Next Article in Journal
Mathematical Modeling of Alzheimer’s Drug Donepezil Hydrochloride Transport to the Brain after Oral Administration
Next Article in Special Issue
Neural Fractional Differential Equations: Optimising the Order of the Fractional Derivative
Previous Article in Journal
Fractional Fourier Series on the Torus and Applications
Previous Article in Special Issue
Numerical Analysis and Computation of the Finite Volume Element Method for the Nonlinear Coupled Time-Fractional Schrödinger Equations
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Difference Approximation for 2D Time-Fractional Integro-Differential Equation with Given Initial and Boundary Conditions

1
College of Big Data Statistics, Guizhou University of Finance and Economics, Guiyang 550025, China
2
School of Mathematics and Physics, Xinjiang Institute of Engineering, Urumqi 830023, China
3
School of Mathematical Sciences, Xinjiang Normal University, Urumqi 830017, China
4
College of Science, Henan University of Technology, Zhengzhou 450001, China
*
Author to whom correspondence should be addressed.
Fractal Fract. 2024, 8(8), 495; https://doi.org/10.3390/fractalfract8080495
Submission received: 6 June 2024 / Revised: 8 August 2024 / Accepted: 19 August 2024 / Published: 22 August 2024

Abstract

In this investigation, a new algorithm based on the compact difference method is proposed. The purpose of this investigation is to solve the 2D time-fractional integro-differential equation. The Riemann–Liouville derivative was utilized to define the time-fractional derivative. Meanwhile, the weighted and shifted Grünwald difference operator and product trapezoidal formula were utilized to construct a high-order numerical scheme. Also, we analyzed the stability and convergence. The convergence order was O ( τ 2 + h x 4 + h y 4 ) , where τ is the time step size, h x and h y are the spatial step sizes. Furthermore, several examples were provided to verify the correctness of our theoretical reasoning.

1. Introduction

Fractional differential equations (FDEs) have attracted much attention for modeling various phenomena in many scientific fields, such as physics [1,2,3], medicine [4,5,6], and electrochemistry [7]. However, analytical solutions in the majority of practical problems are difficult to find. Researchers have used numerical approximations instead of analytical ones to solve FDEs. For this, numerical methods were presented, for example, the finite difference method (FDM) [8,9], finite element method (FEM) [10,11,12], and other efficient methods [13,14,15,16].
In the realm of FDEs, the Caputo and Riemann–Liouville (RL) derivatives are mostly used. Furthermore, RL and Caputo derivatives have different properties. Comparing the definition of the Caputo derivative with the RL one, functions which are derivable in the Caputo sense are much “fewer” than those which are derivable in the RL sense. Physical and geometric interpretations for the RL derivative can be found in [17]. Here, we have no intention of mentioning which derivative is more widely utilized, but we must stress that every derivative has its own serviceable range. Additionally, there are some other special differential definitions, such as Caputo–Fabrizio and Atangana–Baleanu derivatives, without singular kernels [18,19].
Numerical investigations for the time-fractional integro-differential equation (TFIDE) have been studied by some scholars. For the nonlinear time-fractional partial integro-differential equation, Rawani et al. [20] proposed a hybrid method. Cao et al. [21] studied an accurate localized meshless collocation approach. For the linear time-fractional partial integro-differential equation, in [22], Fakhar-Izadi proposed a space–time Spectral–Galerkin method. Kamran et al. [23] proposed a localized transform method to approximate the solution of the equation. Generally speaking, the approximation of the integral term is our focus of research for solving the integro-differential equations. To approximate the integral term, in [24], the second-order fractional quadrature rule was proposed by Lubich. Dehghan and Abbaszadeh [25] studied a new numerical algorithm that combined the finite difference and finite element, with the time convergence order being estimated to be O ( τ 2 ) . Qiao et al. [26] proposed an integral term processing method based on the trapezoidal rule and provided a high-precision ADI difference scheme. Based on the mixed finite element method, Wang et al. [27] considered the second-order scheme in the time direction. Huang et al. [28] studied the orthogonal spline collocation ADI scheme based on Lubich’s second-order quadrature convolution rule. Wang et al. [29,30] considered the trapezoidal PI rule and integration-by-parts method to improve the temporal accuracy, and proposed compact ADI schemes. In [31], Gu and Wu proposed a parallel-in-time iterative algorithm for Volterra partial integral-differential problems with a weakly singular kernel. Since there are much more studies on properties of the Caputo derivative, we focused on the studying the RL derivative in this article, which is helpful in understanding and modeling fractional equations with it.
This study investigates the compact finite difference technique for 2D fractional integro-differential equations. Consider the equation for
D t α 0 u ( x , y , t ) = γ Δ u ( x , y , t ) + I 0 Δ t β u ( x , y , t ) + f ( x , y , t ) , ( x , y ) Ω , t ( 0 , T ] , u ( x , y , t ) = ϕ ( x , y , t ) , ( x , y ) Ω , t ( 0 , T ] , u ( x , y , 0 ) = φ ( x , y ) , ( x , y ) Ω Ω ,
where α , β ( 0 , 1 ) , Δ u ( x , y , t ) is the laplace operator, γ is a positive constant, Ω = { ( x , y ) 0 < x < L 1 , 0 < y < L 2 } with Ω being the boundary, and f ( x , y , t ) , ϕ ( x , y , t ) and φ ( x , y ) are given functions.
In this article, the RL derivative D t α 0 u ( x , y , t ) with 0 < α < 1 is expressed as (one can refer to [32]) follows:
D t α 0 u ( x , y , t ) = 1 Γ ( 1 α ) d d t 0 t ( t s ) α u ( x , y , s ) d s ,
and the Rimann–Liouville fractional integral I t β 0 Δ u ( x , y , t ) with 0 < β < 1 is expressed as
I t β 0 Δ u ( x , y , t ) = 1 Γ ( β ) 0 t ( t s ) β 1 Δ u ( x , y , s ) d s ,
where Γ ( · ) is the Gamma function.
In this article, firstly, we propose a new approach to approximate the RL derivative based on the weighted and shifted Grünwald difference operator. Secondly, a numerical scheme based on compact difference method and product trapezoidal formula is constructed for a 2D TFIDE. Finally, we used theoretical analysis and numerical examples to verify the effectiveness of the proposed method. We need to point out that the proposed method is discussed in view of the sufficient smoothness assumptions on the solution; thus, it is difficult to deal with the typical weak initial singularity of the solution. One can consider the variable-step or the graded mesh method for the typical weak initial singularity of the solution. See [33] for more about the variable-step L1   + method. The following remark ensures the existence and uniqueness of the solution for Equation (1), which provides a guarantee for the subsequent work in this article.
Remark 1
([29,34]). For the sufficiently smooth f ( x , t ) , Equation (1) uses a unique solution and satisfies the regularities as u ( x , y , t ) C 2 ( [ 0 , T ] ; H 2 ( Ω ¯ ) H 0 1 ( Ω ¯ ) ) , for ( x , y , t ) Ω ¯ × [ 0 , T ] , Ω ¯ = Ω Ω .
This article is divided into different sections. The compact finite difference scheme studied in this article is proposed in the next section. Section 3 presents the stability analysis and error estimates of the proposed technique. In Section 4, some numerical tests are presented for comparison with the exact solution in order to verify the effectiveness of the proposed discrete scheme. A brief conclusion is given in the Section 5 of this article.

2. The Numerical Scheme

The compact difference scheme of Equation (1) will be considered in this section. Let t n = n τ ( 0 n N ) , where τ = T / N is the time step size and N is a positive integer. Let x i = i h x ( 0 i M 1 ) and y j = j h y ( 0 j M 2 ) , where h x = L 1 / M 1 and h y = L 2 / M 2 are the spatial step sizes, and M 1 and M 2 are positive integers. The discrete grid Ω h = { ( x i , y j ) | 1 i M 1 1 , 1 j M 2 1 } , Ω h = { ( x i , y j ) | i = 0 o r i = M 1 , j = 0 o r j = M 2 } is the discrete boundary of Ω h , and Ω ¯ h = Ω h Ω h . Then, for the grid function W = { W i j | 1 i M 1 1 , 1 j M 2 1 } on Ω h , if v W , we have
δ x v i 1 2 , j = 1 h x ( v i j v i 1 , j ) , δ x 2 v i , j = 1 h x ( δ x v i 1 2 , j δ x v i + 1 2 , j ) ,
δ y δ x v i 1 2 , j 1 2 = 1 h y ( δ x v i 1 2 , j δ x v i 1 2 , j 1 ) , δ y δ x 2 v i , j 1 2 = 1 h y ( δ x 2 v i , j δ x 2 v i , j 1 ) .
Particularly, for C in this article, C at different positions can represent different constants. Thus, u i j n and U i j n define the exact and approximate solutions of function u at the point ( x i , y j , t n ) , respectively. In order to achieve fourth-order accuracy in the spatial direction, the compact difference operator is required.
H x U j = ( 1 + h x 2 12 δ x 2 ) U i j , 1 i M 1 1 , 0 j M 2 , U i j , i = 0 o r M 1 , 0 j M 2 .
In Equation (3), H x is the compact difference operator in the x direction, and the definition of H y is similar to H x . Further, for any grid function u i j W , define that H u i j = H x H y u i j .
The weighted and shifted Grünwald difference operator and the product trapezoidal formula are used in the following section. Next, some lemmas are introduced.
As in [35], we assume that u L α + 1 ( R ) and let
D t α u ( t ) = 1 Γ ( 1 α ) d d t t ( t s ) α u ( s ) d s .
The following is the shifted Grünwald difference operator:
A τ , p ( α ) u ( t ) = 1 τ α i = 0 g i ( α ) u ( t ( i p ) τ ) ,
and it approximates the RL derivative uniformly with first-order accuracy as τ 0 , i.e.,
A τ , p ( α ) u ( t ) = D u t α ( t ) + O ( τ ) ,
where p is an integer, and the coefficients g i ( α ) ( 0 < α 1 ) are given by
g 0 ( α ) = 1 , g i ( α ) = ( 1 α + 1 i ) g i 1 α , i = 1 , 2 , .
Moreover, we stipulate g 0 ( α ) = 1 and g i ( α ) =0 ( i 1 ) when α = 0 .
Lemma 1
([36]). Let u L 1 ( R ) , and define the weighted and shifted Grünwald difference approximation operator by
D τ , p , q α L u ( t ) = α 2 q 2 ( p q ) A τ , p ( α ) u ( t ) + 2 p α 2 ( p q ) A τ , q ( α ) u ( t ) .
Then, we have
D τ , p , q α L u ( t ) = D u t α ( t ) + O ( τ 2 )
uniformly for t R , where p, q are integers and p q .
Moreover, p and q are symmetric, i.e., D τ , p , q α L = D τ , q , p α L . In [37], the RL derivative combined with Equation (5) was studied. Thus, we have
( 1 + α 2 ) A τ , 0 ( α ) u ( t ) α 2 A τ , 1 ( α ) u ( t ) = 1 τ α k = 0 w k ( α ) u ( t k τ ) = D u t α ( t ) + O ( τ 2 ) ,
where
w 0 ( α ) = ( 1 + α 2 ) g 0 ( α ) = 1 + α 2 , w k ( α ) = ( 1 + α 2 ) g k ( α ) α 2 g k 1 ( α ) , k 1 ,
with these conditions, we have
D t α 0 u ( x i , y j , t n ) = τ α k = 0 n w k ( α ) u ( x i , y j , t n k ) + O ( τ 2 ) .
Lemma 2
([38,39]). Suppose u ( x ) C 6 [ x i 1 , x i + 1 ] and ς ( s ) = 5 ( 1 s ) 3 3 ( 1 s ) 5 ; thus, we have
1 12 ( u x x ( x i + 1 ) + 10 u x x ( x i ) + u x x ( x i 1 ) ) = 1 h x 2 ( u ( x i + 1 ) 2 u ( x i ) + u ( x i 1 ) ) + h x 4 360 0 1 [ f 6 ( x i s h x ) + f 6 ( x i + s h x ) ] ς ( s ) d s .
Lemma 2 gives that
H ( u x x ( x i , y j ) + u y y ( x i , y j ) ) = H y δ x 2 u ( x i , y j ) + H x δ y 2 u ( x i , y j ) + R i j ,
where | R i j | C ( h x 4 + h y 4 ) .
Lemma 3
([40,41]). Suppose u ( t ) C 2 [ 0 , T ] ; thus, there is a positive constant C related only to β when 1 n N . Therefore,
| 0 t n ( t n s ) β 1 u ( s ) d s k = 0 n a n k , n n u ( t n k ) | C max 0 t T | u ( t ) | t n β τ 2 ,
where
a n k , n n = τ β β ( β + 1 ) × 1 , k = 0 , ( k + 1 ) β + 1 2 k β + 1 + ( k 1 ) β + 1 , 1 k n 1 , ( n 1 ) β + 1 ( n 1 β ) n β , k = n .
Lemma 4.
Let a n k , n n under the definition of Equation (8); the sequence { a n k , n n } is monotonically decreasing concerning k.
Proof. 
From Equation (8), we have
a n k , n n = τ β ( ( k + 1 ) β + 1 2 k β + 1 + ( k 1 ) β + 1 ) β ( β + 1 ) ,   with   1 k n 1   and   2 n N .
Let h ( ϵ ) = τ β β ( β + 1 ) × ( ( ϵ + 1 ) β + 1 2 ϵ β + 1 + ( ϵ 1 ) β + 1 ) . Then,
h ( ϵ ) = τ β β × ( ( ϵ + 1 ) β 2 ϵ β + ( ϵ 1 ) β ) .
According to the mean value theorem, we have
h ( ϵ ) = τ β ( ξ 1 β 1 ξ 2 β 1 ) ,
where ξ 1 ( ϵ , ϵ + 1 ) , ξ 2 ( ϵ 1 , ϵ ) ; thus, h ( ϵ ) 0 . Therefore, the sequence { a n k , n n } on [ 1 , ) is monotonically decreasing. □
Lemma 5
([41]). Let β ( 0 < β < 1 ) and a k , n n ( 1 n N ) under the definition of Equation (8); thus,
0 < a 0 , n n < τ β ( β + 1 ) 1 , k = 1 n 1 | a k , n n | T β β 1 .
Assume that u ( x , y , t ) C x , y , t 6 , 6 , 2 ( [ 0 , L 1 ] × [ 0 , L 2 ] × [ 0 , T ] ) and consider Equation (1) on grid point ( x i , y j , t n ) . Thus, with Lemma 1, the left term in Equation (1) has the following form:
D t α 0 u ( x i , y j , t n ) = 1 τ α k = 0 n w k ( α ) u ( x i , y j , t n k ) + O ( τ 2 ) .
From Lemma 3, the second term to the right in Equation (1) can be expressed as follows:
1 Γ ( β ) 0 t n ( t n s ) β 1 Δ u ( x i , y j , s ) d s = 1 Γ ( β ) k = 0 n a n k , n n Δ u ( x i , y j , t n k ) + O ( τ 2 ) .
Then, with Equations (9) and (10), Equation (1) can be rewritten in the following form:
1 τ α k = 0 n w k ( α ) u ( x i , y j , t n k ) = γ Δ u ( x i , y j , t n ) + 1 Γ ( β ) k = 0 n a n k , n n Δ u ( x i , y j , t n k ) + f ( x i , y j , t n ) + O ( τ 2 ) .
Further, applying the compact difference operator H to Equation (11) and by using Equation (7), we can obtain the following difference scheme:
H τ α k = 0 n w k ( α ) u i j n k = γ ( H y δ x 2 u i j n + H x δ y 2 u i j n ) + k = 0 n a n k , n n Γ ( β ) ( H y δ x 2 u i j n k + H x δ y 2 u i j n k ) + H f i j n + R i j n ,
where | R i j n | C ( τ 2 + h x 4 + h y 4 ) .
With Equation (12), by omitting the truncation error R i j n , the compact difference scheme for Equation (1) is as follows:
H k = 0 n w k ( α ) τ α U i j n k = γ ( H y δ x 2 + H x δ y 2 ) U i j n + k = 0 n a n k , n n Γ ( β ) ( H y δ x 2 + H x δ y 2 ) U i j n k + H f i j n , ( x i , y j ) Ω h , n [ 1 , N ] , U i j 0 = ϕ ( x i , y j ) , ( x i , y j ) Ω ¯ h , U i j n = φ ( x i , y j , t n ) , ( x i , y j ) Ω , n [ 1 , N ] .
It is clear to see that the coefficient matrices are strictly diagonally dominant for the compact difference scheme Equation (13). Thus, we have the following result as in [42].
Theorem 1.
The compact difference scheme Equation (13) is uniquely solvable.

3. Stability Analysis and Error Estimates

Now, stability and convergence of the given algorithm are considered. Below are some notations and lemmas to be used. For any grid function U , V W , we have
( U , V ) = h x h y i = 1 M 1 1 j = 1 M 2 1 U i j V i j , U 2 = ( U , U ) , U H x 2 = ( H x U , U ) , u = max 0 j M | u j | .
Lemma 6
([30]). Assume that V W ; thus,
2 3 V 2 V H x 2 V 2   a n d   6 L 1 2 + 6 L 2 2 V 2 δ x V 2 + δ y V 2 .
Lemma 7
([37,43]). If { w k ( α ) } n = 0 under the definition of Equation (6), then for any positive integer m and real vector ( u 0 , u 1 , , u m ) T R m + 1 , it holds that
n = 0 m ( k = 0 n w k ( α ) u n k ) u n 0 .
Lemma 8
([38]). Suppose W , V W , then
( δ x 2 W , V ) = ( δ x W , δ x V ) .
Lemma 9
([38,44]). Let U W ; thus,
δ x U 2 4 h x 2 U 2   a n d   H y U U .
Lemma 10
([41,45]). Assume that { μ n } is a non-negative sequence, which satisfies
μ n k = 0 n 1 η k μ k + b n   w i t h   n 0 ,
where b n is non-decreasing sequence of non-negative numbers, and η k 0 . Thus,
μ n b n exp k = 0 n 1 η k   w i t h   n 0 .
Lemma 11
([46]). If ρ = ( ρ 1 , ρ 2 , , ρ n ) is a non-increasing sequence and ϱ = ( ϱ 1 , ϱ 2 , , ϱ n ) is a non-decreasing sequence, then the following C ˇ e b y s ˇ e v inequality holds
j = 1 m ρ j ϱ j 1 m j = 1 m ρ j j = 1 m ϱ j .
Lemma 12.
Let a n m , n n under the definition of Equation (8). Thus,
m = 1 n 1 | a n m , n n | U n m M * T β ( n 1 ) β m = 1 n 1 U m M m   w i t h   n = 2 , 3 , , N ,
where M * = max 1 m n 1 U n m and M m = U m .
Proof. 
Let M * = max 1 m n 1 U n m and M m = U m . With Lemmas 4 and 11, we have
m = 1 n 1 | a n m , n n | U n m m = 1 n 1 | a n m , n n | M * 1 n 1 m = 1 n 1 | a n m , n n | m = 1 n 1 M * .
From Lemma 5, we have
m = 1 n 1 | a n m , n n | U n m T β ( n 1 ) β m = 1 n 1 M * .
Since the function u continues over Ω × [ 0 , T ] ( Ω = { ( x , y ) 0 x L 1 , 0 y L 2 } ), the function u is bounded on Ω × [ 0 , T ] . If M = max 1 m n 1 M * / M m , then we have
m = 1 n 1 | a n m , n n | U n m T β β 1 ( n 1 ) M * U 1 M 1 + + M * U n 1 M n 1 = T β β 1 ( n 1 ) m = 1 n 1 M * U m M m M T β ( n 1 ) β m = 1 n 1 U n m .
This concludes the proof. □
Theorem 2.
Let U n be the solutions of Equation (13); if τ β / 2 / h 2 C 1 < and τ β / 2 N β C 2 < , then for all τ > 0 , the following estimate holds:
τ n = 0 m U n C T U 0 + τ n = 0 m max 0 n m f n ,
where C, C 1 and C 2 are constants, and C and C 2 are dependent on T.
Proof. 
From Equation (13), we have
H 1 τ α k = 0 n w k ( α ) U n k = γ ( H y δ x 2 + H x δ y 2 ) U n + k = 0 n a n k , n n Γ ( β ) ( H y δ x 2 + H x δ y 2 ) U n k + H f n .
Then, if we take the inner product of Equation (14) with τ U n , we can obtain the following:
τ 1 α k = 0 n w k ( α ) ( H U n k , U n ) = γ τ ( H y δ x 2 U n , U n ) + γ τ ( H x δ y 2 U n , U n ) + τ Γ ( β ) k = 0 n a n k , n n ( H y δ x 2 U n k , U n ) + ( H x δ y 2 U n k , U n ) + τ ( H f n , U n ) .
Since the compact difference operators H x and H y are positive definite and self-joint, there are two positive operators L x and L y , so ( H u , w ) = ( L x L y u , L x L y w ) = ( Q u , Q w ) . Then, for the left term of Equation (15), we have
τ 1 α k = 0 n w k ( α ) ( H U n k , U n ) = τ 1 α k = 0 n w k ( α ) ( Q U n k , Q U n ) .
For the first term on the right in Equation (15), with Lemmas 6 and 8, we have
γ τ ( H y δ x 2 U n , U n ) + γ τ ( H x δ y 2 U n , U n ) = γ τ ( H y δ x U n , δ x U n ) γ τ ( H x δ y U n , δ y U n ) = γ τ δ x U n H y 2 + δ y U n H x 2 2 γ τ 3 δ x U n 2 + δ y U n 2 4 γ τ ( L 2 2 + L 1 2 ) ( L 1 2 L 2 2 ) U n 2 .
For the second term on the right in Equation (15), with Lemmas 6 and 8, we have
τ Γ ( β ) k = 0 n a n k , n n ( H y δ x 2 U n k , U n ) + ( H x δ y 2 U n k , U n ) = τ Γ ( β ) k = 1 n a n k , n n ( H y δ x 2 U n k , U n ) + ( H x δ y 2 U n k , U n ) + τ a n , n n Γ ( β ) ( H y δ x 2 U n , U n ) + ( H x δ y 2 U n , U n ) = τ Γ ( β ) k = 1 n a n k , n n ( H y δ x 2 U n k , U n ) + ( H x δ y 2 U n k , U n ) τ a n , n n Γ ( β ) δ x U n H y 2 + δ y U n H x 2 τ Γ ( β ) k = 1 n a n k , n n ( H y δ x 2 U n k , U n ) + ( H x δ y 2 U n k , U n ) .
With Lemma 8, Lemma 9, and the Cauchy–Schwarz inequality, we have
τ Γ ( β ) k = 1 n a n k , n n ( H y δ x 2 U n k , U n ) + ( H x δ y 2 U n k , U n ) = τ Γ ( β ) k = 1 n a n k , n n ( H y δ x U n k , δ x U n ) + ( H x δ y U n k , δ y U n ) τ Γ ( β ) k = 1 n | a n k , n n | H y δ x U n k δ x U n + H x δ y U n k δ y U n τ Γ ( β ) k = 1 n | a n k , n n | δ x U n k δ x U n + δ y U n k δ y U n 4 τ Γ ( β ) k = 1 n | a n k , n n | U n k U n h x 2 + U n k U n h y 2 .
Let h = min { h x , h y } ; thus, we have
4 τ Γ ( β ) k = 1 n | a n k , n n | U n k U n h x 2 + U n k U n h y 2 8 τ Γ ( β ) h 2 k = 1 n | a n k , n n | U n k U n 8 C 1 τ 1 β 2 Γ ( β ) k = 1 n | a n k , n n | U n k U n .
By inequality (16), inequality (17), and inequality (18), we get
τ 1 α k = 0 n w k ( α ) ( Q U n k , Q U n ) + 4 γ τ ( L 2 2 + L 1 2 ) ( L 1 2 L 2 2 ) U n 2 8 C 1 τ 1 β 2 Γ ( β ) k = 1 n | a n k , n n | U n k U n + τ f n U n .
After deleting factor U n , the following inequality holds:
τ 1 α U n k = 0 n w k ( α ) ( Q U n k , Q U n ) + 4 γ τ ( L 2 2 + L 1 2 ) ( L 1 2 L 2 2 ) U n 8 C 1 τ 1 β 2 Γ ( β ) k = 1 n | a n k , n n | U n k + τ f n .
Summing up inequality (19) for n from 0 to m, the above inequality becomes the following form:
τ 1 α n = 0 m k = 0 n w k ( α ) U n ( Q U n k , Q U n ) + 4 γ τ ( L 2 2 + L 1 2 ) ( L 1 2 L 2 2 ) n = 0 m U n 8 C 1 τ 1 β 2 Γ ( β ) n = 0 m k = 1 n | a n k , n n | U n k + τ n = 0 m f n .
From Lemma 7 and inequality (20), we have
τ n = 0 m U n 2 C 1 τ 1 β 2 L 1 2 L 2 2 γ Γ ( β ) ( L 1 2 + L 2 2 ) n = 0 m k = 1 n | a n k , n n | U n k + τ L 1 2 L 2 2 4 γ ( L 1 2 + L 2 2 ) n = 0 m f n .
With Lemma 5, Lemma 12, and inequality (21), we have
τ n = 0 m U n = 2 C 1 τ 1 β 2 L 1 2 L 2 2 γ Γ ( β ) ( L 1 2 + L 2 2 ) n = 0 m k = 1 n 1 | a n k , n n | U n k + | a 0 , n n | U 0 + τ γ 1 L 1 2 L 2 2 4 ( L 1 2 + L 2 2 ) n = 0 m f n 2 C 1 τ 1 β 2 L 1 2 L 2 2 γ Γ ( β ) ( L 1 2 + L 2 2 ) n = 0 m M T β ( n 1 ) β k = 1 n 1 U n k + τ β U 0 β + 1 + τ L 1 2 L 2 2 4 γ ( L 1 2 + L 2 2 ) n = 0 m f n k = 1 n 1 2 τ C 1 C 2 M L 1 2 L 2 2 γ Γ ( β + 1 ) ( n 1 ) ( L 1 2 + L 2 2 ) n = 0 m U n k + n = 0 m 2 τ β 2 + 1 C 1 L 1 2 L 2 2 γ Γ ( β ) ( β + 1 ) ( L 1 2 + L 2 2 ) U 0 + τ L 1 2 L 2 2 4 γ ( L 1 2 + L 2 2 ) max 0 n m f n .
Let b n = ( 2 τ β / 2 + 1 C 1 L 1 2 L 2 2 γ Γ ( β ) ( β + 1 ) ( L 2 2 + L 2 2 ) ) n = 0 m U 0 + τ L 1 2 L 2 2 4 γ ( L 1 2 + L 2 2 ) n = 0 m max 0 n m f n ; thus, we can check if b n is a positive non-decreasing sequence concerning n. From Lemma 10 and inequality (22), we have
τ n = 0 m U n ( 2 τ β 2 + 1 C 1 L 1 2 L 2 2 γ Γ ( β ) ( β + 1 ) ( L 2 2 + L 2 2 ) n = 0 m U 0 + τ L 1 2 L 2 2 4 γ ( L 1 2 + L 2 2 ) n = 0 m max 0 n m f n ) exp 2 C 1 C 2 M L 1 2 L 2 2 γ Γ ( β + 1 ) ( n 1 ) ( L 2 2 + L 2 2 ) .
Let C = max { 2 τ β / 2 C 1 L 1 2 L 2 2 γ Γ ( β ) ( β + 1 ) ( L 2 2 + L 2 2 ) , L 1 2 L 2 2 4 γ ( L 1 2 + L 2 2 ) } exp ( 2 C 1 C 2 M L 1 2 L 2 2 γ Γ ( β + 1 ) ( n 1 ) ( L 2 2 + L 2 2 ) ) ; thus, the following inequality holds:
τ n = 0 m U n C T U 0 + τ n = 0 m max 0 n m f n .
This concludes the proof. □
Let e i j n = u i j n U i j n { e i j n | 0 i M 1 , 0 j M 2 , and 0 n N } . Now, we consider the convergence theorem of the proposed algorithm.
Theorem 3.
Assume that u n C x , y , t 6 , 6 , 2 ( [ 0 , L 1 ] × [ 0 , L 2 ] × [ 0 , T ] ) and U n Ω ¯ h be the analytical solution and approximation solutions of Equation (1). If τ β / 2 / h 2 C 1 < and τ β / 2 N β C 2 < , then the following holds:
τ n = 0 m e n C T ( τ 2 + h x 4 + h y 4 ) ,
where C, C 1 , and C 2 are constants, and C and C 2 are dependent on T.
Proof. 
With the definition of e n , we can obtain
H 1 τ α k = 0 n w k ( α ) e n k = γ ( H y δ x 2 + H x δ y 2 ) e n + k = 0 n a n k , n n Γ ( β ) ( H y δ x 2 + H x δ y 2 ) e n k + R n .
From Theorem 2 and Equation (23), the following inequality holds:
τ n = 0 m e n C T e 0 + τ n = 0 m R n C τ n = 0 m R n C T ( τ 2 + h x 4 + h y 4 ) .
This concludes proving. □

4. Numerical Examples

To support our theoretical analysis, two tests are used in this section. All tests were carried out in MATLAB 2020 using the AMD Ryzen 4700U with 2.00 GHz and 16 GB of RAM as the hardware configuration.
The L norm error (one can refer to [47]) between the exact and the numerical solutions is shown in the following tables:
Error L = max 1 n N | u ( x i , y j , t n ) U i j n | ,
where u ( x i , y j , t n ) and U i j n are the exact solution and numerical solution, respectively. In addition, the time convergence rate is
Rate t = l o g ( E 1 t / E 2 t ) / l o g ( τ 1 / τ 2 ) ,
where E 1 t and E 2 t are errors that correspond to the grids with mesh sizes τ 1 and τ 2 in time, respectively. Moreover, the space convergence rate is
Rate x = l o g ( E 1 x / E 2 x ) / l o g ( h 1 / h 2 ) ,
where E 1 x and E 2 x are errors corresponding to the grids with mesh sizes h 1 and h 2 in space, respectively.
Example 1.
For the first model, we use h = h x = h y , T = 1 , Ω = [ 0 , 1 ] × [ 0 , 1 ] and γ = 1 with the exact solution as
u ( x , y , t ) = t 4 x 4 ( 1 x ) 4 y 4 ( 1 y ) 4 ,
then, we consider the following force term:
f ( x , y , t ) = 24 t 4 α Γ ( 5 α ) ( x 4 ( 1 x ) 4 y 4 ( 1 y ) 4 ) t 4 + 24 t β + 4 Γ ( 5 + β ) ( y 4 ( 1 y ) 4 ( 12 x 2 ( 1 x 4 ) 32 x 3 ( 1 x ) 3 + 12 x 4 ( 1 x ) 2 ) + ( x 4 ( 1 x ) 4 ) ( 12 y 2 ( 1 y 4 ) 32 y 3 ( 1 y ) 3 + 12 y 4 ( 1 y ) 2 ) ) ,
The above condition Ω = [ 0 , 1 ] × [ 0 , 1 ] means that M 1 = M 2 . Let M = M 1 = M 2 ; Table 1, Table 2, Table 3, Table 4, Table 5 and Table 6 present errors and computational orders for Example 1. Table 1, Table 2 and Table 3 confirm the second-order accuracy in the temporal direction. In Table 1, by setting M = 32 , β = 0.2 , we obtain a different α and N. Similar to β = 0.2 , the results of β = 0.5 and β = 0.8 under the same conditions are shown in Table 2 and Table 3, respectively. It can be seen that the numerical results agree with the theoretical ones. Table 4, Table 5 and Table 6 confirm the fourth-order accuracy in the spatial direction. For N = 400 , in Table 4, by setting β = 0.2 , we obtain a different α and M. Similar to β = 0.2 , the results of β = 0.5 and β = 0.8 under the same conditions are shown in Table 5 and Table 6, respectively. The numerical results obtained are consistent with those obtained theoretically.
In Figure 1a, by setting β = 0.5 , we obtain a different α and demonstrate the results with the L -norm, which attains the accuracy in the temporal direction with O ( τ 2 ) . In Figure 1b, by setting β = 0.2 , we obtain a different α . This demonstrates the results in the L -norm, which attains the accuracy in the spatial direction with O ( h 4 ) . In Figure 2, by setting M 1 = M 2 = 100 and N = 100 , this demonstrates a graph comprising a solution and error with α = β = 0.99 .
For the compact finite difference scheme, the desired accuracy in the temporal direction and the spatial direction can achieve O ( τ 2 ) and O ( h 4 ) for any parameter α and β , which implies that the numerical solution has favorable stability.
Example 2.
For the second example, we consider h = h x = h y , T = 1 , Ω = [ 0 , 1 ] × [ 0 , 1 ] and γ = 1 . This example is used to test the case that the exact solution is unknown. In this example, the initial term is φ ( x , y ) = s i n ( π x ) s i n ( π y ) and the force term is f ( x , y , t ) = 0 .
Table 7, Table 8 and Table 9 confirm the computational orders in the temporal direction with O ( τ 2 ) . In order to verify the accuracy and convergence rate in the temporal direction, we consider the solution on the fine grid ( N = 300 and M = 32 ) as the exact solution. The solutions on coarse grids ( M = 32 and different N) are used as numerical solutions. According to the data in Table 7, Table 8 and Table 9, it can be clearly seen that the second-order accuracy in the temporal direction is verified. Table 10, Table 11 and Table 12 show that the computational order in the spatial direction is O ( h 4 ) . In order to verify the accuracy and convergence rate in the spatial direction, we relate to the solution on the fine grid ( N = 500 and M = 64 ) as the exact solution. The solutions on coarse grids ( N = 500 and different M = 4 , 8 , 16 , 32 ) are used as numerical solutions. According to the data in Table 10, Table 11 and Table 12, it can be clearly seen that the fourth-order accuracy in the spatial direction is verified.
The results of the second example show that the proposed method is still effective as the exact solution is unknown. As the exact solution of this example is unknown, the orders of convergence in the computed solutions can be estimated by using the two-mesh principle; for more detailed information about the two-mesh principle, one can refer to [48,49]. This approach can enhance the reliability of the results, and we will use the two-mesh principle in our future work.

5. Conclusions

A new algorithm for 2D time-fractional integro-differential equations is studied in this manuscript, which is based on a compact difference operator. The time-fractional derivative was approximated by the weighted and shifted Grünwald difference operator; meanwhile, the integral term was approximated with a product of the trapezoidal formula. This paper also discussed the stability and error bounds of the presented method. Numerical investigations showed that the compact finite difference scheme converged with order O ( τ 2 + h x 4 + h y 4 ) , which is in excellent agreement with our theoretical analysis. Note that the numerical accuracy of our proposed method does not depend on parameters α and β . The scientific content of this article is based on uniform meshes. Non-uniform meshes are very meaningful and interesting, and we will study then deeply in the future. In our future work, we would like to investigate other fractional derivatives in temporal and spatial directions, as well as algorithms comprising nonlinear and multi-kernel fractional integro-differential equations.

Author Contributions

Conceptualization, X.Z.; software, Q.T. and L.W.; formal analysis, X.Z., Z.L. and J.L.; investigation, Q.T.; writing—original draft preparation, Z.L.; writing—review and editing, X.Z. and J.L.; visualization, L.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Scientific Research Foundation for Talents Introduced of Guizhou University of Finance and Economics (No. 2023YJ16) and Natural Science Foundation of Xinjiang Uygur Autonomous Region (No. 2022D01E13).

Data Availability Statement

The data analyzed in this study are subject to the following licenses/restrictions: the first author can receive the restrictions. Requests to access these datasets should be directed to liaoyuan1126@163.com.

Acknowledgments

The authors are very grateful to the referee for carefully reading the article and for many valuable comments.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Gurtin, M.E.; Pipkin, A.C. A general theory of heat conduction with finite wave speed. Arch. Ration. Mech. Anal. 1968, 31, 113–126. [Google Scholar] [CrossRef]
  2. Miller, R.K. An integro-differential equation for gird heat conductors with memory. J. Math. Anal. Appl. 1978, 66, 313–332. [Google Scholar] [CrossRef]
  3. Schneider, W.R.; Wyss, W. Fractional diffusion and wave equations. J. Math. Phys. 1989, 30, 134–144. [Google Scholar] [CrossRef]
  4. Kumar, S.; Kumar, A.; Samet, B.; Gómez-Aguilar, J.F.; Osman, M.S. A chaos study of tumor and effector cells in fractional tumor-immune model for cancer treatment. Chaos Solitons Fractals 2020, 141, 110321. [Google Scholar] [CrossRef]
  5. Kumar, S.; Kumar, R.; Osman, M.S.; Samet, B. A wavelet based numerical scheme for fractional order SEIR epidemic of measles by using genocchi polynomials. Numer. Methods Partial. Differ. Equ. 2021, 37, 1250–1268. [Google Scholar] [CrossRef]
  6. Güner, Ö.; Bekir, A. Exact solutions of some fractional differential equations arising in mathematical biology. Int. J. Biomath. 2015, 8, 1550003. [Google Scholar] [CrossRef]
  7. Oldham, K.B. Fractional differential equations in electrochemistry. Adv. Eng. Softw. 2010, 41, 9–12. [Google Scholar] [CrossRef]
  8. López-Marcos, J.C. A difference scheme for a nonliner partial integro-differential equation. SIAM J. Numer. Anal. 1990, 27, 20–31. [Google Scholar] [CrossRef]
  9. Tang, T. A finite difference sheme for partial integro-differential equations with a weakly singular kernel. Appl. Numer. Math. 1993, 11, 309–319. [Google Scholar] [CrossRef]
  10. Li, M.; Gu, X.-M.; Huang, C.; Fei, M.; Zhang, G. A fast linearized conservative finite element method for the strongly coupled nonlinear fractional Schrödinger equations. J. Comput. Phys. 2018, 358, 256–282. [Google Scholar] [CrossRef]
  11. Wang, K. A two-gird method for finite element solution of parabolic integro-differential equations. J. Appl. Math. Comput. 2022, 68, 3473–3490. [Google Scholar] [CrossRef]
  12. Tan, Z.J.; Li, K.; Chen, Y.P. A fully discrete two-grid finite element method for nonlinear hyperbolic integro-differential equation. Appl. Math. Comput. 2022, 413, 126596. [Google Scholar] [CrossRef]
  13. Zhang, X.D.; Yao, L. Numerical approximation of time-dependent fractional convection-diffusion-wave equation by RBF-FD method. Eng. Anal. Bound. Elem. 2021, 130, 1–9. [Google Scholar] [CrossRef]
  14. Zhang, X.D.; Feng, Y.L.; Luo, Z.Y.; Liu, J. A spatial sixth-order numerical scheme for solving fractional partial differential equation. Appl. Math. Lett. 2025, 159, 109265. [Google Scholar] [CrossRef]
  15. Cotta, R.M.; Mikhailov, M.D. Integral transform method. Appl. Math. Model. 1993, 17, 156–161. [Google Scholar] [CrossRef]
  16. Pinheiro, I.F.; Sphaier, L.A.; De B. Alves, L.S. Integral transform solution of integro-differential equations in conduction-radiation problems. Numer. Heat Transf. Part A Appl. 2018, 73, 94–114. [Google Scholar] [CrossRef]
  17. Heymans, N.; Podlubny, I. Physical interpretation of initial conditions for fractional differential equations with Riemann-Liouville fractional derivatives. Rheol. Acta 2006, 45, 765–771. [Google Scholar] [CrossRef]
  18. Mohammadi, H.; Kumar, S.; Rezapour, S.; Etemad, S. A theoretical study of the Caputo-Fabrizio fractional modeling for hearing loss due to mumps virus with optimal control. Chaos Solitons Fractals 2021, 144, 110668. [Google Scholar] [CrossRef]
  19. Singh, D.; Sultana, F.; Pandey, R.K.; Atangana, A. A comparative study of three numerical schemes for solving Atangana-Baleanu fractional integro-differential equation defined in Caputo sense. Eng. Comput. 2022, 38, 149–168. [Google Scholar] [CrossRef]
  20. Rawani, M.K.; Verma, A.K.; Cattani, C. A novel hybrid approach for computing numerical solution of the time-fractional nonlinear one and two-dimensional partial integro-differential equation. Commun. Nonlinear Sci. Numer. Simul. 2023, 118, 106986. [Google Scholar] [CrossRef]
  21. Cao, Y.; Nikan, O.; Avazzadeh, Z. A localized meshless technique for solving 2D nonlinear integro-differential equation with multi-term kernels. Appl. Numer. Math. 2023, 183, 140–156. [Google Scholar] [CrossRef]
  22. Fakhar-Izadi, F. Fully spectral-galerkin method for the one-and two-dimensional fourth-order time-fractional partial integro-differential equations with a weakly singular kernel. Numer. Methods Partial. Differ. Equ. 2022, 38, 160–176. [Google Scholar] [CrossRef]
  23. Kamran, K.; Shah, Z.; Kumam, P.; Alreshidi, N.A. A meshless method based on the laplace transform for the 2D multi-term time fractional partial integro-differential equation. Mathematics 2020, 8, 1972. [Google Scholar] [CrossRef]
  24. Qiao, L.J.; Wang, Z.B.; Xu, D. An alternating direction implicit orthogonal spline collocation method for the two dimensional multi-term time fractional integro-differential equation. Appl. Numer. Math. 2020, 151, 199–212. [Google Scholar] [CrossRef]
  25. Dehghan, M.; Abbaszadeh, M. Error estimate of finite element/finite difference technique for solution of two-dimensional weakly singular integro-partial differential equation with space and time fractional derivatives. J. Comput. Appl. Math. 2019, 356, 314–328. [Google Scholar] [CrossRef]
  26. Qiao, L.J.; Xu, D.; Wang, Z.B. An ADI difference scheme based on fractional trapezoidal rule for fractional integro-differential equation with a weakly singular kernel. Appl. Math. Comput. 2019, 354, 103–114. [Google Scholar] [CrossRef]
  27. Wang, D.; Liu, Y.; Li, H.; Fang, Z.C. Second-order time stepping scheme combined with a mixed element method for a 2D nonlinear fourth-order fractional integro-differential equations. Fractal Fract. 2022, 6, 201. [Google Scholar] [CrossRef]
  28. Huang, Q.; Qiao, L.J.; Tang, B. High-order orthogonal spline collocation ADI scheme for a new complex two-dimensional distributed-order fractional integro-differential equation with two weakly singular kernels. Int. J. Comput. Math. 2023, 100, 703–721. [Google Scholar] [CrossRef]
  29. Wang, Z.B.; Cen, D.; Mo, Y. Sharp error estimate of a compact L1-ADI scheme for the two-dimensional time-fractional integro-differential equation with singular kernels. Appl. Numer. Math. 2021, 159, 190–203. [Google Scholar] [CrossRef]
  30. Wang, Z.B.; Liang, Y.X.; Mo, Y. A novel high order compact ADI scheme for two dimensional fractional integro-differential equations. Appl. Numer. Math. 2021, 167, 257–272. [Google Scholar] [CrossRef]
  31. Gu, X.-M.; Wu, S.L. A parallel-in-time iterative algorithm for Volterra partial integro-differential problems with weakly singular kernel. J. Comput. Phys. 2020, 417, 109576. [Google Scholar] [CrossRef]
  32. Podlubny, I. Fractional Differential Equations; Elsevier: New York, NY, USA, 1999. [Google Scholar]
  33. Liao, H.-L.; Liu, N.; Lyu, T. Discrete gradient structure of a second-order variable-step method for nonlinear integro-differential models. SIAM J. Numer. Anal. 2023, 61, 2157–2181. [Google Scholar] [CrossRef]
  34. Chen, H.B.; Xu, D. A Compact difference scheme for an evolution equation with a weakly singular kernel. Numer. Math.-Theory Methods Appl. 2012, 5, 559–572. [Google Scholar]
  35. Meerschaert, M.M.; Tadjeran, C. Finite difference approximations for fractional advection-dispersion flow equations. J. Comput. Appl. Math. 2004, 172, 65–77. [Google Scholar] [CrossRef]
  36. Tian, W.Y.; Zhou, H.; Deng, W.H. A class of second order difference approximations for solving space fractional diffusion equations. Math. Comput. 2015, 84, 1703–1727. [Google Scholar] [CrossRef]
  37. Gao, G.H.; Sun, H.W.; Sun, Z.Z. Some high-order difference schemes for the distributed-order differential equations. J. Comput. Phys. 2015, 298, 337–359. [Google Scholar] [CrossRef]
  38. Hao, Z.P.; Sun, Z.Z.; Cao, W.R. A three-level linearized compact difference scheme for the Ginzburg-Landau equation. Numer. Methods Partial. Differ. Equ. 2015, 31, 876–899. [Google Scholar] [CrossRef]
  39. Liao, H.L.; Sun, Z.Z. Maximum norm error bounds of ADI and compact ADI methods for solving parabolic equations. Numer. Methods Partial. Differ. Equ. 2010, 26, 37–60. [Google Scholar] [CrossRef]
  40. Diethelm, K.; Ford, N.J.; Freed, A.D. Detailed error analysis for a fractional adams method. Numer. Algorithms 2004, 36, 31–52. [Google Scholar] [CrossRef]
  41. Guo, J.; Xu, D.; Qiu, W.L. A finite difference scheme for the nonlinear time-fractional partial integro-differential equation. Math. Methods Appl. Sci. 2020, 43, 3392–3412. [Google Scholar] [CrossRef]
  42. Qiao, L.J.; Xu, D. Compact alternating direction implicit scheme for integro-differential equations of parabolic type. J. Sci. Comput. 2018, 76, 565–582. [Google Scholar] [CrossRef]
  43. Wang, Z.B.; Vong, S.K. Compact difference schemes for the modified anomalous fractional sub-diffusion equation and the fractional diffusion-wave equation. J. Comput. Physics 2014, 277, 1–15. [Google Scholar] [CrossRef]
  44. Mohebbi, A. Compact finite difference scheme for the solution for a time faractional partial integro-differential equation with a weakly singular kernel. Math. Methods Appl. Sci. 2017, 40, 7627–7639. [Google Scholar] [CrossRef]
  45. Sloan, I.H.; Thomée, V. Time discretization of an integro-differential equation of parabolic type. SIAM J. Numer. Anal. 1986, 23, 1052–1061. [Google Scholar] [CrossRef]
  46. Mitrinović, D.S.; Pečarić, J.E.; Fink, A.M. Classical and New Inequalities in Analysis; Springer: New York, NY, USA, 1993. [Google Scholar]
  47. Maleknejad, K.; Rashidinia, J.; Eftekhari, T. Operational matrices based on hybrid functions for solving general nonlinear two-dimensional fractional integro-differential equations. Comput. Appl. Math. 2020, 39, 1–34. [Google Scholar] [CrossRef]
  48. Stynes, M.; O’Riordan, E.; Gracia, J.L. Error analysis of a finite difference method on graded meshes for a time-fractional diffusion equation. SIAM J. Numer. Anal. 2017, 55, 1057–1079. [Google Scholar] [CrossRef]
  49. Farrell, P.; Hegarty, A.; Miller, J.M.; O’Riordan, E.; Shishkin, G.I. Robust Computational Techniques for Boundary Layers; Appl. Math. 16; Chapman & Hall/CRC: Boca Raton, FL, USA, 2000. [Google Scholar]
Figure 1. Graph of an error with a different α and β for Example 1: (a) β = 0.5 at different α ; (b) β = 0.2 at different α .
Figure 1. Graph of an error with a different α and β for Example 1: (a) β = 0.5 at different α ; (b) β = 0.2 at different α .
Fractalfract 08 00495 g001
Figure 2. Results of Example 1 with α = β = 0.99 , M 1 = M 2 = 100 and N = 100 : (a) exact solution; (b) numerical solution; (c) absolute error; (d) contour plot of absolute error.
Figure 2. Results of Example 1 with α = β = 0.99 , M 1 = M 2 = 100 and N = 100 : (a) exact solution; (b) numerical solution; (c) absolute error; (d) contour plot of absolute error.
Fractalfract 08 00495 g002
Table 1. Example 1: Errors—computational orders with M = 32 and β = 0.2 .
Table 1. Example 1: Errors—computational orders with M = 32 and β = 0.2 .
N α = 0.2 α = 0.5 α = 0.8
Error L Rate t Error L Rate t Error L Rate t
709.1433 × 10−108.1616 × 10−106.6074 × 10   10
807.0624 × 10−101.936.2979 × 10−101.945.0900 × 10   10 1.95
905.6170 × 10   10 1.945.0017 × 10   10 1.964.0318 × 10   10 1.98
1004.5715 × 10   10 1.964.0631 × 10   10 1.973.2638 × 10   10 2.01
Table 2. Example 1: Errors—computational orders with M = 32 and β = 0.5 .
Table 2. Example 1: Errors—computational orders with M = 32 and β = 0.5 .
N α = 0.2 α = 0.5 α = 0.8
Error L Rate t Error L Rate t Error L Rate t
502.0804 × 10   9 1.8721 × 10   9 1.5490 × 10   9
601.4484 × 10   9 1.991.3016 × 10   9 1.991.0744 × 10   9 2.01
701.0651 × 10   9 1.999.5551 × 10   10 2.017.8626 × 10   10 2.03
808.1516 × 10   10 2.007.2984 × 10   10 2.025.9828 × 10   10 2.05
Table 3. Example 1: Errors—computational orders with M = 32 and β = 0.8 .
Table 3. Example 1: Errors—computational orders with M = 32 and β = 0.8 .
N α  = 0.2 α  = 0.5 α  = 0.8
Error L Rate t Error L Rate t Error L Rate t
104.3329 × 10   8 3.8059 × 10   8 2.9837 × 10   8
201.0931 × 10   8 1.999.5612 × 10   9 1.997.4416 × 10   9 2.00
304.8663 × 10   9 2.004.2472 × 10   9 2.003.2920 × 10   9 2.01
402.7359 × 10   9 2.002.3829 × 10   9 2.011.8394 × 10   9 2.02
Table 4. Example 1: Errors—computational orders with N = 400 and β = 0.2 .
Table 4. Example 1: Errors—computational orders with N = 400 and β = 0.2 .
M α  = 0.2 α  = 0.5 α  = 0.8
Error L Rate x Error L Rate x Error L Rate x
121.6977 × 10   8 1.6792 × 10   8 1.6534 × 10   8
165.4909 × 10   9 3.925.4417 × 10   9 3.925.3726 × 10   9 3.91
202.2737 × 10   9 3.952.2499 × 10   9 3.962.2164 × 10   9 3.97
241.0913 × 10   9 4.031.0808 × 10   9 4.021.0660 × 10   9 4.02
Table 5. Example 1: Errors—computational orders with N = 400 and β = 0.5 .
Table 5. Example 1: Errors—computational orders with N = 400 and β = 0.5 .
M α  = 0.2 α  = 0.5 α  = 0.8
Error L Rate x Error L Rate x Error L Rate x
121.6913 × 10   8 1.6700 × 10   8 1.6405 × 10   8
165.4740 × 10   9 3.925.4172 × 10   9 3.915.3382 × 10   9 3.90
202.2657 × 10   9 3.962.2384 × 10   9 3.962.2002 × 10   9 3.97
241.0879 × 10   9 4.021.0759 × 10   9 4.021.0589 × 10   9 4.01
Table 6. Example 1: Errors—computational orders with N = 400 and β = 0.8 .
Table 6. Example 1: Errors—computational orders with N = 400 and β = 0.8 .
M α  = 0.2 α  = 0.5 α  = 0.8
Error L Rate x Error L Rate x Error L Rate x
121.6856 × 10   8 1.6617 × 10   8 1.6287 × 10   8
165.4588 × 10   9 3.925.3952 × 10   9 3.915.3069 × 10   9 3.90
202.2583 × 10   9 3.962.2276 × 10   9 3.962.1849 × 10   9 3.98
241.0847 × 10   9 4.021.0712 × 10   9 4.021.0523 × 10   9 4.01
Table 7. Example 2: Errors—computational orders with M = 32 and α = 0.1 .
Table 7. Example 2: Errors—computational orders with M = 32 and α = 0.1 .
N β  = 0.1 β  = 0.5 β  = 0.9
Error L Rate t Error L Rate t Error L Rate t
1307.9699 × 10   5 3.2922 × 10   4 7.1783 × 10   4
1406.9555 × 10   5 1.842.8754 × 10   4 1.836.2756 × 10   4 1.81
1506.0783 × 10   6 2.002.5147 × 10   4 1.955.4930 × 10   4 1.93
1605.3123 × 10   6 2.012.1993 × 10   4 2.014.8079 × 10   4 2.06
Table 8. Example 2: Errors—computational orders with M = 32 and α = 0.5 .
Table 8. Example 2: Errors—computational orders with M = 32 and α = 0.5 .
N β  = 0.1 β  = 0.5 β  = 0.9
Error L Rate t Error L Rate t Error L Rate t
1408.7400 × 10   5 4.8492 × 10   4 1.0980 × 10   3
1507.7033 × 10   5 1.834.2813 × 10   4 1.819.7004 × 10   4 1.80
1606.7865 × 10   5 1.963.7777 × 10   4 1.948.5644 × 10   4 1.93
1705.9694 × 10   5 2.113.3277 × 10   4 2.097.5481 × 10   4 2.08
Table 9. Example 2: Errors—computational orders with M = 32 and α = 0.9 .
Table 9. Example 2: Errors—computational orders with M = 32 and α = 0.9 .
N β  = 0.1 β  = 0.5 β  = 0.9
Error L Rate t Error L Rate t Error L Rate t
1501.3776 × 10   4 6.9117 × 10   4 1.5866 × 10   3
1601.2261 × 10   4 1.816.1698 × 10   4 1.801.4272 × 10   3 2.02
1701.0889 × 10   4 1.965.4946 × 10   4 1.911.2628 × 10   3 2.02
1809.6385 × 10   5 2.134.8762 × 10   4 2.081.1212 × 10   3 2.03
Table 10. Example 2: Errors—computational orders with N = 500 and α = 0.1 .
Table 10. Example 2: Errors—computational orders with N = 500 and α = 0.1 .
M β  = 0.1 β  = 0.5 β  = 0.9
Error L Rate x Error L Rate x Error L Rate x
43.4174 × 10   9 1.7859 × 10   8 4.5051 × 10   8
82.0959 × 10   10 4.031.0955 × 10   9 4.032.7635 × 10   9 4.03
161.2991 × 10   11 4.016.7905 × 10   11 4.011.7130 × 10   10 4.01
327.6333 × 10   13 4.083.9099 × 10   12 4.081.0065 × 10   11 4.08
Table 11. Example 2: Errors—computational orders with N = 500 and α = 0.5 .
Table 11. Example 2: Errors—computational orders with N = 500 and α = 0.5 .
M β  = 0.1 β  = 0.5 β  = 0.9
Error L Rate L Error L Rate L Error L Rate L
46.2105 × 10   8 5.9446 × 10   7 1.4047 × 10   6
83.8108 × 10   9 4.023.6463 × 10   8 4.038.6159 × 10   8 4.03
162.3622 × 10   10 4.012.2602 × 10   9 4.015.3406 × 10   9 4.01
321.3880 × 10   11 4.081.3280 × 10   10 4.083.1380 × 10   10 4.08
Table 12. Example 2: Errors—computational orders with N = 500 and α = 0.9 .
Table 12. Example 2: Errors—computational orders with N = 500 and α = 0.9 .
M β  = 0.1 β  = 0.5 β  = 0.9
Error L Rate L Error L Rate L Error L Rate L
41.5936 × 10   6 9.8327 × 10   6 2.3640 × 10   5
89.7745 × 10   8 4.036.0304 × 10   7 4.031.4498 × 10   6 4.03
166.0587 × 10   9 4.013.7379 × 10   8 4.018.9865 × 10   8 4.01
323.5599 × 10   10 4.082.1963 × 10   9 4.085.2802 × 10   9 4.08
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhang, X.; Luo, Z.; Tang, Q.; Wei, L.; Liu, J. Difference Approximation for 2D Time-Fractional Integro-Differential Equation with Given Initial and Boundary Conditions. Fractal Fract. 2024, 8, 495. https://doi.org/10.3390/fractalfract8080495

AMA Style

Zhang X, Luo Z, Tang Q, Wei L, Liu J. Difference Approximation for 2D Time-Fractional Integro-Differential Equation with Given Initial and Boundary Conditions. Fractal and Fractional. 2024; 8(8):495. https://doi.org/10.3390/fractalfract8080495

Chicago/Turabian Style

Zhang, Xindong, Ziyang Luo, Quan Tang, Leilei Wei, and Juan Liu. 2024. "Difference Approximation for 2D Time-Fractional Integro-Differential Equation with Given Initial and Boundary Conditions" Fractal and Fractional 8, no. 8: 495. https://doi.org/10.3390/fractalfract8080495

APA Style

Zhang, X., Luo, Z., Tang, Q., Wei, L., & Liu, J. (2024). Difference Approximation for 2D Time-Fractional Integro-Differential Equation with Given Initial and Boundary Conditions. Fractal and Fractional, 8(8), 495. https://doi.org/10.3390/fractalfract8080495

Article Metrics

Back to TopTop