5!} \frac{{\partial}^5 u}{\partial x^5}\Big{|}^{+} + O( h^5), \end{eqnarray}$

(2.12)

$\begin{eqnarray} \frac{\partial u}{\partial x}\Big{|}^{-} & = & \frac{u^{-} - u_{k-1} }{h} + \frac{ h }{2!} \frac{{\partial}^2 u}{\partial x^2}\Big{|}^{-} - \frac{ h^2}{3!} \frac{{\partial}^3 u}{\partial x^3}\Big{|}^{-} + \frac{ h^3}{4!} \frac{{\partial}^4 u}{\partial x^4}\Big{|}^{-} \\ &-& \frac{ h^4}{5!} \frac{{\partial}^5 u}{\partial x^5}\Big{|}^{-} + O( h^5). \end{eqnarray}$

(2.13)

According to the governing equation, we have

$\begin{equation} u^{+}_{xx} = \frac{1}{\beta^{+}} ( - \beta^{+}_x u^{+}_x + c^{+} u^{+} - f^{+} ), \quad u^{-}_{xx} = \frac{1}{\beta^{-}} ( - \beta^{-}_x u^{-}_x + c^{-} u^{-} - f^{-} ). \end{equation}$

(2.14)

Keeping the first two terms on the right-hand side of (2.12) and substituting (2.14) into (2.12), the second-order accurate approximate scheme of $\frac{\partial u}{\partial x}\Big{|}^{+}$ can be given as

$\begin{equation} \frac{\partial u}{\partial x}\Big{|}^{+} = \rho_{r, 1} u^{+} + \rho_{r, 2} u_{k+1} + \rho_{r, 3} + O(h^2), \end{equation}$

(2.15)

where

$\begin{eqnarray*} \rho_{r, 1} = -(\frac{1}{h} + \frac{h}{2}\frac{c^{+}}{\beta^{+}})\Big{/}D_r , \quad \rho_{r, 2} = \frac{1}{h} \Big{/}D_r , \quad \rho_{r, 3} = \frac{h}{2}\frac{f^{+}}{\beta^{+}}\Big{/}D_r, \quad D_r = (1-\frac{h}{2}\frac{\beta^{+}_{x}}{\beta^{+}}). \end{eqnarray*}$

To obtain the higher-order approximate of $\frac{\partial u}{\partial x}\Big{|}^{+}$ , it is necessary to handle the third- and fourth-order derivatives in (2.12). From (2.4) and (2.5), the expressions of $u^{+}_{xxx}$ and $u^{+}_{xxxx}$ can be given as

$\begin{equation} u^{+}_{xxx} = \Phi^{+}_1 u^{+}_{x} + \Phi^{+}_2 u^{+} + \Phi^{+}_3, \end{equation}$

(2.16)

and

$\begin{equation} u^{+}_{xxxx} = \Psi^{+}_1 u^{+}_{x} + \Psi^{+}_2 u^{+} + \Psi^{+}_3, \end{equation}$

(2.17)

where

$\begin{eqnarray*} & & \Phi^{+}_1 = - \frac{\beta^{+}_x}{\beta^{+}} \phi^{+}_1 + \phi^{+}_2, \quad \Phi^{+}_2 = \frac{c^{+}}{\beta^{+}} \phi^{+}_1 + \phi^{+}_3, \quad \Phi^{+}_3 = (-\frac{c^{+}}{\beta^{+}}\phi^{+}_1 + \phi^{+}_4) f^{+} + \phi^{+}_5 f^{+}_x, \\ & & \Psi^{+}_1 = - \frac{\beta^{+}_x}{\beta^{+}} \psi^{+}_1 + \psi^{+}_2, \quad \Psi^{+}_2 = \frac{c^{+}}{\beta^{+}} \psi^{+}_1 + \psi^{+}_3, \quad \Psi^{+}_3 = \frac{f^{+}}{\beta^{+}}\psi^{+}_1 + \psi^{+}_4. \end{eqnarray*}$

Keeping the first three terms on the right-hand side of (2.12), we establish the third-order accurate approximate of $\frac{\partial u}{\partial x}\Big{|}^{+}$ as follows:

$\begin{equation} \frac{\partial u}{\partial x}\Big{|}^{+} = \bar{\rho}_{r, 1} u^{+} + \bar{\rho}_{r, 2} u_{k+1} + \bar{\rho}_{r, 3} + O(h^3), \end{equation}$

(2.18)

where

$\begin{eqnarray*} \bar{\rho}_{r, 1} & = & -\left(\frac{1}{h} + \frac{h}{2} \frac{c^{+}}{\beta^{+}} + \frac{h^2}{6}\Phi^{+}_2 \right)\big{/}\bar{D}_r, \quad \bar{\rho}_{r, 2} = \frac{1}{h}\big{/} \bar{D}_r , \\ \bar{\rho}_{r, 3} & = & \left(\frac{h}{2}\frac{f^{+}}{\beta^{+}} - \frac{h^2}{6} \Phi^{+}_3 \right)\big{/}\bar{D}_r, \quad \bar{D}_r = 1 - \frac{h}{2} \frac{\beta^{+}_x}{\beta^{+}} + \frac{h^2}{6} \Phi^{+}_1. \end{eqnarray*}$

Further, the fourth-order approximate of $\frac{\partial u}{\partial x}\Big{|}^{+}$ can be given as

$\begin{equation} \frac{\partial u}{\partial x}\Big{|}^{+} = \bar{\bar{\rho}}_{r, 1} u^{+} + \bar{\bar{\rho}}_{r, 2} u_{k+1} + \bar{\bar{\rho}}_{r, 3} + O(h^4), \end{equation}$

(2.19)

where

$\begin{eqnarray*} \bar{\bar{\rho}}_{r, 1} & = & - \left(\frac{1}{h} + \frac{h}{2} \frac{c^{+}}{\beta^{+}} + \frac{h^2}{6} \Phi^{+}_2 + \frac{h^3}{24}\Psi^{+}_2 \right) \big{/} \bar{\bar{D}}_r, \quad \bar{\bar{\rho}}_{r, 2} = \frac{1}{h} \big{/} \bar{\bar{D}}_r , \\ \bar{\bar{\rho}}_{r, 3} & = & \left(\frac{h}{2}\frac{f^{+}}{\beta^{+}} + \frac{h^2}{6} \Phi^{+}_3 + \frac{h^3}{24} \Psi^{+}_3 \right)\big{/} \bar{\bar{D}}_r, \quad \bar{\bar{D}}_r = 1 - \frac{h}{2}\frac{\beta^{+}_x}{\beta^{+}} - \frac{h^2}{6} \Phi^{+}_1 + \frac{h^3}{24} \Psi^{+}_1. \end{eqnarray*}$

Similarly, the second-, third- and fourth-order approximation of $\frac{\partial u}{\partial x}\Big{|}^{-}$ can be given as

$\begin{equation} \frac{\partial u}{\partial x}\Big{|}^{-} = \rho_{l, 1} u^{-} + \rho_{l, 2} u_{k-1} + \rho_{l, 3} + O(h^2), \end{equation}$

(2.20)

$\begin{equation} \frac{\partial u}{\partial x}\Big{|}^{-} = \bar{\rho}_{l, 1} u^{-} + \bar{\rho}_{l, 2} u_{k-1} + \bar{\rho}_{l, 3} + O(h^3), \end{equation}$

(2.21)

$\begin{equation} \frac{\partial u}{\partial x}\Big{|}^{-} = \bar{\bar{\rho}}_{l, 1} u^{-} + \bar{\bar{\rho}}_{l, 2} u_{k-1} + \bar{\bar{\rho}}_{l, 3} + O(h^4), \end{equation}$

(2.22)

where

$\begin{eqnarray*} \rho_{l, 1} & = & (\frac{1}{h} + \frac{h}{2} \frac{c^{-}}{\beta^{-}} ) \Big{/} D_l, \quad \rho_{l, 2} = -\frac{1}{h} \Big{/} D_l, \quad \rho_{l, 3} = -\frac{h}{2}\frac{f^{-}}{\beta^{-}}\Big{/} D_l, \quad D_l = 1 + \frac{h}{2} \frac{\beta^{-}_x}{\beta^{-}} , \\ \bar{\rho}_{l, 1} & = & \left(\frac{1}{h} + \frac{h}{2} \frac{c^{-}}{\beta^{-}} - \frac{h^2}{6} \Phi^{-}_2 \right) \big{/} \bar{D}_l, \quad \bar{\rho}_{l, 2} = -\frac{1}{h} \big{/} \bar{D}_l, \\ \bar{\rho}_{l, 3} & = & \left( -\frac{h}{2}\frac{f^{-}}{\beta^{-}} - \frac{h^2}{6} \Phi^{-}_3 \right) \big{/} \bar{D}_l, \quad \bar{D}_l = 1 + \frac{h}{2} \frac{\beta^{-}_x}{\beta^{-}} + \frac{h^2}{6} \Phi^{-}_1 , \\ \bar{\bar{\rho}}_{l, 1} & = & \left(\frac{1}{h} + \frac{h}{2} \frac{c^{-}}{\beta^{-}} - \frac{h^2}{6} \Phi^{-}_2 + \frac{h^3}{24} \Psi^{-}_2 \right) \big{/} \bar{\bar{D}}_l, \quad \bar{\bar{\rho}}_{l, 2} = -\frac{1}{h} \big{/} \bar{\bar{D}}_l, \\ \bar{\bar{\rho}}_{l, 3} & = & \left(-\frac{h}{2}\frac{f^{-}}{\beta^{-}} + \frac{h^2}{6} \Phi^{-}_3 + \frac{h^3}{24} \Psi^{-}_3 \right) \big{/} \bar{\bar{D}}_l , \quad \bar{\bar{D}}_l = 1 + \frac{h}{2} \frac{\beta^{-}_x}{\beta^{-}} - \frac{h^2}{6} \Phi^{-}_1 + \frac{h^3}{24} \Psi^{-}_1. \end{eqnarray*}$

2.2. Interface unfitted with mesh

When the interface is located within one of the computational mesh, it appears as shown in . The irregular nodes are $k$ and $k+1$ , and the construction of the higher order scheme is more complicated. Owing to the interface arbitrarily cutting through the computational mesh, the computational mesh of these irregular nodes is nonuniform, as shown in Figure 4.

Figure 3. Interface unfitted with a body.

DownLoad: Full-Size Img PowerPoint

Figure 4. Nonuniform mesh.

DownLoad: Full-Size Img PowerPoint

We assume the function $u(x)$ is smooth enough, expand the $u(x_{i+1}), u(x_{i-1})$ at point $x_i$ by Taylor series

$\begin{eqnarray} u_{i+1} & = & u_{i} + h_f \frac{\partial u}{\partial x}\Big{|}_{i} + \frac{ h^2_f}{2!} \frac{{\partial}^2 u}{\partial x^2}\Big{|}_{i} + \frac{ h^3_f}{3!} \frac{{\partial}^3 u}{\partial x^3}\Big{|}_{i} + \frac{ h^4_f}{4!} \frac{{\partial}^4 u}{\partial x^4}\Big{|}_{i} \\ &+& \frac{h^5_f}{5!} \frac{{\partial}^5 u}{\partial x^5}\Big{|}_{i} + O(h^6_f), \end{eqnarray}$

(2.23)

and

$\begin{eqnarray} u_{i-1} & = & u_{i} - h_b \frac{\partial u}{\partial x}\Big|_{i} + \frac{ h^2_b}{2!} \frac{{\partial}^2 u}{\partial x^2}\Big|_{i} - \frac{ h^3_b}{3!} \frac{{\partial}^3 u}{\partial x^3}\Big|_{i} + \frac{ h^4_b}{4!} \frac{{\partial}^4 u}{\partial x^4}\Big|_{i} \\ &-& \frac{h^5_b}{5!} \frac{{\partial}^5 u}{\partial x^5}\Big|_{i} + O(h^6_b). \end{eqnarray}$

(2.24)

From (2.23) and (2.24), we can obtain

$\begin{eqnarray} \frac{{\partial}^2 u}{\partial x^2}\Big|_{i} & = & 2 \frac{ h_b u_{i+1} + h_f u_{i-1} -( h_f + h_b) u_i }{h_f h_b (h_f + h_b)} - \frac{h_f - h_b}{3} \frac{{\partial}^3 u}{\partial x^3}\Big|_{i} - \frac{1}{12} \frac{h^3_f + h^3_b }{h_f + h_b} \frac{{\partial}^4 u}{\partial x^4}\Big|_{i} \\ &-& \frac{1}{60}(h^2_f + h^2_b)(h_f + h_b) \frac{{\partial}^5 u}{\partial x^5}\Big|_{i} + O(\frac{ h^5_f + h^5_b }{h_f + h_b}), \end{eqnarray}$

(2.25)

and

$\begin{eqnarray} \frac{{\partial} u}{\partial x}\Big|_{i} & = & \frac{ h^2_b u_{i+1} - h^2_f u_{i-1} + (h^2_f - h^2_b) u_i }{h^2_b h_f + h^2_f h_b} - \frac{ h_f h_b}{6} \frac{{\partial}^3 u}{\partial x^3}\Big|_{i} - \frac{h_f h_b (h_f - h_b) }{24} \frac{{\partial}^4 u}{\partial x^4}\Big|_{i} \\ &+& O(\frac{ (h^3_f + h^3_b)h_f h_b }{ h_f + h_b }). \end{eqnarray}$

(2.26)

Define the difference operators

$\begin{equation} \delta_{x} u_i = \frac{ h^2_b u_{i+1} - h^2_f u_{i-1} + (h^2_f - h^2_b) u_i }{h^2_b h_f + h^2_f h_b}, \end{equation}$

(2.27)

$\begin{equation} \delta^2_{x} u_i = 2 \frac{ h_b u_{i+1} + h_f u_{i-1} -( h_f + h_b) u_i }{h_f h_b (h_f + h_b)}. \end{equation}$

(2.28)

Equations (2.25) and (2.26) can be rewritten as

$\begin{eqnarray} \frac{{\partial}^2 u}{\partial x^2}\Big|_{i} & = & \delta^2_{x} u_i - \frac{h_f - h_b}{3} \frac{{\partial}^3 u}{\partial x^3}\Big|_{i} - \frac{1}{12} \frac{h^3_f + h^3_b }{h_f + h_b} \frac{{\partial}^4 u}{\partial x^4}\Big|_{i} \\ &-& \frac{1}{60}(h^2_f + h^2_b)(h_f + h_b) \frac{{\partial}^5 u}{\partial x^5}\Big|_{i} + O(\frac{ h^5_f + h^5_b }{h_f + h_b}), \end{eqnarray}$

(2.29)

and

$\begin{eqnarray} \frac{{\partial} u}{\partial x}\Big|_{i} = \delta_{x} u_i - \frac{ h_f h_b}{6} \frac{{\partial}^3 u}{\partial x^3}\Big|_{i} - \frac{h_f h_b (h_f - h_b) }{24} \frac{{\partial}^4 u}{\partial x^4}\Big|_{i} + O(\frac{ (h^3_f + h^3_b)h_f h_b }{ h_f + h_b }). \end{eqnarray}$

(2.30)

Substituting Eqs (2.29) and (2.30) into the governing equation (1.1), we can get the difference scheme for the elliptic interface equation

$\begin{equation} - \beta \delta^2_{x} u_i - \beta_{x} \delta_{x} u_i + k_i u_i + \tau = f_i, \end{equation}$

(2.31)

where

$\begin{eqnarray} \tau = Q_1 \frac{{\partial}^3 u}{\partial x^3}\Big|_{i} + Q_2 \frac{{\partial}^4 u}{\partial x^4}\Big|_{i} + Q_3 \frac{{\partial}^5 u}{\partial x^5}\Big|_{i} + O(h^4), \end{eqnarray}$

(2.32)

and

$\begin{eqnarray*} Q_1 & = & \beta \frac{h_f - h_b}{3} - \beta_x \frac{h_f h_b}{6}, \quad Q_2 = \beta \frac{1}{12} \frac{h^3_f + h^3_b}{h_f + h_b} - \beta_x \frac{h_f h_b (h_f - h_b)}{24} , \\ Q_3 & = & \beta \frac{1}{60}(h^2_f + h^2_b)(h_f + h_b). \end{eqnarray*}$

To obtain the higher-order accuracy discrete scheme, it is necessary to handle the third- and fourth-order derivatives of the term $\tau$ . Substituting (2.4) and (2.5) into (2.31) and rearranging it, we have

$\begin{equation} \mathcal{A}_i \delta^2_{x} u_i + \mathcal{B} \delta_x u_i + \mathcal{C}_i u_i = \mathcal{F}_i + O((h^2_f + h^2_b)(h_f + h_b)), \end{equation}$

(2.33)

where

$\begin{eqnarray*} \mathcal{A}_i = -\beta + Q_1 \phi_1 + Q_2 \psi_1, \quad \mathcal{B}_i = -\beta_x + Q_1 \phi_2 + Q_2 \psi_2, \\ \mathcal{C}_i = c_i + Q_1 \phi_3 + Q_2 \psi_3, \quad \mathcal{F}_i = f_i - Q_1 (\phi_4 f_i + \phi_5 f_{x, i}) - Q_2 \psi_4. \end{eqnarray*}$

For the irregular points $k$ and $k+1$ , we use the interface $\alpha$ as one of the points in the compact stencil as shown in . Thus, the 3-point compact stencil for nodes $k$ and $k+1$ are $k-1, k, \alpha^{-}$ and $\alpha^{+}, k+1, k+2$ , respectively. From the above analysis, the difference operators for the first- and second-order derivatives for nodes $k$ and $k+1$ are respectively as follows

$\begin{equation} \delta_{x} u_k = \frac{ h^2_b u^{-} - h^2_f u_{k-1} + (h^2_f - h^2_b) u_k }{h^2_b h_f + h^2_f h_b}, \quad \delta^2_{x} u_k = 2 \frac{ h_b u^{-} + h_f u_{k-1} -( h_f + h_b) u_k }{h_f h_b (h_f + h_b)}, \end{equation}$

(2.34)

Figure 5. Stencil of interface unfitted with a body.

DownLoad: Full-Size Img PowerPoint

and

$\begin{equation} \delta_{x} u_{k+1} = \frac{ h^2_b u_{k+2} - h^2_f u^{+} + (h^2_f - h^2_b) u_{k+1} }{h^2_b h_f + h^2_f h_b}, \quad \delta^2_{x} u_{k+1} = 2 \frac{ h_b u_{k+2} + h_f u^{+} -( h_f + h_b) u_{k+1} }{h_f h_b (h_f + h_b)}. \end{equation}$

(2.35)

Substituting (2.34) and (2.35) into (2.33), we can get the higher-order compact scheme for the irregular point $k$ :

$\begin{equation} \bar{A}_k u_{k-1} + \bar{B}_k u_{k} + \bar{C}_k u^{-} = \bar{F}_k + O((h^2_f + h^2_b)(h_f + h_b)), \end{equation}$

(2.36)

where

$\begin{eqnarray*} &&\bar{A}_k = \frac{2}{h_b (h_f + h_b) } \mathcal{A}_k - \frac{h^2_f }{h^2_b h_f + h^2_f h_b } \mathcal{B}_k , \quad \bar{B}_k = - \frac{2}{h_f h_b } \mathcal{A}_k + \frac{ h^2_f - h^2_b }{h^2_b h_f + h^2_f h_b } \mathcal{B}_k + \mathcal{C}_k, \\ &&\bar{C}_k = \frac{2}{h_f (h_f + h_b)} \mathcal{A}_k + \frac{h^2_b}{h^2_b h_f + h^2_f h_b} \mathcal{B}_k , \quad \bar{F}_k = \mathcal{F}_k. \end{eqnarray*}$

Similarly, the higher-order compact scheme for the irregular point $k+1$ is as follows:

$\begin{equation} \bar{A}_{k+1} u^{+} + \bar{B}_{k+1} u_{k+1} + \bar{C}_{k+1} u_{k+2} = \bar{F}_{k+1}+ O((h^2_f + h^2_b)(h_f + h_b)), \end{equation}$

(2.37)

where

$\begin{eqnarray*} &&\bar{A}_{k+1} = \frac{2}{h_b (h_f + h_b)} \mathcal{A}_{k+1} - \frac{h^2_f}{h^2_b h_f + h^2_f h_b} \mathcal{B}_{k+1} , \\ &&\bar{B}_{k+1} = - \frac{2}{h_f h_b} \mathcal{A}_{k+1} + \frac{h^2_f - h^2_b}{h^2_b h_f + h^2_f h_b} \mathcal{B}_{k+1} + \mathcal{C}_{k+1}, \\ &&\bar{C}_{k+1} = \frac{2}{h_f (h_f + h_b)} \mathcal{A}_{k+1} + \frac{h^2_b}{h^2_b h_f + h^2_f h_b} \mathcal{B}_{k+1} , \quad \bar{F}_{k+1} = \mathcal{F}_{k+1}. \end{eqnarray*}$

From the above analysis, we have constructed the higher-order compact scheme at regular points and irregular points. It should be noted that there are two auxiliary unknown quantities $u^{+}$ and $u^{-}$ in the scheme for the irregular points $k$ and $k+1$ . Therefore, we have to find a way to deal with the interface connection conditions, so as to obtain a high-precision approximation for the auxiliary unknowns $u^{+}$ and $u^{-}$ on both sides of the interface.

3. Approximation of the normal derivative on the interface

The first order derivative on both sides of the interface, $\frac{\partial u}{\partial x}\Big{|}^{+}$ and $\frac{\partial u}{\partial x}\Big{|}^{-}$ , can respectively be given as

$\begin{eqnarray} \frac{\partial u}{\partial x}\Big{|}^{+} & = & \frac{u_{k+1} - u^{+}}{h_{r1}} - \frac{ h_{r1} }{2!} \frac{{\partial}^2 u}{\partial x^2}\Big{|}^{+} - \frac{ h^2_{r1}}{3!} \frac{{\partial}^3 u}{\partial x^3}\Big{|}^{+} - \frac{ h^3_{r1}}{4!} \frac{{\partial}^4 u}{\partial x^4}\Big{|}^{+} \\ &-& \frac{ h^4_{r1}}{5!} \frac{{\partial}^5 u}{\partial x^5}\Big{|}^{+} + O( h^5_r), \end{eqnarray}$

(3.1)

and

$\begin{eqnarray} \frac{\partial u}{\partial x}\Big{|}^{-} & = & \frac{u^{-} - u_{k} }{h_{l1}} + \frac{ h_{l1} }{2!} \frac{{\partial}^2 u}{\partial x^2}\Big{|}^{-} - \frac{ h^2_{l1}}{3!} \frac{{\partial}^3 u}{\partial x^3}\Big{|}^{-} + \frac{ h^3_{l1}}{4!} \frac{{\partial}^4 u}{\partial x^4}\Big{|}^{-} \\ &-& \frac{ h^4_{l1}}{5!} \frac{{\partial}^5 u}{\partial x^5}\Big{|}^{-} + O( h^5_l), \end{eqnarray}$

(3.2)

where $h_{l1} = \alpha - x_k$ , $h_{r1} = x_{k+1} - \alpha$ , $h_{l2} = h_{l1} + h$ and $h_{r2} = h_{r1} + h$ , as shown in Figure 6.

Figure 6. Interface and the irregular nodes.

DownLoad: Full-Size Img PowerPoint

Further, the second-order derivative on both sides of the interface, $\frac{\partial^2 u}{\partial x^2}\Big{|}^{+}$ and $\frac{\partial^2 u}{\partial x^2}\Big{|}^{-}$ , can respectively be given as

$\begin{eqnarray} \frac{\partial^2 u}{\partial x^2}\Big{|}^{+} & = & \frac{-2}{h_{r1} h_{r2} h} \left(h_{r2}u_{k+1} - h_{r1} u_{k+2} - h u^{+} \right) - \frac{ h_{r1} + h_{r2}}{3} \frac{{\partial}^3 u}{\partial x^3}\Big{|}^{+} \\ &-& \frac{h^2_{r1} + h_{r1}h_{r2} + h^{2}_{r2}}{12} \frac{{\partial}^4 u}{\partial x^4}\Big{|}^{+} + O( h^3_{r1}+ h^{2}_{r2} h_{r1} + h^{2}_{r1} h_{r2} + h^{3}_{r2}), \end{eqnarray}$

(3.3)

and

$\begin{eqnarray} \frac{\partial^2 u}{\partial x^2}\Big{|}^{-} & = & \frac{-2}{h_{l1} h_{l2} h} \left( h_{l2} u_{k} - h_{l1} u_{k-1} - h u^{-} \right) + \frac{ h_{l1} + h_{l2}}{3} \frac{{\partial}^3 u}{\partial x^3}\Big{|}^{-} \\ &-& \frac{h^2_{l1} + h_{l1}h_{l2} + h^{2}_{l2}}{12} \frac{{\partial}^4 u}{\partial x^4}\Big{|}^{-} + O( h^3_{l1}+h^{2}_{l2}h_{l1}+h^{2}_{l1}h_{l2}+h^{3}_{l2}). \end{eqnarray}$

(3.4)

Substituting (3.3) and (3.4) into (3.1) and (3.2) respectively, we can get

$\begin{eqnarray} \frac{\partial u}{\partial x}\Big{|}^{+} & = & \big\{ -(\frac{1}{h_{r1}} + \frac{1}{h_{r2}}) u^{+} + ( \frac{1}{h_{r1}} + \frac{1}{h}) u_{k+1} - \frac{h_{r1}}{h_{r2} h} u_{k+2} \big\} + \frac{ h_{r1}h_{r2}}{3!} \frac{{\partial}^3 u}{\partial x^3}\Big{|}^{+} \\ &+& \frac{h^2_{r1}h_{r2} + h^{2}_{r2} h_{r1}}{4!} \frac{{\partial}^4 u}{\partial x^4}\Big{|}^{+} + O( h^3_{r1}+h^{2}_{r2}h_{r1}+h^{2}_{r1}h_{r2}+h^{3}_{r2}), \end{eqnarray}$

(3.5)

and

$\begin{eqnarray} \frac{\partial u}{\partial x}\Big{|}^{-} & = & \big\{ (\frac{1}{h_{l1}} + \frac{1}{h_{l2}}) u^{-} - (\frac{1}{h_{l1}} + \frac{1}{h}) u_{k} + \frac{h_{l1}}{h_{l2} h} u_{k-1} \big\} + \frac{ h_{l1}h_{l2}}{3!} \frac{{\partial}^3 u}{\partial x^3}\Big{|}^{+} \\ &-& \frac{h^2_{l1}h_{l2} + h^{2}_{l2} h_{l1}}{4!} \frac{{\partial}^4 u}{\partial x^4}\Big{|}^{+} + O( h^3_{l1}+h^{2}_{l2}h_{l1}+h^{2}_{l1}h_{l2}+h^{3}_{l2}). \end{eqnarray}$

(3.6)

3.1. Second-order scheme

Keep the first term in (3.6) and (3.5) and discard the other terms. We can get the approximation of the first order derivative on both sides of the interface with second-order accuracy:

$\begin{equation} u^{-}_{x} = \rho_{l, 1} u^{-} + \rho_{l, 2} u_{k} + \rho_{l, 3}u_{k-1} + O(h^2), \end{equation}$

(3.7)

and

$\begin{equation} u^{+}_{x} = \rho_{r, 1} u^{+} + \rho_{r, 2} u^{k+1} + \rho_{r, 3}u_{k+2} + O(h^2), \end{equation}$

(3.8)

where

$\begin{eqnarray*} \rho_{l, 1} = \frac{1}{h_{l1}} + \frac{1}{h_{l2}} , \quad \rho_{l, 2} = - (\frac{1}{h_{l1}} + \frac{1}{h}), \quad \rho_{l, 3} = \frac{h_{l1}}{h h_{l2}}, \\ \rho_{r, 1} = -(\frac{1}{h_{r1}} + \frac{1}{h_{r2}}), \quad \rho_{r, 2} = \frac{1}{h_{r1}} + \frac{1}{h}, \quad \rho_{r, 3} = -\frac{h_{r1}}{h h_{r2}}. \end{eqnarray*}$

3.2. Third-order scheme

Keeping the first two terms in (3.6) and (3.5) and discarding the other terms, and according to the (2.16), we can get the approximation of the first derivative with third order accuracy

$\begin{equation} u^{-}_{x} = \bar{\rho}_{l, 1} u^{-} + \bar{\rho}_{l, 2} u_{k} + \bar{\rho}_{l, 3} u_{k-1} + \bar{\rho}_{l, 4} + O(h^3), \end{equation}$

(3.9)

and

$\begin{equation} u^{+}_{x} = \bar{\rho}_{r, 1} u^{+} + \bar{\rho}_{r, 2} u_{k+1} + \bar{\rho}_{r, 3} u_{k+2} + \bar{\rho}_{r, 4} + O(h^3), \end{equation}$

(3.10)

where

$\begin{eqnarray*} & & \bar{\rho}_{l, 1} = \left(\rho_{1, 1} + \frac{h_{l1} h_{l2}}{6} \Phi^{-}_2 \right) \big/\bar{D}_l, \quad \bar{\rho}_{l, 2} = \rho_{1, 2} \big/\bar{D}_l , \quad \bar{\rho}_{l, 3} = \rho_{1, 3}\big/\bar{D}_l, \\ & & \bar{\rho}_{l, 4} = \frac{h_{l1} h_{l2}}{6} \Phi^{-}_3 \big/\bar{D}_l , \quad \bar{D}_l = 1 - \frac{h_{l1} h_{l2}}{6} \Phi^{-}_1 , \\ & & \bar{\rho}_{r, 1} = \left(\rho_{r, 1} + \frac{h_{r1} h_{r2}}{6} \Phi^{+}_2 \right) \big/\bar{D}_r, \quad \bar{\rho}_{r, 2} = \rho_{r, 2}\big/ \bar{D}_r , \quad \bar{\rho}_{r, 3} = \rho_{r, 3}\big/\bar{D}_r , \\ & & \bar{\rho}_{l, 4} = \frac{h_{r1} h_{r2}}{6} \Phi^{+}_3 \big/\bar{D}_r, \quad \bar{D}_r = 1 - \frac{h_{r1} h_{r2}}{6} \Phi^{+}_1. \end{eqnarray*}$

3.3. Fourth order scheme

Following the same manner as in the above subsection, we can get the approximation of the first derivative with fourth-order accuracy

$\begin{equation} u^{-}_{x} = \bar{\bar{\rho}}_{l, 1} u^{-} + \bar{\bar{\rho}}_{l, 2} u_{k} + \bar{\bar{\rho}}_{l, 3} u_{k-1} + \bar{\bar{\rho}}_{l, 4} + O(h^4), \end{equation}$

(3.11)

and

$\begin{equation} u^{+}_{x} = \bar{\bar{\rho}}_{r, 1} u^{+} + \bar{\bar{\rho}}_{r, 2} u_{k+1} + \bar{\bar{\rho}}_{r, 3} u_{k+2} + \bar{\bar{\rho}}_{r, 4}+ O(h^4), \end{equation}$

(3.12)

where

$\begin{eqnarray*} & & \bar{\bar{\rho}}_{l, 1} = \left( \rho_{l, 1} + \frac{h_{l1} h_{l2}}{6} \Phi^{-}_2 + \frac{h^{2}_{l2} h_{l1} + h^{2}_{l1} h_{l2}}{24} \Psi^{-}_2 \right)\big/\bar{\bar{D}}_l, \quad \bar{\bar{\rho}}_{l, 2} = \rho_{l, 2} \big/\bar{\bar{D}}_l, \\ & & \bar{\bar{\rho}}_{l, 3} = \rho_{l, 3} \big/\bar{\bar{D}}_l, \quad \bar{\bar{\rho}}_{l, 4} = \left( \frac{h_{l1} h_{l2}}{6} \Phi^{-}_3 + \frac{h^{2}_{l2} h_{l1} + h^{2}_{l1} h_{l2}}{24} \Psi^{-}_3 \right)\big/ \bar{\bar{D}}_l, \\ & & \bar{\bar{\rho}}_{r, 1} = \left( \rho_{r, 1} - \frac{h_{l1} h_{l2}}{6} \Phi^{-}_1 - \frac{h^{2}_{l2} h_{l1} + h^{2}_{l1} h_{l2}}{24} \Psi^{-}_1 \right)\big/\bar{\bar{D}}_r, \quad \bar{\bar{\rho}}_{r, 2} = \rho_{r, 2} \big/\bar{\bar{D}}_r , \\ & & \bar{\bar{\rho}}_{r, 3} = \rho_{r, 3} \big/\bar{\bar{D}}_r, \quad \bar{\bar{\rho}}_{r, 4} = \left( \frac{h_{r1} h_{r2}}{6} \Phi^{+}_3 - \frac{h^{2}_{r2} h_{r1} + h^{2}_{r1} h_{r2}}{24} \Psi^{+}_3 \right)\big/\bar{\bar{D}}_r, \\ & & \bar{\bar{D}}_l = 1 - \frac{h_{l1} h_{l2}}{6} \Phi^{-}_1 - \frac{h^{2}_{r2} h_{r1} + h^{2}_{r1} h_{r2}}{24} \Psi^{+}_1 , \\ & & \bar{\bar{D}}_r = 1 - \frac{h_{r1} h_{r2}}{6} \Phi^{+}_1 - \frac{h^{2}_{r2} h_{r1} + h^{2}_{r1} h_{r2}}{24} \Psi^{+}_1 . \end{eqnarray*}$

4. Compute $u^{+}$ and $u^{-}$ on the interface

The implicit connecting condition on the interface is given as

$\begin{equation} \begin{cases} [u] = \lambda \beta^{-} u^{-}_{x}, \\ \beta^{+}u^{+}_{x} = \beta^{-} u^{-}_{x}. \end{cases} \end{equation}$

(4.1)

For the interface fitted mesh, the approximation of the first-order derivative on both sides of the interface can respectively be written as

$\begin{eqnarray} & & {\frac{\partial u}{\partial x}}\big{|}^{-} = \rho_{l, 1} u^{-} + \rho_{l, 2} u_{k-1} + \rho_{l, 3} + O(h^{m}_l), \end{eqnarray}$

(4.2)

$\begin{eqnarray} & & {\frac{\partial u}{\partial x}}\big{|}^{+} = \rho_{r, 1} u^{+} + \rho_{r, 2} u_{k+1} + \rho_{r, 3} + O(h^{m}_r). \end{eqnarray}$

(4.3)

Discarding the terms $O(h^{m}_l)$ and $O(h^{m}_r)$ in (4.2) and (4.3), and substituting them into (4.1), the discrete format of the implicit connecting condition on the interface can be given as

$\begin{equation*} \label{jumpcond3} \begin{cases} u^{+} - u^{-} = \lambda \beta^{-} (\rho_{l, 1} u^{-} + \rho_{l, 2} u_{k-1} + \rho_{l, 3}) , \\ \beta^{+}(\rho_{r, 1} u^{+} + \rho_{r, 2} u_{k+1} + \rho_{r, 3}) = \beta^{-} (\rho_{l, 1} u^{-} + \rho_{l, 2} u_{k-1} + \rho_{l, 3}) . \end{cases} \end{equation*}$

For brevity, rearrange the above equation and establish linear equations with $u^{+}$ and $u^{-}$ as variables as follows:

$\begin{equation*} \begin{cases} a_{11} u^{+} - a_{12} u^{-} = b_{1}, \\ a_{21} u^{+} + a_{22} u^{-} = b_{2}, \end{cases} \end{equation*}$

where

$\begin{eqnarray*} && a_{11} = 1, \quad a_{12} = 1 + \lambda {\beta}^{-} \rho_{l, 1}, \quad a_{21} = {\beta}^{+} \rho_{r, 1}, \quad a_{22} = -{\beta}^{-} \rho_{l, 1}. \\ && b_1 = \lambda {\beta}^{-} (\rho_{l, 2} u_{k-1} + \rho_{l, 3} ), \\ && b_2 = {\beta}^{-} (\rho_{l, 2} u_{k-1} + \rho_{l, 3} ) - {\beta}^{+} (\rho_{r, 2} u_{k+1} + \rho_{r, 3}). \end{eqnarray*}$

Solving the above linear system, we can get the approximate of each of the unknowns $u^{+}$ and $u^{-}$ on both sides of the interface with $m$ -th order accuracy:

$\begin{eqnarray} u^{+} & = & \tau^{+}_1 u_{k-1} + \tau^{+}_2 u_{k+1} + \tau^{+}_3, \end{eqnarray}$

(4.4)

$\begin{eqnarray} u^{-} & = & \tau^{-}_1 u_{k-1} + \tau^{-}_2 u_{k+1} + \tau^{-}_3, \end{eqnarray}$

(4.5)

where

$\begin{eqnarray*} &&\tau^{+}_{1} = \frac{ (a_{22} \lambda + a_{12})\beta^{-}\rho_{l, 2} }{a_{11}a_{22} + a_{12}a_{21}}, \quad \tau^{+}_{2} = -\frac{ a_{12} \beta^{+} \rho_{r, 2} }{a_{11}a_{22} + a_{12}a_{21}}, \\ &&\tau^{+}_{3} = \frac{ (a_{22} \lambda + a_{12})\beta^{-}\rho_{l, 3} - a_{12} \beta^{+} \rho_{r, 3} }{a_{11}a_{22} + a_{12}a_{21}} , \\ &&\tau^{-}_{1} = \frac{ (a_{11} - a_{21} \lambda)\beta^{-}\rho_{l, 2} }{a_{11}a_{22} + a_{12}a_{21}}, \quad \tau^{-}_{2} = -\frac{ a_{11} \beta^{+}\rho_{r, 2} }{a_{11}a_{22} + a_{12}a_{21}}, \\ &&\tau^{-}_{3} = \frac{ (a_{11} - a_{21} \lambda)\beta^{-}\rho_{l, 3} - a_{11} \beta^{+} \rho_{r, 3} }{a_{11}a_{22} + a_{12}a_{21}}. \end{eqnarray*}$

For the interface-cut mesh, the approximation of the first-order derivative on both sides of the interface can be written as

$\begin{eqnarray} & & {\frac{\partial u}{\partial x}}\big{|}^{-} = \rho_{l, 1} u^{-} + \rho_{l, 2} u_{k} + \rho_{l, 3} u_{k-1} + \rho_{l, 4} + O(h^{m}_l), \end{eqnarray}$

(4.6)

$\begin{eqnarray} & & {\frac{\partial u}{\partial x}}\big{|}^{+} = \rho_{r, 1} u^{+} + \rho_{r, 2} u_{k+1} + \rho_{r, 3}u_{k+2} + \rho_{r, 4} + O(h^{m}_r). \end{eqnarray}$

(4.7)

In a similar manner as above, the approximate of each of the unknowns $u^{+}$ and $u^{-}$ on both sides of the interface with $m$ -th-order accuracy is given as

$\begin{eqnarray} u^{+} & = & \tau^{+}_1 u_{k-1} + \tau^{+}_2 u_{k} + \tau^{+}_3 u_{k+1} + \tau^{+}_4 u_{k+2} + \tau^{+}_5, \end{eqnarray}$

(4.8)

$\begin{eqnarray} u^{-} & = & \tau^{-}_1 u_{k-1} + \tau^{-}_2 u_{k} + \tau^{-}_3 u_{k+1} + \tau^{-}_4 u_{k+2} + \tau^{-}_5, \end{eqnarray}$

(4.9)

where

$\begin{eqnarray*} &&\tau^{+}_{1} = \frac{ (a_{22} \lambda + a_{12})\beta^{-}\rho_{l, 3} }{a_{11}a_{22} + a_{12}a_{21}}, \quad \tau^{+}_{2} = \frac{ (a_{22} \lambda + a_{12})\beta^{-}\rho_{l, 2} }{a_{11}a_{22} + a_{12}a_{21}}, \quad \tau^{+}_{3} = \frac{ - a_{12} \beta^{+}\rho_{r, 2} }{a_{11}a_{22} + a_{12}a_{21}}, \\ &&\tau^{+}_{4} = \frac{ - a_{12} \beta^{+}\rho_{r, 3} }{a_{11}a_{22} + a_{12}a_{21}}, \quad \tau^{+}_{5} = \frac{ a_{22} \lambda \beta^{-} \rho_{l, 4} + a_{12} ( \beta^{-} \rho_{l, 4} - \beta^{+} \rho_{r, 4} ) }{a_{11}a_{22} + a_{12}a_{21}}, \\ &&\tau^{-}_{1} = \frac{ (a_{11} - a_{21} \lambda)\beta^{-}\rho_{l, 3} }{a_{11}a_{22} + a_{12}a_{21}}, \quad \tau^{-}_{2} = \frac{ (a_{11} - a_{21} \lambda)\beta^{-}\rho_{l, 2} }{a_{11}a_{22} + a_{12}a_{21}}, \quad \tau^{-}_{3} = \frac{ - a_{11} \beta^{+} \rho_{r, 2} }{a_{11}a_{22} + a_{12}a_{21}}, \\ &&\tau^{-}_{4} = \frac{ - a_{11} \beta^{+} \rho_{r, 3} }{a_{11}a_{22} + a_{12}a_{21}}, \quad \tau^{-}_{5} = \frac{ a_{11}( \beta^{-}\rho_{l, 4} - \beta^{+}\rho_{r, 4}) - a_{21} \lambda \beta^{-}\rho_{l, 4} }{a_{11}a_{22} + a_{12}a_{21}}. \end{eqnarray*}$

and

$\begin{eqnarray*} && a_{11} = 1, \quad a_{12} = 1 + \lambda {\beta}^{-} \rho_{l, 1}, \quad a_{21} = {\beta}^{+} \rho_{r, 1}, \quad a_{22} = -{\beta}^{-} \rho_{l, 1}. \\ && b_1 = \lambda {\beta}^{-} (\rho_{l, 2} u_{k} + \rho_{l, 3} u_{k-1} + \rho_{l, 4} ), \\ && b_2 = {\beta}^{-} (\rho_{l, 2} u_{k} + \rho_{l, 3} u_{k-1} + \rho_{l, 4} ) - {\beta}^{+} (\rho_{r, 2} u_{k+1} + \rho_{r, 3}u_{k+2} + \rho_{r, 4}). \end{eqnarray*}$

5. Discrete system

From the above sections, we establish the fourth-order compact finite-difference scheme for non-irregular mesh nodes:

$\begin{equation} A_i u_{i-1} + B_{i} u_i + C_{i} u_{i+1} = F_{i}, \end{equation}$

(5.1)

for

$\begin{equation*} i \neq \begin{cases} \{k-1, k+1 \}, \quad \quad {\rm{for\; body-fitted\; mesh, }} \\ \{k, k+1\}, \qquad \qquad{\rm{for \;non-body-fitted\; mesh}}. \end{cases} \end{equation*}$

In the interface-fitted mesh case, the higher-order compact schemes at the irregular mesh nodes $k-1$ and $k+1$ are given as (2.7) and (2.8), respectively. Eliminating the auxiliary qualities $u^{+}$ and $u^{-}$ by substituting the formulas (4.4) and (4.5) into (2.7) and (2.8). The schemes for the irregular mesh nodes are given as

$\begin{equation} A_{k-1} u_{k-2} + B_{k-1} u_{k-1} + C_{k-1} u_{k+1} = F_{k-1}, \end{equation}$

(5.2)

and

$\begin{equation} A_{k+1} u_{k-1} + B_{k+1} u_{k+1} + C_{k+1} u_{k+2} = F_{k+1}, \end{equation}$

(5.3)

where

$\begin{eqnarray*} & & A_{k-1} = \bar{A}_{k-1} , \quad B_{k-1} = \bar{B}_{k-1} + \bar{C}_{k-1} \tau^{-}_1, \quad C_{k-1} = \bar{C}_{k-1} \tau^{-}_2, \\ & & A_{k+1} = \bar{A}_{k+1} \tau^{+}_1 , \quad B_{k+1} = \bar{B}_{k+1} + \bar{A}_{k+1} \tau^{+}_2 , \quad C_{k+1} = \bar{C}_{k+1}, \\ & & F_{k-1} = \bar{F}_{k-1} - \bar{C}_{k-1} \tau^{-}_3, \quad F_{k+1} = \bar{F}_{k+1} - \bar{A}_{k+1} \tau^{+}_3 . \end{eqnarray*}$

Similarly, in the non-body-fitted case, the higher-order compact schemes at the irregular mesh nodes $k$ and $k+1$ are respectively given as (2.36) and (2.37). Substitute the formulas (4.8) and (4.9) into (2.36) and (2.37) to eliminate the auxiliary qualities $u^{+}$ and $u^{-}$ . The higher-order compact schemes at the irregular mesh nodes $k$ and $k+1$ are respectively given as

$\begin{equation} A_{k} u_{k-1} + B_{k} u_{k} + C_{k} u_{k+1} + D_{k} u_{k+2} = F_{k} , \end{equation}$

(5.4)

and

$\begin{equation} A_{k+1} u_{k-1} + B_{k+1}u_{k} + C_{k+1} u_{k+1} + D_{k+1} u_{k+2} = F_{k+1}, \end{equation}$

(5.5)

where

$\begin{eqnarray*} & & A_{k} = \bar{A}_{k} + \bar{C} \tau^{-}_1, \quad B_{k} = \bar{B}_{k} + \bar{C} \tau^{-}_2, \quad C_{k} = \bar{C}_{k} \tau^{-}_3, \quad D_{k} = \bar{C}_{k} \tau^{-}_4, \\ & & A_{k+1} = \bar{A}_{k+1} \tau^{+}_1, \quad B_{k+1} = \bar{A}_{k+1} \tau^{+}_2, \quad C_{k+1} = \bar{B}_{k+1} + \bar{A}_{k+1} \tau^{+}_3, \\ & & D_{k+1} = \bar{C}_{k+1} + \bar{A}_{k+1} \tau^{+}_4, \quad F_{k} = \bar{F}_{k} - \bar{C} \tau^{-}_5, \quad F_{k+1} = \bar{F}_{k+1} - \bar{A}_{k+1} \tau^{+}_5. \end{eqnarray*}$

Obviously, the resulting discrete linear equations have a tri-diagonal form for the body-fitted case and block-diagonal form for the non-body-fitted case, which can be efficiently solved by using existing numerical methods, such as the forward and backward sweep method, BiCGstab method, etc.

6. Numerical experiments

In this section, we use several numerical experiments to demonstrate the performance of the discrete schemes.

Example 1. Consider the computational domain $\Omega = [0, 1]$ , and the solution is separated into two parts by the interface at $x = \alpha$ , where $\alpha = 0.5$ and $\alpha = 0.52323$ , for the interface-fitted mesh and interface-cut mesh cases, respectively. The analytical solution of this problem is given by

$\begin{equation*} u(x, y) = \begin{cases} e^{x^2} , & \qquad x \in (0, \alpha), \\ \kappa e^{x^2}, & \qquad x \in (\alpha, 1). \end{cases} \end{equation*}$

The diffusion coefficient is defined as follows

$\begin{equation*} \beta = \begin{cases} \kappa , & \qquad x \in (0, \alpha), \\ 1, & \qquad x \in (\alpha, 1). \end{cases} \end{equation*}$

Conservation of the flux on the interface satisfies

$\begin{equation*} [\beta \frac{\partial u}{\partial x}] = \beta^{+}\frac{\partial u^{+}}{\partial x} - \beta^{-} \frac{\partial u^{-}}{\partial x} = 2 \kappa x e^{x^2} - 2 \kappa x e^{x^2} = 0, \quad at \quad x = \alpha. \end{equation*}$

The coefficient $\lambda$ is given as

$\begin{equation*} \lambda = \frac{\kappa e^{\alpha^2}-\alpha^2}{2\kappa \alpha e^{\alpha^2}}. \end{equation*}$

and compare the $L_{2}$ and $L_{\infty}$ errors in second-, third- and fourth-order formats on fitted and non-fitted mesh, respectively. The tables show that the numerical results of the three methods achieve theoretical accuracy on both fitted and non-fitted mesh.

Table 1. Comparison of

$L_{2}$ and

$L_{\infty}$ errors in second-, third- and fourth-order formats on fitted mesh for Example 1,

$\kappa$ = 100.

Mesh	Second-order scheme				Third-order scheme				Fourth-order scheme
Mesh	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate
11	4.8e-1		8.7e-1		2.2e-2		4.7e-2		2.4e-3		4.5e-3
21	1.2e-1	2.2	2.2e-1	2.1	2.4e-3	3.2	5.6e-3	3.1	1.4e-4	4.1	2.8e-4	4.0
41	2.8e-2	2.1	5.5e-2	2.0	2.8e-4	3.1	6.8e-4	3.0	8.8e-6	4.0	1.7e-5	4.0
81	6.9e-3	2.0	1.4e-2	2.0	3.4e-5	3.1	8.4e-5	3.0	5.4e-7	4.0	1.1e-6	4.0
161	1.7e-3	2.0	3.4e-3	2.0	4.2e-6	3.0	1.0e-5	3.0	3.3e-8	4.0	6.8e-8	4.0

| Show Table

DownLoad: CSV

Table 2. Comparison of

$L_{2}$ and

$L_{\infty}$ errors in second-, third- and fourth-order formats on non-fitted mesh Example 1,

$\kappa$ = 100.

Mesh	Second-order scheme				Third-order scheme				Fourth-order scheme
Mesh	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate
11	8.9e-2		1.8e-1		9.8e-2		2.7e-1		2.4e-3		4.5e-3
21	6.9e-2	0.4	1.2e-1	0.6	2.4e-2	2.0	4.5e-2	2.6	1.5e-4	4.0	2.8e-4	4.0
41	2.8e-2	1.3	5.5e-2	1.1	3.5e-3	2.8	7.3e-3	2.6	8.8e-6	4.0	1.8e-5	4.0
81	6.8e-3	2.2	1.3e-2	2.0	4.3e-4	3.1	8.4e-4	3.1	5.4e-7	4.0	1.1e-6	4.0
161	1.5e-3	2.1	3.0e-3	2.2	5.5e-5	2.9	1.0e-4	3.0	3.4e-8	4.0	6.8e-8	5.0

| Show Table

DownLoad: CSV

Due to the strong discontinuity of physical quantities at the interface, it is crucial for the numerical scheme to be stable and robust. and compare the errors of $u^{+}$ and $u^{-}$ for different methods on fitted and non-fitted mesh, respectively. It can be seen from the tables that the physical quantities on both sides of the interface differ in magnitude by a factor of 100, but the numerical scheme proposed can capture the discontinuity of physical quantities on both sides of the interface with a high degree of accuracy.

Table 3. Comparison of the errors of

$u^{+}$ and

$u^{-}$ for different methods on fitted mesh for Example 1,

$\kappa$ = 100.

	$(u^{-}, u^{+})$ = (1.284025,128.4025)
	Mesh	11	21	41	81	161
Second	$e_{\infty}$	(8.6e-4, 8.7e-1)	(2.1e-4, 2.2e-1)	(5.3e-5, 5.5e-2)	(1.3e-5, 1.4e-2)	(3.3e-6, 3.4e-3)
Second	$e^{2h}_{\infty}/e^{h}_{\infty}$	-	(4.0, 4.0)	(4.0, 4.0)	(4.0, 4.0)	(4.0, 4.0)
Third	$e_{\infty}$	(8.6e-4, 4.7e-2)	(1.1e-4, 5.6e-3)	(1.4e-5, 6.8e-4)	(1.7e-6, 8.4e-5)	(2.1e-7, 1.0e-5)
Third	$e^{2h}_{\infty}/e^{h}_{\infty}$	-	(8.0, 8.4)	(7.9, 8.2)	(8.0, 8.1)	(8.0, 8.1)
Fourth	$e_{\infty}$	(1.2e-5, 4.5e-3)	(7.3e-7, 2.8e-4)	(4.6e-8, 1.8e-5)	(2.9e-9, 1.1e-6)	(1.8e-10, 6.8e-8)
Fourth	$e^{2h}_{\infty}/e^{h}_{\infty}$	-	(15.9, 16.0)	(16.0, 16.0)	(16.0, 16.1)	(16.1, 16.0)

| Show Table

DownLoad: CSV

Table 4. Comparison of the errors of

$u^{+}$ and

$u^{-}$ for different methods on non-fitted mesh for Example 1,

$\kappa$ = 100.

	$(u^{-}, u^{+})$ = (1.284025,128.4025)
	mesh	11	21	41	81	161
Second	$e_{\infty}$	(4.7e-3, 1.3e-1)	(1.6e-3, 1.1e-1)	(1.3e-3, 5.5e-2)	(2.7e-4, 1.3e-2)	(6.4e-5, 3.0e-3)
Second	$e^{2h}_{\infty}/e^{h}_{\infty}$	-	(3.6, 4.2)	(3.8, 4.1)	(3.9, 4.1)	(4.2, 4.5)
Third	$e_{\infty}$	(8.7e-4, 4.7e-2)	(1.1e-4, 5.6e-3)	(1.4e-5, 6.8e-4)	(1.7e-6, 8.4e-5)	(2.1e-7, 1.0e-5)
Third	$e^{2h}_{\infty}/e^{h}_{\infty}$	-	(8.0, 8.4)	(7.9, 8.2)	(8.0, 8.1)	(8.0, 8.1)
Fourth	$e_{\infty}$	(1.2e-5, 4.5e-3)	(7.3e-7, 2.8e-4)	(4.6e-8, 1.8e-5)	(2.9e-9, 1.1e-6)	(1.8e-10, 6.8e-8)
Fourth	$e^{2h}_{\infty}/e^{h}_{\infty}$	-	(15.9, 16.0)	(16.0, 16.0)	(16.0, 16.1)	(16.1, 16.0)

| Show Table

DownLoad: CSV

compares the exact and numerical solutions on fitted and non-fitted mesh for Problem 1 at $k = 100$ with a number of intervals of $32$ . It is shown that the numerical solutions are well matched with the exact solutions, in spite of the fact that the amount of physical jumps on both sides of the interface are much larger. Figures 8 and 9 errors by the fourth-order, second-order and third-order scheme on fitted and non-fitted mesh, respectively. It can be seen that the errors are decrease as the number of mesh points increases.

Figure 7. Comparison of exact and numerical solutions on fitted and non-fitted mesh for Problem 1.

DownLoad: Full-Size Img PowerPoint

Figure 8. Comparison of errors by the fourth-order scheme on fitted mesh and non-fitted mesh for Example 1.

DownLoad: Full-Size Img PowerPoint

Figure 9. Comparison of errors by the second-order and third-order schemes for Example 1.

DownLoad: Full-Size Img PowerPoint

Example 2. Consider the elliptic interface problem with a variable diffusion coefficient, and the solution is separated into two parts by the interface at $x = \alpha$ , where $\alpha = 0.5$ and $\alpha = 0.53232$ , for the interface-fitted mesh and interface-cut mesh cases, respectively. The analytical solution of this problem is given by

$\begin{equation*} u(x, y) = \begin{cases} x^2 e^{x} , & \qquad x \in (0, \alpha), \\ \kappa e^{x}, & \qquad x \in (\alpha, 1). \end{cases} \end{equation*}$

The diffusion coefficient is defined as follows

$\begin{equation*} \beta = \begin{cases} \kappa , & \qquad x \in (0, \alpha), \\ 2 x + x^2, & \qquad x \in (\alpha, 1). \end{cases} \end{equation*}$

Conservation of the flux on the interface satisfies

$\begin{equation*} [\beta \frac{\partial u}{\partial x}] = \beta^{+}\frac{\partial u^{+}}{\partial x} - \beta^{-} \frac{\partial u^{-}}{\partial x} = \kappa (2x + x^2) e^{x} - \kappa (2x + x^2) e^{x} = 0, \quad at \quad x = \alpha. \end{equation*}$

The coefficient $\lambda$ is given as

$\begin{equation*} \lambda = \frac{\kappa - \alpha^2}{\kappa (2\alpha +\alpha^2)}. \end{equation*}$

and compare the $L_{2}$ and $L_{\infty}$ errors in second-, third- and fourth-order formats on fitted and non-fitted mesh, respectively. It can be seen that the numerical format is able to achieve theoretical accuracy on fitted mesh. The numerical accuracy on the non-fitted mesh is slightly lower than the theoretical accuracy, probably due to the unbalanced grid step on both sides of the interface, which leads to the instability of the format.

Table 5. Comparison of

$L_{2}$ and

$L_{\infty}$ errors in second-, third- and fourth-order formats on fitted mesh for Example 2,

$\kappa$ = 100.

Mesh	Second-order scheme				Third-order scheme				Fourth-order scheme
Mesh	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate
21	1.4e-2		2.9e-2		9.9e-5		2.4e-4		3.2e-6		6.1e-6
41	3.4e-3	2.0	7.3e-3	2.0	1.1e-5	3.1	2.9e-5	3.1	2.0e-7	4.0	4.0e-7	4.0
81	8.3e-4	2.0	1.8e-3	2.0	1.4e-6	3.1	3.5e-6	3.0	1.2e-8	4.0	2.5e-8	4.0
161	2.1e-4	2.0	4.6e-4	2.0	1.7e-7	3.0	4.3e-7	3.0	7.5e-10	4.0	1.5e-9	4.1
321	5.1e-5	2.0	1.1e-4	2.0	2.1e-8	3.0	5.4e-8	3.0	4.6e-11	4.0	9.2e-11	4.1

| Show Table

DownLoad: CSV

Table 6. Comparison of

$L_{2}$ and

$L_{\infty}$ errors in second-, third- and fourth-order formats on non-fitted mesh for Example 2,

$\kappa$ = 100.

Mesh	Second-order scheme				Third-order scheme				Fourth-order scheme
Mesh	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate
21	8.9e-3		1.5e-2		1.3e-3		5.9e-3		1.1e-3		5.2e3
41	3.6e-3	1.3	7.6e-3	1.0	3.2e-5	5.3	7.5e-5	6.3	3.2e-5	5.1	8.8e-5	5.9
81	8.3e-4	2.1	1.8e-3	2.1	3.7e-6	3.1	1.3e-5	2.5	3.8e-6	3.1	1.5e-5	2.6
161	1.8e-4	2.2	4.0e-4	2.2	4.2e-7	3.1	3.8e-6	1.8	4.8e-7	3.0	4.0e-6	1.9
321	2.8e-5	2.7	5.2e-5	2.9	9.5e-8	2.2	1.7e-6	1.2	9.9e-8	2.3	1.7e-6	1.3

| Show Table

DownLoad: CSV

compares the exact and numerical solutions on fitted and non-fitted mesh for Problem 1 at $k = 100$ with a number of intervals of $41$ . It is shown that the numerical solutions are well matched with the exact solutions, in spite of the fact that the amount of physical jumps on both sides of the interface are much larger. Figure 11 compares errors by the fourth-order and third-order schemes on fitted and non-fitted mesh, respectively. As seen in the table, the errors in the fourth-order scheme are significantly lower than those in the third-order scheme.

Figure 10. Comparison of exact and numerical solutions for Problem 2.

DownLoad: Full-Size Img PowerPoint

Figure 11. Comparison of errors under different kinds schemes for Problem 2.

DownLoad: Full-Size Img PowerPoint

Table 7 compares the length ratio of the interface-cut mesh interval. The table shows that as the number of grids changes, the spacing ratio changes significantly and affects the accuracy of the numerical scheme.

Table 7. Comparison of the length ratio

$h_f/h_b$ of interface-cut mesh.

Mesh	21	41	81	161	321
$h_f / h_b$	1.48	2.80	1.72e-2	1.36	2.30

| Show Table

DownLoad: CSV

compares the condition numbers of the linear system matrix $A$ for the fitted and non-fitted meshes. It can be seen that the condition number of the matrix does not differ much for the same grid number, so the proposed numerical scheme is stable.

Table 8. Comparison of condition numbers of the linear system on fitted and non-fitted mesh.

Mesh	21	41	81	161	321
Fitted	2098	6746	20081	63749	147695
Non-fitted	4108	14355	31339	107609	341350

| Show Table

DownLoad: CSV

Example 3. Consider the elliptic interface problem with a variable diffusion coefficient, and the solution is separated into two parts by the interface at $x = \alpha$ , where $\alpha = 0.5$ and $\alpha = 0.53232$ , for the interface-fitted mesh and interface-cut mesh cases, respectively. The analytical solution of this problem is given by

$\begin{equation*} u(x, y) = \begin{cases} \sin x e^{x^2}, & \qquad x \in (0, \alpha), \\ \kappa e^{x^2}, & \qquad x \in (\alpha, 1). \end{cases} \end{equation*}$

The diffusion coefficient is defined as follows

$\begin{equation*} \beta = \begin{cases} 2 \kappa x , & \qquad x \in (0, \alpha), \\ \cos x + 2x \sin x, & \qquad x \in (\alpha, 1). \end{cases} \end{equation*}$

Conservation of the flux on the interface satisfies

$\begin{equation*} [\beta \frac{\partial u}{\partial x}] = \beta^{+}\frac{\partial u^{+}}{\partial x} - \beta^{-} \frac{\partial u^{-}}{\partial x} = 2 \kappa x (2x \sin x + \cos x) e^{x^2} - 2 \kappa x (2x \sin x + \cos x) e^{x^2} = 0. \end{equation*}$

The coefficient $\lambda$ is given as

$\begin{equation*} \lambda = \frac{\kappa - \sin \alpha }{2 \kappa \alpha (\cos \alpha + 2\alpha \sin \alpha) }. \end{equation*}$

and compare the $L_{2}$ and $L_{\infty}$ errors in second-, third- and fourth-order formats on fitted and non-fitted mesh, respectively. It is shows that the numerical format is able to achieve theoretical accuracy on fitted mesh.

Table 9. Comparison of

$L_{2}$ and

$L_{infty}$ errors in second-, third- and fourth-order formats on fitted mesh for Example 3,

$\kappa$ = 1000.

Mesh	Second-order scheme				Third-order scheme				Fourth-order scheme
Mesh	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate
21	1.4		2.6		2.1e-2		5.0e-2		1.8e-3		3.4e-3
41	3.4e-1	2.0	6.4e-1	2.0	2.4e-3	3.1	6.0e-3	3.1	1.1e-4	4.0	2.1e-4	4.0
81	8.4e-2	2.0	1.6e-1	2.0	2.9e-4	3.1	7.4e-4	3.0	6.9e-6	4.0	1.3e-5	4.0
161	2.1e-2	2.0	4.0e-2	2.0	3.5e-5	3.0	9.1e-5	3.0	4.4e-7	4.0	8.5e-7	4.0
321	5.2e-3	2.0	1.0e-2	2.0	4.3e-6	3.0	1.1e-5	3.0	2.9e-8	3.9	5.5e-8	3.9

| Show Table

DownLoad: CSV

Table 10. Comparison of

$L_{2}$ and

$L_{infty}$ errors in second-, third- and fourth-order formats on non-fitted mesh for Example 3,

$\kappa$ = 1000.

Mesh	Second-order scheme				Third-order scheme				Fourth-order scheme
Mesh	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate	$L_2$	Rate	$L_{\infty}$	Rate
21	9.9e-1		1.7		4.0e-2		1.4e-1		1.4e-2		6.3e-2
41	3.4e-1	1.6	6.4e-1	1.4	8.2e-4	5.6	1.9e-3	6.2	1.1e-3	3.7	2.5e-3	4.6
81	8.1e-2	2.1	1.6e-1	2.0	5.3e-5	4.0	1.2e-4	4.0	1.1e-4	3.2	2.9e-4	3.1
161	1.9e-2	2.1	3.6e-2	2.1	1.0e-5	2.4	5.9e-5	1.0	1.1e-5	3.3	3.0e-5	3.3
321	3.6e-3	2.4	6.3e-3	2.5	1.8e-6	2.5	3.9e-5	0.6	1.2e-6	3.3	1.7e-6	4.1

| Show Table

DownLoad: CSV

compares the exact and numerical solutions on fitted and non-fitted mesh for Problem 1 at $k = 100$ with a number of intervals of $41$ . It can be seen that the error in matching the numerical solution to the exact solution is essentially small despite the large physical jumps on both sides of the interface. Figure 13 compares errors by the fourth-order and third-order schemes on fitted and non-fitted mesh, respectively. As can be seen in the table, the numerical accuracy of the fourth-order scheme is better than that of the third-order scheme.

Figure 12. Comparison of exact and numerical solutions for Problem 3.

DownLoad: Full-Size Img PowerPoint

Figure 13. Errors under the third-order scheme (a) and fourth-order scheme (b) for Problem 3, with different mesh sizes.

DownLoad: Full-Size Img PowerPoint

7. Conclusions

The 1D elliptic interface problem with imperfect contact is characterized by the fact that the jump quantity of the solution is unknown and is related to the flux across the interface. In this paper, a class of higher-order finite-difference schemes is constructed for interface fitted and non-fitted meshes, respectively. The second-, third-, and fourth-order approximations of the jump conditions are provided by using the connected jump conditions and their higher-order derivatives. Some numerical experiments were carried out to illustrate the accuracy and stability of the present method. The numerical results show that the proposed scheme is able to capture the solution of the discontinuous case of the interface body with theoretical accuracy. The theoretical accuracy has been achieved for the elliptic interface problem with implicit interface connection conditions.

Acknowledgments

This work was partly supported by the National Natural Science Foundation (12161067, 12261067, 12001015, 62201298, 51961031), the National Natural Science Foundation of China Youth Fund Project (11801287), the Inner Mongolia Autonomous Region Youth Science and Technology Talents Support Program (NJYT20B15), the Inner Mongolia Scientific Fund Project (2020MS06010, 2021LHMS01006, 2022MS01008), and Innovation Fund Project of Inner Mongolia University of Science and Technology-Excellent Youth Science Fund Project (2019YQL02).

Conflict of interest

The authors declare that the publication of this paper does coincide with any confilict of interest.

References

[1]	B. Eleonora, P. Pietro, The language of chaos, Int. J. Bifurcat. Chaos, 16 (2006), 523–557. https://doi.org/10.1142/S0218127406014988 doi: 10.1142/S0218127406014988
[2]	R. B. Leipnik, T. A. Newton, Double strange attractors in rigid body motion, Phys. Lett. A, 86 (1981), 63–67. https://doi.org/10.1016/0375-9601(81)90165-1 doi: 10.1016/0375-9601(81)90165-1
[3]	X. Wang, L. Tian, Bifurcation analysis and linear control of the Newton-Leipnik system, Chaos Soliton. Fract., 27 (2006), 31–38. https://doi.org/10.1016/j.chaos.2005.04.009 doi: 10.1016/j.chaos.2005.04.009
[4]	H. K. Chen, C. I. Lee, Anti-control of chaos in rigid body motion, Chaos Soliton. Fract., 21 (2004), 957–965. https://doi.org/10.1016/j.chaos.2003.12.034 doi: 10.1016/j.chaos.2003.12.034
[5]	H. Richter, Controlling chaotic system with multiple strange attractors, Phys. Lett. A, 300 (2002), 182–188. https://doi.org/10.1016/S0375-9601(02)00183-4 doi: 10.1016/S0375-9601(02)00183-4
[6]	L. J. Sheu, H. K. Chen, J. H. Chen, L. M. Tam, W. C. Chen, K. T. Lin, et al., Chaos in the Newton-Leipnik system with fractional order, Chaos Soliton. Fract., 36 (2008), 98–103. https://doi.org/10.1016/j.chaos.2006.06.013 doi: 10.1016/j.chaos.2006.06.013
[7]	K. M. S. Tavazoei, M. Haeri, A necessary condition for double scroll attractor existence in fractional order systems, Phys. Lett. A, 367 (2007), 102–113. https://doi.org/10.1016/j.physleta.2007.05.081 doi: 10.1016/j.physleta.2007.05.081
[8]	H. K. Chen, Chaos and chaos synchronization of a symmetric gyro with linear-plus-cubic damping, J. Sound. Vib., 255 (2002), 719–740. https://doi.org/10.1006/jsvi.2001.4186 doi: 10.1006/jsvi.2001.4186
[9]	A. Khalid, Splines solutions of boundary value problems that arises in sculpturing electrical process of motors with two rotating mechanism circuit, Phys. Scripta, 96 (2021), 104001. https://doi.org/10.1088/1402-4896/ac0bd0 doi: 10.1088/1402-4896/ac0bd0
[10]	M. Caputo, M. Fabrizio, A new definition of fractional derivative without singular kernel, Prog. Fract. Dier. Appl., 2 (2015), 1–13. Available from: https://digitalcommons.aaru.edu.jo/pfda/vol1/iss2/1
[11]	W. Deng, C. Li, J. Lu, Stability analysis of linear fractional differential system with multiple time delays, Nonlinear Dynam., 48 (2007), 409–416. https://doi.org/10.1007/s11071-006-9094-0 doi: 10.1007/s11071-006-9094-0
[12]	S. Rashid, K. T. Kubra, S. Sultana, P. Agarwal, M. S. Osman, An approximate analytical view of physical and biological models in the setting of Caputo operator via Elzaki transform decomposition method, J. Comput. Appl. Math., 413 (2022), 114378. https://doi.org/10.1016/j.cam.2022.114378 doi: 10.1016/j.cam.2022.114378
[13]	V. D. Gejji, Y. Sukale, S. Bhalekar, A new predictor-corrector method for fractional differential equations, Appl. Math. Comput., 244 (2014), 158–182. https://doi.org/10.1016/j.amc.2014.06.097 doi: 10.1016/j.amc.2014.06.097
[14]	K. M. Owolabi, Mathematical modelling and analysis of two-component system with Caputo fractional derivative order, Chaos Soliton. Fract., 103 (2017), 544–554. https://doi.org/10.1016/j.chaos.2017.07.013 doi: 10.1016/j.chaos.2017.07.013
[15]	C. Li, C. Tao, On the fractional adams method, Comput. Math. Appl., 58 (2009), 1573–1588. https://doi.org/10.1016/j.camwa.2009.07.050 doi: 10.1016/j.camwa.2009.07.050
[16]	V. D. Gejji, H. Jafari, Analysis of a system of non autonomous fractional differential equations involving Caputo derivatives, J. Math. Anal. Appl., 328 (2007), 1026–1033. https://doi.org/10.1016/j.jmaa.2006.06.007 doi: 10.1016/j.jmaa.2006.06.007
[17]	A. Atangana, J. F. G. Aguilar, Decolonisation of fractional calculus rules: Breaking commutativity and associativity to capture more natural phenomena, Eur. Phys. J. Plus, 133 (2018), 1–23. https://doi.org/10.1140/epjp/i2018-12021-3 doi: 10.1140/epjp/i2018-12021-3
[18]	A. Atangana, Non validity of index law in fractional calculus: A fractional differential operator with Markovian and non-Markovian properties, Physica A, 505 (2018), 688–706. https://doi.org/10.1016/j.physa.2018.03.056 doi: 10.1016/j.physa.2018.03.056
[19]	M. S. Tavazoei, M. Haeri, Chaotic attractors in incommensurate fractional order systems, Physica D, 237 (2008), 2628–2637. https://doi.org/10.1016/j.physd.2008.03.037 doi: 10.1016/j.physd.2008.03.037
[20]	H. M. Baskonus, T. Mekkaoui, Z. Hammouch, H. Bulut, Active control of a chaotic fractional order economic system, Entropy, 17 (2015), 5771–5783. https://doi.org/10.3390/e17085771 doi: 10.3390/e17085771
[21]	M. Toufik, A. Atangana, New numerical approximation of fractional derivative with non-local and non-singular kernel: Application to chaotic models, Eur. Phys. J. Plus, 132 (2017), 1–16. https://doi.org/10.1140/epjp/i2017-11717-0 doi: 10.1140/epjp/i2017-11717-0
[22]	L. Galeone, R. Garrappa, Fractional adams-moulton methods, Math. Comput. Simulat., 79 (2008), 1358–1367. https://doi.org/10.1016/j.matcom.2008.03.008 doi: 10.1016/j.matcom.2008.03.008
[23]	K. Hattaf, Stability of fractional differential equations with new generalized hattaf fractional derivative, Math. Probl. Eng., 2021 (2021), 8608447. https://doi.org/10.1155/2021/8608447 doi: 10.1155/2021/8608447
[24]	K. Hattaf, Z. Hajhouji, M. A. Ichou, N. Yousfi, A Numerical method for fractional differential equations with new generalized hattaf fractional derivative, Math. Probl. Eng., 2022 (2022). https://doi.org/10.1155/2022/3358071 doi: 10.1155/2022/3358071
[25]	K. Hattaf, On the stability and numerical scheme of fractional differential equations with application to biology, Computation, 10 (2022), 97. https://doi.org/10.3390/computation10060097 doi: 10.3390/computation10060097
[26]	K. Hattaf, A new generalized definition of fractional derivative with non-singular kernel, Computation, 8 (2020), 49. https://doi.org/10.3390/computation8020049 doi: 10.3390/computation8020049
[27]	K. Hattaf, A new class of generalized fractal and fractal-fractional derivatives with non-singular kernels, Fractal Fract., 7 (2023), 395. https://doi.org/10.3390/fractalfract7050395 doi: 10.3390/fractalfract7050395
[28]	M. H. Alshehri, S. Saber, F. Z. Duraihem, Dynamical analysis of fractional-order of IVGTT glucose-insulin interaction, Int. J. Nonlin. Sci. Num., 24 (2023), 1123–1140. https://doi.org/10.1515/ijnsns-2020-0201 doi: 10.1515/ijnsns-2020-0201
[29]	M. H. Alshehri, F. Z. Duraihem, A. Alalyani, S. Saber, A Caputo (discretization) fractional-order model of glucose-insulin interaction: Numerical solution and comparisons with experimental data, J. Taibah Univ. Sci., 15 (2021), 26–36. https://doi.org/10.1080/16583655.2021.1872197 doi: 10.1080/16583655.2021.1872197
[30]	S. Saber, A. M. Alghamdi, G. A. Ahmed, K. M. Alshehri, Mathematical modelling and optimal control of pneumonia disease in sheep and goats in Al-Baha region with cost-effective strategies, AIMS Math., 7 (2022), 12011–12049. https://doi.org/10.3934/math.2022669 doi: 10.3934/math.2022669
[31]	S. Saber, A. Alalyani, Stability analysis and numerical simulations of IVGTT glucose-insulin interaction models with two time delays, Math. Model. Anal., 27 (2022), 383–407. https://doi.org/10.3846/mma.2022.14007 doi: 10.3846/mma.2022.14007
[32]	A. Alalyani, S. Saber, Stability analysis and numerical simulations of the fractional COVID-19 pandemic model, Int. J. Nonlin. Sci. Num., 24 (2023), 989–1002. https://doi.org/10.1515/ijnsns-2021-0042 doi: 10.1515/ijnsns-2021-0042
[33]	T. W. Zhang, L. L. Xiong, Periodic motion for impulsive fractional functional differential equations with piecewise Caputo derivative, Appl. Math. Lett., 101 (2020), 106072. https://doi.org/10.1016/j.aml.2019.106072 doi: 10.1016/j.aml.2019.106072
[34]	K. I. A. Ahmed, H. D. S. Adam, M. Y. Youssif, S. Saber, Different strategies for diabetes by mathematical modeling: Modified Minimal Model, Alex. Eng. J., 80 (2023), 74–87. https://doi.org/10.1016/j.aej.2023.07.050 doi: 10.1016/j.aej.2023.07.050
[35]	K. I. A. Ahmed, H. D. S. Adam, M. Y. Youssif, S. Saber, Different strategies for diabetes by mathematical modeling: Applications of fractal-fractional derivatives in the sense of Atangana-Baleanu, Results Phys., 2023, 106892. https://doi.org/10.1016/j.rinp.2023.106892 doi: 10.1016/j.rinp.2023.106892
[36]	S. G. Samko, Fractional integration and differentiation of variable order, Anal, Math., 21 (1995), 213–236. https://doi.org/10.1007/s11071-012-0485-0 doi: 10.1007/s11071-012-0485-0
[37]	J. E. S. Pérez, J. F. G. Aguilar, A. Atangana, Novel numerical method for solving variable-order fractional differential equations with power, exponential and Mittag-Leffler laws, Chaos Soliton. Fract., 114 (2018), 175–185. https://doi.org/10.1016/j.chaos.2018.06.032 doi: 10.1016/j.chaos.2018.06.032
[38]	B. S. T. Alkahtani, I. Koca, A. Atangana, A novel approach of variable order derivative: Theory and methods, J. Nonlinear Sci. Appl., 9 (2016), 4867–4876. http://dx.doi.org/10.22436/jnsa.009.06.122 doi: 10.22436/jnsa.009.06.122
[39]	A. Atangana, On the stability and convergence of the time-fractional variable-order telegraph equation, J. Comput. Phys., 293 (2015), 104–114. https://doi.org/10.1016/j.jcp.2014.12.043 doi: 10.1016/j.jcp.2014.12.043
[40]	A. Atangana, D. Baleanu, New fractional derivatives with nonlocal and non-singular kernel: Theory and application to heat transfer model, Therm. Sci., 20 (2016), 763–769. https://doi.org/10.48550/arXiv.1602.03408 doi: 10.48550/arXiv.1602.03408
[41]	S. Kumar, A. Kumar, D. Baleanu, Two analytical methods for time-fractional nonlinear coupled Boussinesq-Burger's equations arise in propagation of shallow water waves, Nonlinear Dyn., 1 (2016), 1–17. https://doi.org/10.1007/s11071-016-2716-2 doi: 10.1007/s11071-016-2716-2
[42]	P. Zhuang, F. Liu, V. Anh, I. Turner, Numerical methods for the variable-order fractional advection-diffusion equation with a nonlinear source term, SIAM J. Numer. Anal., 47 (2009), 1760–1781. https://doi.org/10.1137/080730597 doi: 10.1137/080730597
[43]	A. H. Bhrawy, M. A. Zaky, Numerical simulation for two-dimensional variable-order fractional nonlinear cable equation, Nonlinear Dyn., 80 (2015), 101–116. https://doi.org/10.1007/s11071-014-1854-7 doi: 10.1007/s11071-014-1854-7
[44]	S. Djennadi, N. Shawagfeh, M. Inc, M. S. Osman, The Tikhonov regularization method for the inverse source problem of time fractional heat equation in the view of ABC-fractional technique, Phys. Scr., 96 (2021), 094006. https://doi.org/10.1088/1402-4896/ac0867 doi: 10.1088/1402-4896/ac0867
[45]	B. P. Moghaddam, S. Yaghoobi, J. T. Machado, An extended predictor-corrector algorithm for variable-order fractional delay differential equations, J. Comput. Nonlinear Dyn., 1 (2016), 1–11. https://doi.org/10.1115/1.4032574 doi: 10.1115/1.4032574
[46]	M. F. Danca, Lyapunov exponents of a discontinuous 4D hyperchaotic system of integer or fractional order, Entropy, 20 (2018), 337. https://doi.org/10.3390/e20050337 doi: 10.3390/e20050337
[47]	M. F. Danca, N. Kuznetsov, Matlab code for Lyapunov exponents of fractional-order systems, Int. J. Bif. Chaos, 28 (2018), 1850067. https://doi.org/10.1142/S0218127418500670 doi: 10.1142/S0218127418500670
[48]	L. Shi, S. Tayebi, O. A. Arqub, M. S. Osman, P. Agarwal, W. Mahamoud, et al., The novel cubic B-spline method for fractional Painleve and Bagley-Trovik equations in the Caputo, Caputo-Fabrizio, and conformable fractional sense, Alex. Eng. J., 65 (2023), 413–426. https://doi.org/10.1016/j.aej.2022.09.039 doi: 10.1016/j.aej.2022.09.039
[49]	A. Atangana, K. M. Owolabi, New numerical approach for fractional differential equations, Math. Model. Nat. Phenom., 13 (2018), 1–17. https://doi.org/10.48550/arXiv.1707.08177 doi: 10.48550/arXiv.1707.08177
[50]	A. Atangana, I. S. Araz, New numerical method for ordinary differential equations: Newton polynomial, J. Comput. Appl. Math., 372 (2019). https://doi.org/10.1016/j.cam.2019.112622 doi: 10.1016/j.cam.2019.112622
[51]	A. Atangana, I. S. Araz, New numerical scheme with newton polynomial, theory, methods, and applications, 1 Eds., Academic Press, 2021.
[52]	B. S. T. Alkahtani, A new numerical scheme based on Newton polynomial with application to fractional nonlinear differential equations, Alex. Eng. J., 59 (2019), 1893–1907. https://doi.org/10.1016/j.aej.2019.11.008 doi: 10.1016/j.aej.2019.11.008
[53]	T. W. Zhang, Y. K. Li, Exponential Euler scheme of multi-delay Caputo-Fabrizio fractional-order differential equations, Appl. Math. Lett., 124 (2022), 107709. https://doi.org/10.1016/j.aml.2021.107709 doi: 10.1016/j.aml.2021.107709
[54]	K. K. Ali, M. A. A. Salam, E. M. H. Mohamed, B. Samet, S. Kumar, Numerical solution for generalized nonlinear fractional integro-differential equations with linear functional arguments using Chebyshev series, Adv. Differ. Equ., 494 (2020). https://doi.org/10.1186/s13662-020-02951-z doi: 10.1186/s13662-020-02951-z
[55]	S. Rashid, K. T. Kubra, S. Sultana, P. Agarwal, M. S. Osman, An approximate analytical view of physical and biological models in the setting of Caputo operator via Elzaki transform decomposition method, J. Comput. Appl. Math., 413 (2022), 114378
[56]	S. Qureshi, A. Soomro, E. Hincal, J. R. Lee, C. Park, M. S. Osman, An efficient variable stepsize rational method for stiff, singular and singularly perturbed problems, Alex. Eng. J., 61 (2022), 10953–10963. https://doi.org/10.1016/j.aej.2022.03.014 doi: 10.1016/j.aej.2022.03.014
[57]	O. A. Arqub, M. S. Osman, C. Park, J. R. Lee, H. Alsulam, M. Alhodaly, Development of the reproducing kernel Hilbert space algorithm for numerical pointwise solution of the time-fractional nonlocal reaction-diffusion equation, Alex. Eng. J., 61 (2022), 10539–10550. https://doi.org/10.1016/j.aej.2022.04.008 doi: 10.1016/j.aej.2022.04.008
[58]	O. A. Arqub, S. Tayebi, D. Baleanu, M. S. Osman, W. Mahmoud, H. Alsulami, A numerical combined algorithm in cubic B-spline method and finite difference technique for the time-fractional nonlinear diffusion wave equation with reaction and damping terms, Results Phys., 41 (2022), 105912. https://doi.org/10.1016/j.rinp.2022.105912 doi: 10.1016/j.rinp.2022.105912
[59]	N. Djeddi, S. Hasan, M. A. Smadi, S. Momani, Modified analytical approach for generalized quadratic and cubic logistic models with Caputo-Fabrizio fractional derivative, Alex. Eng. J., 59 (2020), 5111–5122. https://doi.org/10.1016/j.aej.2020.09.041 doi: 10.1016/j.aej.2020.09.041
[60]	A. Khalid, A. S. A. Alsubaie, M. Inc, A. Rehan, W. Mahmoud, M. S. Osman, Cubic spline solutions of the higher-order boundary value problems arise in sandwich panel theory, Results Phys., 39 (2022), 105726. https://doi.org/10.1016/j.rinp.2022.105726 doi: 10.1016/j.rinp.2022.105726

This article has been cited by:

Shougui Zhang, Xiyong Cui, Guihua Xiong, Ruisheng Ran, An Optimal ADMM for Unilateral Obstacle Problems, 2024, 12, 2227-7390, 1901, 10.3390/math12121901

Reader Comments

Your name:*

Email:*
© 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)