Next Article in Journal
The Müntz–Legendre Wavelet Collocation Method for Solving Weakly Singular Integro-Differential Equations with Fractional Derivatives
Next Article in Special Issue
Finite Representations of the Wright Function
Previous Article in Journal
The Iterative Properties for Positive Solutions of a Tempered Fractional Equation
Previous Article in Special Issue
The Analytical Solutions to the Fractional Kraenkel–Manna–Merle System in Ferromagnetic Materials
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Numerical Solution of Advection–Diffusion Equation of Fractional Order Using Chebyshev Collocation Method

1
Department of Mathematics, Islamia College Peshawar, Jamrod Road, Peshawar 25120, Khyber PakhtunKhwa, Pakistan
2
Robotics and Internet-of-Things Laboratory, Prince Sultan University, Riyadh 11586, Saudi Arabia
3
RIADI Laboratory, National School of Computer Sciences, University of Manouba, Manouba 2010, Tunisia
4
Department of Mathematics and Sciences, Prince Sultan University, Riyadh 11586, Saudi Arabia
*
Authors to whom correspondence should be addressed.
Fractal Fract. 2023, 7(10), 762; https://doi.org/10.3390/fractalfract7100762
Submission received: 6 September 2023 / Revised: 9 October 2023 / Accepted: 13 October 2023 / Published: 17 October 2023

Abstract

:
This work presents a highly accurate method for the numerical solution of the advection–diffusion equation of fractional order. In our proposed method, we apply the Laplace transform to handle the time-fractional derivative and utilize the Chebyshev spectral collocation method for spatial discretization. The primary motivation for using the Laplace transform is its ability to avoid the classical time-stepping scheme and overcome the adverse effects of time steps on numerical accuracy and stability. Our method comprises three primary steps: (i) reducing the time-dependent equation to a time-independent equation via the Laplace transform, (ii) employing the Chebyshev spectral collocation method to approximate the solution of the transformed equation, and (iii) numerically inverting the Laplace transform. We discuss the convergence and stability of the method and assess its accuracy and efficiency by solving various problems in two dimensions.

1. Introduction

Fractional calculus (FC) is the generalization of classical calculus, which includes integrals and derivatives in non-integer order [1,2]. In recent years, the concept of fractional derivatives has been applied with great success to model various real-life phenomena in many scientific fields [3]. Fractional order operators include the history of a physical phenomenon from the initial state to the current state. Therefore, fractional order operators are often applied to model systems that describe the influence of memory effects [4]. Fractional order operators have broadened the concept of integer order operators [5]. In the last ten years, PDEs of fractional order have attracted the remarkable attention of the research community due to their ability to model various problems, such as neural networks [6], optimal voltage controllers for electronic vehicle charging stations [7], robotics [8], optimal controls for impulsive systems [9], hydrology [10], medical sciences [11], and many others [12].
The time-fractional advection–diffusion equations (TFADEs) have been widely used in many fields of science because of their ability to model memory and non-local properties [13]. ADEs combine diffusion and advection terms to describe physical events in which particles, energy, or other physical quantities are moved inside a physical system by two processes, i.e., diffusion and advection. The concentration of substances for mass and heat transfer is represented by the advection and diffusion models. This equation has been used to model heat transfer in draining films [14], flow in porous media [15], and air pollution [16], etc. There have been many attempts to solve the TFADEs accurately. In [17], a second-order implicit difference method was proposed for solving a time–space-fractional ADE. Mardani et al. [18] studied a meshless method for the solution of ADEs with variable coefficients. A fully implicit finite difference scheme based on cubic B-splines was developed in [19] for solving TFADEs. Abbaszadeh and Amjadian [20] studied the spectral element method for the solution of TFADEs. In the current study, we investigate the numerical solution of two-dimensional TFADEs using the Chebyshev spectral collocation method (CSCM) coupled with the Laplace transform (LT).
Spectral collocation methods have attracted the attention of the research community due to their highly accurate solutions to partial differential equations. They converge exponentially fast as compared to the algebraic convergence rates for FEM and FDM. Spectral methods are global in nature. This means that the optimal accuracy can be obtained with few grid points. For the first time, spectral methods were introduced by Gottlieb and Orszag [21]. Their work paved the way for more advanced extensions of the method for a larger variety of problems [22,23]. The most important tool for spectral methods is the computation of differentiation matrices, discussed briefly by Welfert [24]. The spectral collocation methods have been used by many authors. For example, in [25], the authors discussed the Hermite spectral methods for unbounded domains. The solution of the fractional advection–dispersion equation by the Chebyshev spectral method was incorporated by Sweilam et al. [26]. In [27], the spectral Legendre–Chebyshev collocation method for variable coefficients was examined. Tian et al. [28] investigated the polynomial spectral collocation method for space-fractional ADEs. The spectral collocation method and the non-standard finite difference technique for the solution of TFADEs were introduced by [29].
In the aforementioned methods, the time discretization is carried out using the finite difference time-stepping method. The main shortcoming of the time-stepping technique is that it may not always result in a stable solution. A finite difference time-stepping scheme is stable if the errors calculated at a one-time step do not cause the errors to be increased during the calculations. In other words, the time-stepping scheme is stable whether the errors remain constant or decay during the computations. Furthermore, for optimal accuracy, we need to decrease the time step, which results in an increased computational time, and, thus, has low efficiency in simulating the time-fractional problems. To overcome this drawback of the finite difference time-stepping scheme, the LT has been used as one of the best alternative mathematical tools [30,31]. However, while using the LT, the main difficulty is its inversion. In the literature, many algorithms have been developed for the numerical inversion of the LT [32,33,34]. In the current study, we use Stehfest’s method [35] and Talbot’s method [32,36] for the numerical inversion of the LT.

2. Numerical Scheme

In our numerical scheme, a two-dimensional TFADE is taken into account. First, the LT is employed to reduce the TFADE to a time-independent problem in the LT domain. Then, the CSCM is employed to obtain the approximate solution to the transformed problem. Finally, the time domain solution is retrieved from the LT domain solution via the inverse LT.

2.1. Two-Dimensional TFADE

We consider a two-dimensional time-fractional advection–diffusion equation of the form
D τ α U ( x ¯ , τ ) = λ 1 Δ U ( x ¯ , τ ) + ( u · ) U ( x ¯ , τ ) λ 2 U ( x ¯ , τ ) + Q ( x ¯ , τ ) , x ¯ Ω , 0 τ 1 ,
with boundary conditions
U ( x ¯ , τ ) = h ( x ¯ , τ ) , x ¯ Ω , 0 τ 1 ,
and the initial condition
U ( x ¯ , 0 ) = U 0 ( x ¯ ) , x ¯ Ω ,
where   u = ( u 1 , u 2 ) is a known vector, λ 1 and λ 2 are arbitrary constants, Q ( x ¯ , τ ) , U 0 ( x ¯ ) , h ( x ¯ , τ ) are given continuous functions, Ω R 2 is a bounded domain with a smooth boundary Ω , and D τ α represents Caputo’s derivative, defined as [3]
D τ α U ( x ¯ , τ ) = 1 Γ ( 1 α ) 0 τ U ( x ¯ , z ) z ( τ z ) α d z , 0 < α 1 .

2.2. Time Discretization

In this section, the LT is used to reduce the TFADE to an equivalent time-independent problem in the LT domain. The LT of the function U ( x ¯ , τ ) is denoted by U ^ ( x ¯ , s ) and is defined as
U ^ ( x ¯ , s ) = 0 e s τ U ( x ¯ , τ ) d τ ,
and the LT of Caputo’s derivative is defined as
L D τ α U ( x ¯ , τ ) = s α U ^ ( x ¯ , s ) s α 1 U 0 .
Applying the LT to (1)–(3) gives
s α U ^ ( x ¯ , s ) s α 1 U 0 = λ 1 Δ U ^ ( x ¯ , s ) + ( u · ) U ^ ( x ¯ , s ) λ 2 U ^ ( x ¯ , s ) + Q ^ ( x ¯ , s ) ,
and
U ^ ( x ¯ , s ) = h ^ ( x ¯ , s ) ,
Equations (4) and (5) can be rearranged as
( s α I L ) U ^ ( x ¯ , s ) = G ^ ( x ¯ , s ) ,
and
U ^ ( x ¯ , s ) = h ^ ( x ¯ , s ) .
where L = λ 1 Δ + ( u · ) λ 2 I is a linear differential operator, G ^ ( x ¯ , s ) = s α 1 U 0 + Q ^ ( x ¯ , s ) , and I denotes the identity operator. To obtain the approximate solution of the system defined in (6) and (7), first, the linear spatial operator L is discretized using the CSCM. Then, the fully discrete system is solved for each node s in the LT domain. Finally, the numerical inversion of the LT is performed to obtain the solution of the problem defined in Equations (1)–(3).

2.3. Chebyshev Spectral Collocation Method

In the CSCM, we interpolate the data x j , U ^ ( x j ) by using the Lagrange interpolation polynomial (LIP) l j ( x ) of a degree of at most n [23,37]
I n ( x ) = j = 0 n l j ( x ) U ^ j
where l j ( x ) is the LIP at the point x j ( j = 0 , 1 , , n ) , given as
l j ( x ) = ( x x 0 ) ( x x j 1 ) ( x x j + 1 ) ( x x n ) ( x j x 0 ) ( x j x j 1 ) ( x j x j + 1 ) ( x j x n )
where U ^ j = U ^ ( x j ) . In the CSCM, the Chebyshev nodes are selected as the interpolation nodes. The domain [ 1 , 1 ] is discretized using the Chebyshev points defined as
x j = cos j π n , j = 0 , 1 , , n .
The approximation of the first derivative U ^ ( x ) x on the Chebyshev points can be obtained as
U ^ ( x ) x D n U ^ ,
where the entries of the matrix D n are
[ D n ] i , j = l j ( x i ) , i , j = 1 , 2 , , n .
The off-diagonal entries of [ D n ] i , j are obtained as
[ D n ] i , j = α j α i ( x i x j ) , i j ,
where α j 1 = i j n ( x i x j ) , and the diagonal elements are obtained as
[ D n ] i , j = j = 0 , j i n [ D n ] i , j , i = 0 , 1 , 2 , . . , n .
Next, the entries of the μ t h -order differentiation matrix D n μ can be obtained analytically as
[ D n μ ] i , j = l j μ ( x i ) , i , j = 1 , 2 , , n .
A more accurate and stable evaluation can be found in [24,38]. In [24], the author derived a useful recursion relation for the differentiation matrices, given as
[ D n μ ] k j = μ x k x j α j α k [ D n ( μ 1 ) ] k k [ D n ( μ 1 ) ] k j , k j
For square domain [ 1 , 1 ] 2 , assume the points x ¯ i j are defined as
x ¯ i j = cos i π n , cos j π n , i , j = 0 , 1 , 2 , , n
The LIP corresponding to the above Chebyshev points are given as
l i j ( x ¯ ) = l i ( x ) l j ( y ) ,
where l i j ( x ¯ i j ) = δ i j , i , j = 0 , 1 , 2 , , n . In terms of x and y, the second derivatives of the LIPs (10) are given as
2 l i j ( x ¯ r s ) x 2 = l i ( x r ) l j ( y s ) = [ D n 2 ] r i δ j s ,
2 l i j ( x ¯ r s ) y 2 = l i ( x r ) l j ( y s ) = δ r i [ D n 2 ] s j ,
where D n 2 denotes the second-order Chebyshev differentiation matrix. Applying the linear operator L on the LIP at the Chebyshev nodes, { x ¯ r s } gives
L ( l i j ( x ¯ r s ) ) = λ 1 [ D n 2 ] r i δ j s + δ r i [ D n 2 ] s j + u 1 [ D n ] r i δ j s + u 2 δ r i [ D n ] s j λ 2 δ r i δ s j ,
Therefore, the discretized form of the linear differential operator L via the CSCM can be expressed as
L D = λ 1 I n D n 2 + D n 2 I n + u 1 I n D n + u 2 D n I n λ 2 I n I n .
By using Equation (12) in Equation (6), we have
( s α I L D ) U ^ ( x ¯ , s ) = G ^ ( x ¯ , s ) ,
where we incorporate the boundary conditions in Equation (7) by taking the matrix L D , based on all collocation points x ¯ , and then replace the rows of L D corresponding to collocation at boundary points with unit vectors that have a one in the position corresponding to the diagonal of L D . Thus, the boundary condition U ( x ¯ , s ) = h ^ ( x ¯ , s ) in (7) will be explicitly enforced at this point as soon as we set the right-hand side to the corresponding value of h ^ ( x ¯ , s ) [23]. After incorporating the boundary conditions and solving Equation (13) for each node s, we obtain the approximate solution U ^ ( x ¯ , s ) in the LT domain. In Section 2.4, we describe the numerical inverse LT method used to evaluate the approximate time domain solution U ( x ¯ , τ ) for the original problem in Equations (1)–(3).

Stability and Error Analysis

Given the Chebyshev nodes in Equation (9) and the Lagrange interpolation polynomials defined in Equation (8), the interpolation operator is defined as follows [37]:
I n : C ( Ω ) P n , I n ( U ) = j = 0 n U ( x j ) l j ( x ) .
The steps proposed by Borm et al. [39] for constructing the error bound are utilized here. Let M n be a constant with the stability estimate given as
I n ( U ) M n U , f o r a l l U C [ 1 , 1 ] .
I n ( U ) = U , f o r a l l U P n .
For the Chebyshev interpolation, we have
M n = l n ( n + 1 ) ( π 2 ) + 1 ( n + 1 ) .
Thus, the stability constant depends on n and develops extremely slowly. Additionally, the following approximation error bound [39] holds for a function U C ( n + 1 ) [ 1 , 1 ] ,
U I n ( U ) 2 n ( n + 1 ) ! U ( n + 1 ) .
Theorem 1
([39]). If (15) and (18) hold for U C ( n + 1 ) [ 1 , 1 ] , and considering k { 0 , 1 , 2 , , n } , then
U ( k ) I n ( U ) ( k ) 2 ( M n ( k ) + 1 ) ( n k + 1 ) ! 1 2 ( n k + 1 ) U ( n + 1 )
depending on the stability constant
M n ( k ) = M n k ! n ! ( n k ) ! .
Now, use (18) and (19) for the TFADE (1) in one dimension. For the 1 D case, the linear differential operator L is of the form L = λ 1 2 x 2 + u 1 x λ 2 I . Thus, we find the following error estimate
E = ( D t α U L U ) ( D t α I n ( U ) L I n ( U ) ) = ( D t α ( U I n ( U ) ) L ( U I n ( U ) ) ( D t α ( U I n ( U ) ) ) + L ( U I n ( U ) ) D t α ( U I n ( U ) ) + | λ 1 | U x x I n ( U ) x x + | u 1 | U x I n ( U ) x + | λ 2 | U I n ( U ) E D t α ( U I n ( U ) ) + | λ 1 | 2 ( M n ( 2 ) + 1 ) ( n 1 ) ! 1 2 n 1 U ( n + 1 ) + | u 1 | 2 ( M n ( 1 ) + 1 ) n ! 1 2 n U ( n + 1 ) + | λ 2 | 2 n ( n + 1 ) ! U ( n + 1 )
The time derivative is accurately evaluated. Therefore, the error bound of ( D t α ( U I n ( U ) ) ) is the same order of ( U I n ( U ) ) . Finally, we have [37]:
E C U ( n + 1 ) .
where C is constant, resulting from calculation of the coefficients of U ( n + 1 ) . We can use the tensor product interpolation operators for two-dimensional problems.

2.4. Numerical Inversion of Laplace Transform

In this section, we implement the inverse Laplace transform method to convert the CSCM solution U ^ ( x ¯ , s ) from the Laplace domain to the time domain as follows:
U ( x ¯ , τ ) = 1 2 π i ρ i ρ + i e s τ U ^ ( x ¯ , s ) d s , ρ > ρ 0 .
Here the transform U ^ ( x ¯ , s ) needs to be inverted, ρ 0 is the converging abscissa, and ρ > ρ 0 . This means that all of the singularities of U ^ ( x ¯ , s ) lie in the open half-plane R e ( s ) < ρ . We aim to approximate the integral defined in Equation (20). In most cases, it is quite difficult to evaluate the integral defined in Equation (20) analytically; thus, a numerical method must be used. There are several numerical algorithms in the literature that can be used to evaluate the integral defined in Equation (20). Among them, we can list the Fourier series method [40], the de Hoog method [41], Stehfest’s method [35,42], Talbot’s method [32,36], and the Weeks method [43,44], etc. Each approach has an identifiable use and is appropriate for a particular function. In this article, we use two popular inversion algorithms: the improved Talbot’s and Stehfest’s methods will be presented in the following sections.

2.4.1. TheImproved Talbot’s Method

Here, we use Talbot’s method to approximate the U ( x ¯ , τ )
U ( x ¯ , τ ) = 1 2 π i ρ i ρ + i e s τ U ^ ( x ¯ , s ) d s = 1 2 π i C e s τ U ^ ( x ¯ , s ) d s , R e ( s ) > 0 .
where C is a suitably chosen line. It is very difficult to solve the above integral due to the highly oscillatory exponential function e s τ and the slow-decaying transform U ^ ( x ¯ , s ) . Talbot [36] suggested that by utilizing contour integration, this problem can be resolved. He further said that the integration might be carried out so that the real parts of the contours begin and stop on the left side of the complex plane, which contains all of the singularities in the transformed function U ^ ( x ¯ , s ) . Due to the exponential element, the integrand decays quickly on such contours, making integral Equation (21) suitable for approximation with the midpoint rule. Consider a contour of the form [32]:
Γ : s = s ( η ) , π η π ,
where R e s ( ± π ) = , and s ( η ) is given as
s ( η ) = M T τ ζ ( η ) , ζ ( η ) = δ + ϱ η c o t ( μ η ) + γ i η ,
where μ ,   ϱ ,   and   γ will be chosen precisely. Using Equation (23) in Equation (21), we obtain
U ( x ¯ , τ ) = 1 2 π i π π e s ( η ) τ U ^ ( x ¯ , s ( η ) ) s ( η ) d η .
The integral defined in (24) is approximated via the midpoint rule with spacing h = 2 π M T as
U A p p ( x ¯ , τ ) 1 M T i j = 1 M T e s ( η j ) τ U ^ ( x ¯ , s ( η j ) ) s ( η j ) , η j = π + ( j 1 2 ) h .

Convergence

The approximate value of the defined integral in Equation (24) depends on the contour of integration Γ ; different rates of convergence are attained for the suggested numerical scheme. Additionally, the quadrature step h determines that the suggested scheme will converge. In order to have optimal results, we need to search for the optimal contour of integration, which can be found by using the optimal values for the parameters involved in (23). In [32], the authors have proposed optimal values for the parameters as
δ = 0.6122 , γ = 0.2645 , μ = 0.6407 , and ϱ = 0.5017 ,
For the improved Talbot’s method, the error estimate is given as
E r r o r e s t = | U A p p ( x ¯ , τ ) U ( x ¯ , τ ) | = O ( e 1.35800 M T ) .

2.4.2. TheStehfest’s Method

The Gaver–Stehfest method is one of the most important techniques for the numerical inversion of the Laplace transform. It was designed in the second half of the 1960s. It has gained prominence in a number of disciplines, such as chemistry, finance, economics, and computational physics, due to its effectiveness and ease. The series of Gaver approximants, as determined by Gaver [42], is the foundation of the Gaver–Stehfest method. Acceleration was required because the Gaver approximants’ convergence was basically logarithmic. Stehfest [35] used the Salzer acceleration method to offer a linear acceleration method. The Gaver–Stehfest approach uses a series of functions to approximate U ( x ¯ , τ ) as
U A p p ( x ¯ , τ ) = l n 2 τ j = 1 M S α j U ^ x ¯ , l n 2 τ j ,
where the coefficients α j are defined by
α j = ( 1 ) M S 2 + j = j + 1 2 m i n ( j , M S 2 ) M S 2 ( 2 ) ! ( M S 2 ) ! ! ( 1 ) ! ( j ) ! ( 2 j ) ! .
Solve Equation (6) for the corresponding Laplace parameters s = l n 2 τ j , j = 1 , 2 , 3 , , M S . The solution to the given problem in Equation (1) can be obtained by (26). There are a few appealing features of the Gaver–Stehfest algorithm: (i) U A p p ( x ¯ , τ ) are linear in terms of the values of U ( x ¯ , s ) ; (ii) the values of U ( x ¯ , s ) are needed only for the real value of s; (iii) the process of computing the coefficients is quite simple; and (iv) for constant functions, this approach yields highly accurate approximations. In the literature, this technique has been used by many authors in [45,46], where it has been demonstrated that this method converges very fast to U A p p ( x ¯ , τ ) (provided U A p p ( x ¯ , τ ) is non-oscillatory).

Convergence

In [45], the author has derived two conditions for U ( x ¯ , τ ) , which guarantee the convergence of U A p p ( x ¯ , τ ) . The conditions are given in the following theorem.
Theorem 2.
Let U : ( 0 , ) R be a locally integrable function and the approximate solution U A p p ( x ¯ , τ ) be defined by (26), and let its LT U ^ ( x ¯ , s ) be defined for s > 0 :
1. 
U A p p ( x ¯ , τ ) converges for the values of U A p p ( x ¯ , τ ) in the neighborhood of τ.
2. 
Let for some real number m and some 0 < ε < 1 / 4
0 ε | U ( τ l o g 2 ( 1 / 2 + ξ ) ) + U ( τ l o g 2 ( 1 / 2 ξ ) ) 2 m | ξ 1 d ξ < .
Then, U m as M S + .
3. 
Allow that U A p p ( x ¯ , τ ) has a bounded variation in the neighborhood of τ . Then,
U A p p ( τ ) U ( τ + 0 ) + U ( τ 0 ) 2 as M S + .
Corollary 1.
Under the above assumptions, if
U ( τ + ξ ) U ( τ ) = O | ξ | υ ,
ξ and some υ in the neighborhood of τ , then U A p p ( x ¯ , τ ) U ( x ¯ , τ ) , as M S + .
Additionally, the authors in [47] performed various experiments for the parameter effect on the accuracy of the numerical scheme, and they concluded in their findings that “If p significant digits are required, then let M S be the positive integer 2.2 p . Set the system precision to q = 1.1 M S and, for a given M S , calculate α i , 1 i M S , using (27). Then, for the given transform U ^ ( x ¯ , s ) and the argument τ , calculate the U A p p ( x ¯ , τ ) in (26)”. According to these conclusions, if the error of input data is 10 ( q + 1 ) U ^ U U 10 q , with an even positive integer M S via q = 1.1 M S , then the final error is 10 ( p + 1 ) U ^ U U 10 p , where M S = 2.2 p [48]. Next, we present Algorithm 1 for the proposed numerical scheme.
Algorithm 1: Algorithm for the proposed numerical scheme
 1:
Input: The computational domain, the fractional order derivative, the final time, the optimal parameters for the inverse laplace transform methods, the inhomogeneous function, and other conditions.
 2:
Step i: Apply the Laplace transform to problems (1) and (2), and obtain the time-independent problems (6) and (7). 
 3:
Step ii: Obtain L D by discretizing the linear differential operator L using the CSCM via (12). 
 4:
Step iii: Solve Equation (13) with the boundary conditions given in Equation (7) for each point in the LT domain. 
 5:
Step iv: Compute the approximate solution to the problem in Equations (1)–(3) using Equation (21) or Equation (26). 
 6:
Step v: The approximate solution is U A p p ( x ¯ , τ ) .  

3. Numerical Results and Discussion

The numerical results and discussions are addressed in this section. Three examples are used to evaluate and validate the effectiveness and accuracy of the Laplace-transformed CSCM. We performed our experiments in MATLAB R2019a on a Windows 10 (64-bit) PC equipped with an Intel(R) Core TM i5-4310M CPU @ 2.70 GHz with 4 GB of RAM. Three error norms—the relative L 2 error, the maximum absolute error, and the RMS error—are used to evaluate the accuracy of the proposed method and are defined below:
e r 2 = j = 1 n ( U ( x ¯ j , τ ) U A p p ( x ¯ j , τ ) ) 2 j = 1 n ( U ( x ¯ j , τ ) ) 2 ,
e r = max 1 j n | U ( x ¯ j , τ ) U A p p ( x ¯ j , τ ) | ,
e r r m s = j = 1 n ( U ( x ¯ j , τ ) U A p p ( x ¯ j , τ ) ) 2 n ,
where U ( x ¯ j , τ ) and U A p p ( x ¯ j , τ ) represent the analytic and numerical solutions, respectively.
Example 1.
Consider the 2D time-fractional advection–diffusion Equation (1) with an exact solution given as
U ( x , y , τ ) = x 2 + y 2 + τ 2 , Ω = [ 1 , 1 ] 2 , τ [ 0 , 1 ] ,
where λ 1 = 1 , u = ( 1 , 1 ) T , and λ 2 = 0 . The forcing term Q ( x , y , τ ) , and the initial-boundary data are extracted from the exact solution. The numerical results obtained via Talbot’s and Stehfest’s methods are presented in Table 1 and Table 2 by using different values of α , M S , M T , and n at τ = 1 . The approximate solution obtained by the proposed method is shown in Figure 1a. The plots for the comparison of errors e r 2 , e r , and e r r m s using Stehfest’s method for several values of α , M S , and τ at n = 25 are shown in Figure 1b–d, respectively. Figure 2a–c show the plots of errors e r 2 , e r , and e r r m s obtained using Talbot’s method with different values of α , M T , and τ at n = 25 .  Figure 2d presents the contour plot of the absolute error using Talbot’s method for different values of α with M T = 26 , τ = 1 , and n = 25 . The accuracy of Stehfest’s method enhances for M S = 8 , 10 , 12 , 14 and then gradually diminishes for values exceeding M S = 14 . From all of the obtained results presented in the figures and tables, we conclude that the proposed method is stable, accurate, and efficient. Clearly, the numerical results show that the accuracy of Talbot’s method is better than that of Stehfest’s method.
Example 2.
Consider the 2D time-fractional advection–diffusion Equation (1) with an exact solution given as
U ( x , y , τ ) = s i n ( x + y ) + τ 2 , Ω = [ 1 , 1 ] 2 , τ [ 0 , 1 ] ,
where λ 1 = 1 , u = ( 1 , 1 ) T , and λ 2 = 0 . The forcing term Q ( x , y , τ ) , and the initial-boundary data are extracted from the exact solution. The numerical results obtained via Talbot’s and Stehfest’s methods are presented in Table 3 and Table 4 by using different values for α , M S , M T , and n at τ = 1 . The approximate solution obtained by the proposed method is shown in Figure 3a. The plots of the comparison of errors e r 2 , e r , and e r r m s using Stehfest’s method for several values of α , M S , and τ at n = 25 are shown in Figure 3b–d, respectively. Figure 4a–c show the plots of errors e r 2 , e r , and e r r m s by using Talbot’s method with different values of α , M T , and τ at n = 25 . Figure 3e presents the contour plot of the absolute error obtained by using Stehfest’s method for different values of α with M S = 12 , τ = 1 , and n = 25 . The accuracy of Stehfest’s method enhances for M S = 8 , 10 , 12 , 14 and then gradually diminishes for values exceeding M S = 14 . From all of the obtained results presented in the figures and tables, we conclude that the proposed method is stable, accurate, and efficient. Clearly, the results show that the accuracy of Talbot’s method is better than that of Stehfest’s method.
Example 3.
Consider the 2D time-fractional advection–diffusion Equation (1) with an exact solution given as
U ( x , y , τ ) = τ 2 e ( x + y ) , Ω = [ 1 , 1 ] 2 , τ [ 0 , 1 ] ,
where λ 1 = 1 , u = ( 1 , 1 ) T , and λ 2 = 0 . The forcing term Q ( x , y , τ ) , and the initial-boundary data are extracted from the exact solution. The numerical results obtained via Talbot’s and Stehfest’s methods are presented in Table 5 and Table 6 by using different values of α , M S , a n d M T , with n at τ = 1 . The approximate solution obtained by the proposed method is shown in Figure 5a. The plots of the comparison of the errors e r 2 , e r , and e r r m s using Stehfest’s method for several values of α , M S , and τ at n = 25 are shown in Figure 5b–d, respectively. Figure 6a–c show the plots of errors e r 2 , e r , and e r r m s via Talbot’s method with different values of α , M T , and τ at n = 25 .  Figure 6d presents the contour plot of the absolute error by using Talbot’s method for different values of α with M T = 26 , τ = 1 , and n = 25 . The accuracy of Stehfest’s method enhances for M S = 8 , 10 , 12 , 14 and then gradually diminishes for values exceeding than M S = 14 . From all the obtained results presented in the figures and tables, we conclude that the proposed method is stable, accurate, and efficient. Clearly, the results show that the accuracy of Talbot’s method is better than that of Stehfest’s method.

4. Conclusions

In this article, a numerical method based on the LT coupled with the CSCM for the numerical solution of two-dimensional TFADEs in the Caputo sense has been developed. It first uses the LT to convert a TFADE into a time-independent inhomogeneous equation in the LT space. Then, it uses the CSCM to discretize the spatial derivatives of this Laplace-transformed inhomogeneous equation. Finally, it chooses Stehfest’s method and Talbot’s method for the numerical inversion of the LT to retrieve the numerical solutions of the TFADE from the corresponding CSCM solutions in the LT domain. Compared with the finite difference time-stepping scheme, the proposed method introduces the LT technique to avoid the calculation of a costly convolution integral in the approximation of time-fractional derivation and to avoid the effect of the time step on the stability and accuracy. Hence, it can successfully solve the long time history TFADEs. The proposed advection–diffusion equation of fractional order presents significant potential for extension into uncertainty propagation and sensitivity analysis, enabling a more comprehensive and nuanced understanding of system responses to parameter variations and initial conditions [49,50,51].

Author Contributions

Conceptualization, F.A.S. and K.; methodology, F.A.S. and K.; software, F.A.S. and K.;validation, W.B., A.K. and N.M.; formal analysis, W.B., A.K. and N.M.; investigation, K., W.B., A.K. and N.M.; resources, W.B., A.K. and N.M.; data curation, F.A.S., K., W.B., A.K. and N.M.; writing—original draft preparation, F.A.S., K.; writing—review and editing, F.A.S. and K.; visualization, W.B., N.M.; supervision, W.B., A.K. and N.M.; project administration, W.B. and N.M.; funding acquisition, W.B. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

All data required for this research are included within the paper.

Acknowledgments

The authors W. Boulila, A. Koubaa and N. Mlaiki would like to thank the Prince Sultan University for paying the publication fees for this work through RIOTU LAB.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Khan, Z.; Shah, K.; Abdalla, B.; Abdeljawad, T. A numerical Study of Complex Dynamics of a Chemostat Model Under Fractal-Fractional Derivative. Fractals 2023, 31, 2340181. [Google Scholar] [CrossRef]
  2. Shah, K.; Amin, R.; Abdeljawad, T. Utilization of Haar wavelet collocation technique for fractal-fractional order problem. Heliyon. 2023, 9, e17123. [Google Scholar] [CrossRef] [PubMed]
  3. Podlubny, I. Fractional Differential Equations: An Introduction to Fractional Derivatives, Fractional Differential Equations, to Methods of Their Solution and Some of Their Applications; Elsevier: Amsterdam, The Netherlands, 1998. [Google Scholar]
  4. Oldham, K.; Spanier, J. The Fractional Calculus Theory and Applications of Differentiation and Integration to Arbitrary Order; Elsevier: Amsterdam, The Netherlands, 1974. [Google Scholar]
  5. Tarasov, V.E. Geometric interpretation of fractional-order derivative. Fract. Calc. Appl. Anal. 2016, 19, 1200–1221. [Google Scholar] [CrossRef]
  6. Roohi, M.; Zhang, C.; Taheri, M.; Basse-ÓConnor, A. Synchronization of Fractional-Order Delayed Neural Networks Using Dynamic-Free Adaptive Sliding Mode Control. Fractal Fract. 2023, 7, 682. [Google Scholar] [CrossRef]
  7. Zaid, S.A.; Bakeer, A.; Albalawi, H.; Alatwi, A.M.; AbdelMeguid, H.; Kassem, A.M. Optimal Fractional-Order Controller for the Voltage Stability of a DC Microgrid Feeding an Electric Vehicle Charging Station. Fractal Fract. 2023, 7, 677. [Google Scholar] [CrossRef]
  8. Gharab, S.; Feliu Batlle, V. Fractional Control of a Class of Underdamped Fractional Systems with Time Delay-Application to a Teleoperated Robot with a Flexible Link. Fractal Fract. 2023, 7, 646. [Google Scholar] [CrossRef]
  9. Guechi, S.; Dhayal, R.; Debbouche, A.; Malik, M. Analysis and Optimal Control of ϕ-Hilfer Fractional Semilinear Equations Involving Nonlocal Impulsive Conditions. Symmetry 2021, 13, 2084. [Google Scholar] [CrossRef]
  10. Benson, D.A.; Wheatcraft, S.W.; Meerschaert, M.M. Application of a fractional advection-dispersion equation. Water Resour. Res. 2000, 36, 1403–1412. [Google Scholar] [CrossRef]
  11. Magin, R.L. Fractional Calculus in Bioengineering; Begell House Publishers: Danbury, CT, USA, 2006. [Google Scholar]
  12. Sun, H.; Zhang, Y.; Baleanu, D.; Chen, W.; Chen, Y. A new collection of real world applications of fractional calculus in science and engineering. Commun. Nonlinear Sci. Numer. Simul. 2018, 64, 213–231. [Google Scholar] [CrossRef]
  13. Owolabi, K.M. High-dimensional spatial patterns in fractional reaction diffusion system arising in biology. Chaos Solitons Fractals 2020, 134, 109723. [Google Scholar] [CrossRef]
  14. Isenberg, J.; Gutfinger, C. Heat transfer to a draining film. Int. J. Heat Mass Transf. 1973, 16, 505–512. [Google Scholar] [CrossRef]
  15. Kumar, N. Unsteady flow against dispersion in finite porous media. J. Hydrol. 1983, 63, 345–358. [Google Scholar] [CrossRef]
  16. Lanser, D.; Verwer, J.G. Analysis of operator splitting for advection–diffusion–reaction problems from air pollution modelling. J. Comput. Appl. Math. 1999, 111, 201–216. [Google Scholar] [CrossRef]
  17. Zhao, Y.L.; Huang, T.Z.; Gu, X.M.; Luo, W.H. A fast second-order implicit difference method for time space fractional advection diffusion equation. Numer. Funct. Anal. Optim. 2020, 41, 257–293. [Google Scholar] [CrossRef]
  18. Mardani, A.; Hooshmandasl, M.R.; Heydari, M.H.; Cattani, C. A meshless method for solving the time fractional advection diffusion equation with variable coefficients. Comput. Math. Appl. 2018, 75, 122–133. [Google Scholar] [CrossRef]
  19. Mohyud-Din, S.T.; Akram, T.; Abbas, M.; Ismail, A.I.; Ali, N.H. A fully implicit finite difference scheme based on extended cubic B-splines for time fractional advection-diffusion equation. Adv. Differ. Equations 2018, 2018, 1–17. [Google Scholar] [CrossRef]
  20. Abbaszadeh, M.; Amjadian, H. Second order finite difference or spectral element formulation for solving the fractional advection diffusion equation. Commun. Appl. Math. Comput. 2020, 2, 653–669. [Google Scholar] [CrossRef]
  21. Gottlieb, D.; Orszag, S.A. Numerical Analysis of Spectral Methods: Theory and Applications; Society for Industrial and Applied Mathematics: Philadelphia, PA, USA, 1977. [Google Scholar]
  22. Canuto, C.; Hussaini, M.Y.; Quarteroni, A.; Zang, T.A. Spectral Methods, Evolution to Complex Geometries and Applications to Fluid Dynamics; Springer Science & Business Media: Berlin/Heidelberg, Germany, 2007. [Google Scholar]
  23. Trefethen, L.N. Spectral Methods in MATLAB, Volume 10 of Software, Environments, and Tools; Society for Industrial and Applied Mathematics (SIAM): Philadelphia, PA, USA, 2000; Volume 24, p. 57. [Google Scholar]
  24. Welfert, B.D. Generation of pseudospectral differentiation matrices I. Siam J. Numer. Anal. 1997, 34, 1640–1657. [Google Scholar] [CrossRef]
  25. Mao, Z.; Shen, J. Hermite spectral methods for fractional PDEs in unbounded domains. Siam J. Sci. Comput. 2017, 39, A1928–A1950. [Google Scholar] [CrossRef]
  26. Sweilam, N.H.; Khader, M.M.; Adel, M. Chebyshev pseudo-spectral method for solving fractional advection-dispersion equation. Appl. Math. 2014, 5, 3240. [Google Scholar] [CrossRef]
  27. Bhrawy, A.H.; Baleanu, D. A spectral Legendre-Gauss-Lobatto collocation method for a space-fractional advection diffusion equations with variable coefficients. Rep. Math. Phys. 2013, 72, 219–233. [Google Scholar] [CrossRef]
  28. Tian, W.; Deng, W.; Wu, Y. Polynomial spectral collocation method for space fractional advection-diffusion equation. Numer. Methods Partial. Differ. Equ. 2014, 30, 514–535. [Google Scholar] [CrossRef]
  29. Sweilam, N.H.; El Sayed, A.A.E.; Boulaaras, S. Fractional order advection dispersion problem solution via the spectral collocation method and the non standard finite difference technique. Chaos Solitons Fractals 2021, 144, 110736. [Google Scholar] [CrossRef]
  30. Kamran; Khan, S.; Alhazmi, S.E.; Alotaibi, F.M.; Ferrara, M.; Ahmadian, A. On the Numerical Approximation of Mobile-Immobile Advection-Dispersion Model of Fractional Order Arising from Solute Transport in Porous Media. Fractal Fract. 2022, 6, 445. [Google Scholar] [CrossRef]
  31. Davies, A.J.; Crann, D.; Kane, S.J.; Lai, C.H. A hybrid Laplace transform/finite difference boundary element method for diffusion problems. Comput. Model. Eng. Sci. 2007, 18, 79–86. [Google Scholar]
  32. Dingfelder, B.; Weideman, J.A.C. An improved Talbot method for numerical Laplace transform inversion. Numer. Algorithms 2015, 68, 167–183. [Google Scholar] [CrossRef]
  33. Kamran; Kamal, R.; Rahmat, G.; Shah, K. On the Numerical Approximation of Three-Dimensional Time Fractional Convection-Diffusion Equations. Math. Probl. Eng. 2021, 2021, 1–16. [Google Scholar] [CrossRef]
  34. Kamal, R.; Kamran; Rahmat, G.; Ahmadian, A.; Arshad, N.I.; Salahshour, S. Approximation of linear one dimensional partial differential equations including fractional derivative with non-singular kernel. Adv. Differ. Equ. 2021, 2021, 1–15. [Google Scholar] [CrossRef]
  35. Stehfest, H. Algorithm 368: Numerical inversion of Laplace transforms [D5]. Commun. Acm 1970, 13, 47–49. [Google Scholar] [CrossRef]
  36. Talbot, A. The accurate numerical inversion of Laplace transforms. Ima J. Appl. Math. 1979, 23, 97–120. [Google Scholar] [CrossRef]
  37. Shokri, A.; Mirzaei, S. A pseudo-spectral based method for time-fractional advection-diffusion equation. Comput. Methods Differ. Equ. 2020, 8, 454–467. [Google Scholar]
  38. Baltensperger, R.; Trummer, M.R. Spectral differencing with a twist. Siam J. Sci. Comput. 2003, 24, 1465–1487. [Google Scholar] [CrossRef]
  39. Börm, S.; Grasedyck, L.; Hackbusch, W. Introduction to hierarchical matrices with applications. Eng. Anal. Bound. Elem. 2003, 27, 405–422. [Google Scholar] [CrossRef]
  40. Crump, K.S. Numerical inversion of Laplace transforms using a Fourier series approximation. J. ACM 1976, 23, 89–96. [Google Scholar] [CrossRef]
  41. De Hoog, F.R.; Knight, J.H.; Stokes, A.N. An improved method for numerical inversion of Laplace transforms. Siam J. Sci. Stat. Comput. 1982, 3, 357–366. [Google Scholar] [CrossRef]
  42. Gaver, D.P., Jr. Observing stochastic processes, and approximate transform inversion. Oper. Res. 1966, 14, 444–459. [Google Scholar] [CrossRef]
  43. Weeks, W.T. Numerical inversion of Laplace transforms using Laguerre functions. J. ACM 1966, 13, 419–429. [Google Scholar] [CrossRef]
  44. Kamran; Khan, S.U.; Haque, S.; Mlaiki, N. On the Approximation of Fractional-Order Differential Equations Using Laplace Transform and Weeks Method. Symmetry 2023, 15, 1214. [Google Scholar] [CrossRef]
  45. Kuznetsov, A. On the Convergence of the Gaver–Stehfest Algorithm. SIAM J. Numer. Anal. 2013, 51, 2984–2998. [Google Scholar] [CrossRef]
  46. Davies, B.; Martin, B. Numerical inversion of the Laplace transform: A survey and comparison of methods. J. Comput. Phys. 1979, 33, 1–32. [Google Scholar] [CrossRef]
  47. Abate, J.; Whitt, W. A unified framework for numerically inverting Laplace transforms. Informs J. Comput. 2006, 18, 408–421. [Google Scholar] [CrossRef]
  48. Fu, Z.J.; Chen, W.; Yang, H.T. Boundary particle method for Laplace transformed time fractional diffusion equations. J. Comput. Phys. 2013, 235, 52–66. [Google Scholar] [CrossRef]
  49. Boulila, W.; Ayadi, Z.; Farah, I. Sensitivity analysis approach to model epistemic and aleatory imperfection: Application to Land Cover Change prediction model. J. Comput. Sci. 2017, 23, 58–70. [Google Scholar] [CrossRef]
  50. Ferchichi, A.; Boulila, W.; Farah, I. Propagating aleatory and epistemic uncertainty in land cover change prediction process. Ecol. Inform. 2017, 37, 24–37. [Google Scholar] [CrossRef]
  51. Ferchichi, A.; Boulila, W.; Farah, I. Reducing uncertainties in land cover change models using sensitivity analysis. Knowl. Inf. Syst. 2018, 55, 719–740. [Google Scholar] [CrossRef]
Figure 1. (a) Numerical solution of Example 1. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different α , with M S = 10 and n = 25 at τ = 1 for Example 1. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different M S with α = 0.3 and n = 25 at τ = 1 for Example 1. (d) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different τ with M S = 10 , α = 0.3 , and n = 25 for Example 1.
Figure 1. (a) Numerical solution of Example 1. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different α , with M S = 10 and n = 25 at τ = 1 for Example 1. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different M S with α = 0.3 and n = 25 at τ = 1 for Example 1. (d) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different τ with M S = 10 , α = 0.3 , and n = 25 for Example 1.
Fractalfract 07 00762 g001
Figure 2. (a) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different α with M T = 22 and n = 25 at τ = 1 for Example 1. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different M T with α = 0.3 and n = 25 at τ = 1 for Example 1. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different τ with M T = 22 , a l p h a = 0.3 , and n = 25 for Example 1. (d) Contour plot of the absolute error obtained with M T = 26 , τ = 1 a l p h a = 0.3 , and n = 25 using Talbot’s method for Example 1.
Figure 2. (a) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different α with M T = 22 and n = 25 at τ = 1 for Example 1. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different M T with α = 0.3 and n = 25 at τ = 1 for Example 1. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different τ with M T = 22 , a l p h a = 0.3 , and n = 25 for Example 1. (d) Contour plot of the absolute error obtained with M T = 26 , τ = 1 a l p h a = 0.3 , and n = 25 using Talbot’s method for Example 1.
Fractalfract 07 00762 g002
Figure 3. (a) Numerical solution of Example 2. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different α , with M S = 10 and n = 25 at τ = 1 for Example 2. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different M S with α = 0.3 and n = 25 at τ = 1 for Example 2. (d) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different τ with M S = 10 , α = 0.3 , and n = 25 for Example 2. (e) Contour plot of the absolute error obtained using Stehfest’s method with α = 0.5 , M S = 12 , τ = 1 , and n = 25 for Example 2.
Figure 3. (a) Numerical solution of Example 2. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different α , with M S = 10 and n = 25 at τ = 1 for Example 2. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different M S with α = 0.3 and n = 25 at τ = 1 for Example 2. (d) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different τ with M S = 10 , α = 0.3 , and n = 25 for Example 2. (e) Contour plot of the absolute error obtained using Stehfest’s method with α = 0.5 , M S = 12 , τ = 1 , and n = 25 for Example 2.
Fractalfract 07 00762 g003aFractalfract 07 00762 g003b
Figure 4. (a) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different α , with M T = 22 and n = 25 at τ = 1 for Example 2. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different M T with α = 0.3 and n = 25 at τ = 1 for Example 2. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different τ with M T = 22 , a l p h a = 0.3 , and n = 25 for Example 2.
Figure 4. (a) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different α , with M T = 22 and n = 25 at τ = 1 for Example 2. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different M T with α = 0.3 and n = 25 at τ = 1 for Example 2. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different τ with M T = 22 , a l p h a = 0.3 , and n = 25 for Example 2.
Fractalfract 07 00762 g004aFractalfract 07 00762 g004b
Figure 5. (a) Numerical solution of Example 3. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different α , with M S = 10 and n = 25 at τ = 1 for Example 3. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different M S with α = 0.3 and n = 25 at τ = 1 for Example 3. (d) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different τ with M S = 10 , α = 0.3 , and n = 25 for Example 3.
Figure 5. (a) Numerical solution of Example 3. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different α , with M S = 10 and n = 25 at τ = 1 for Example 3. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different M S with α = 0.3 and n = 25 at τ = 1 for Example 3. (d) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Stehfest’s method for different τ with M S = 10 , α = 0.3 , and n = 25 for Example 3.
Fractalfract 07 00762 g005
Figure 6. (a) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different α , with M T = 26 and n = 25 at τ = 1 for Example 3. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different M T with α = 0.3 and n = 25 at τ = 1 for Example 3. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different τ with M T = 26 , a l p h a = 0.3 , and n = 25 for Example 3. (d) Contour plot of the absolute error obtained with M T = 26 , τ = 1 a l p h a = 0.5 , and n = 25 using Talbot’s method for Example 3.
Figure 6. (a) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different α , with M T = 26 and n = 25 at τ = 1 for Example 3. (b) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different M T with α = 0.3 and n = 25 at τ = 1 for Example 3. (c) The plot shows a comparison of e r 2 ,   e r , and e r r m s using Talbot’s method for different τ with M T = 26 , a l p h a = 0.3 , and n = 25 for Example 3. (d) Contour plot of the absolute error obtained with M T = 26 , τ = 1 a l p h a = 0.5 , and n = 25 using Talbot’s method for Example 3.
Fractalfract 07 00762 g006
Table 1. The e r 2 , e r , and e r r m s obtained using Stehfest’s method for different values of α , n , and M S at τ = 1 for Example 1.
Table 1. The e r 2 , e r , and e r r m s obtained using Stehfest’s method for different values of α , n , and M S at τ = 1 for Example 1.
α = 0.30 n M S er 2 er er rms CPU(s)
21129.7994 × 10 4 6.0556 × 10 5 4.6664 × 10 5 0.233261
23 1.1084 × 10 3 8.2038 × 10 5 4.8190 × 10 5 0.200611
25 1.1581 × 10 3 7.1935 × 10 5 4.6325 × 10 5 0.266045
25081.0812 × 10 1 4.1584 × 10 3 4.3246 × 10 3 0.196162
101.4553 × 10 3 5.8142 × 10 5 5.8213 × 10 5 0.249590
121.1581 × 10 3 7.1935 × 10 5 4.6325 × 10 5 0.255773
α = 0.50
21129.8850 × 10 4 5.6517 × 10 5 4.7071 × 10 5 0.167258
23 1.0923 × 10 3 7.1906 × 10 5 4.7492 × 10 5 0.204245
25 1.2186 × 10 3 1.3279 × 10 4 4.8744 × 10 5 0.302795
25081.0812 × 10 1 4.1584 × 10 4 4.3246 × 10 3 0.196037
101.4539 × 10 3 5.7437 × 10 5 5.8156 × 10 5 0.241286
121.2186 × 10 3 1.3279 × 10 4 4.8744 × 10 5 0.377014
α = 0.80
21121.0540 × 10 3 9.2987 × 10 5 5.0191 × 10 5 0.142862
23 9.8247 × 10 4 6.2113 × 10 5 4.2716 × 10 5 0.235252
25 1.2741 × 10 3 1.5227 × 10 4 5.0962 × 10 5 0.287223
25081.0811 × 10 1 4.1585 × 10 3 4.3246 × 10 5 0.214762
101.4514 × 10 3 5.7719 × 10 5 5.8056 × 10 5 0.234821
121.2741 × 10 3 1.5227 × 10 4 5.0962 × 10 5 0.287223
Table 2. The e r 2 , e r , and e r r m s obtained via Talbot’s method for different values of α , n , and M T at τ = 1 for Example 1.
Table 2. The e r 2 , e r , and e r r m s obtained via Talbot’s method for different values of α , n , and M T at τ = 1 for Example 1.
α = 0.30 n M T er 2 er er rms CPU(s)
21263.6788 × 10 10 9.4858 × 10 11 1.7518 × 10 11 0.477112
23 5.1649 × 10 10 1.5900 × 10 10 2.2456 × 10 11 0.710826
25 8.1265 × 10 10 3.4762 × 10 10 3.2506 × 10 11 1.118694
25223.6227 × 10 8 1.5662 × 10 9 1.4491 × 10 9 0.955487
243.0969 × 10 9 3.1978 × 10 10 1.2388 × 10 10 0.997716
268.1265 × 10 10 3.4762 × 10 10 3.2506 × 10 11 1.082287
α = 0.50
21264.5560 × 10 10 1.7228 × 10 10 2.1695 × 10 11 0.468985
23 5.0871 × 10 10 1.8159 × 10 10 2.2118 × 10 11 0.741330
25 8.4173 × 10 10 3.3289 × 10 10 3.3669 × 10 11 1.042339
25223.6042 × 10 8 1.4790 × 10 9 1.4417 × 10 9 0.952621
242.8473 × 10 9 3.0892 × 10 10 1.1389 × 10 10 0.987540
268.4173 × 10 10 3.3289 × 10 10 3.3669 × 10 11 1.042339
α = 0.80
21263.9392 × 10 10 8.8192 × 10 11 1.8758 × 10 11 0.479675
23 4.8220 × 10 10 1.9319 × 10 10 2.0965 × 10 11 0.761089
25 1.1098 × 10 9 7.7091 × 10 10 4.4393 × 10 11 1.094016
25223.6116 × 10 8 1.5182 × 10 9 1.4446 × 10 9 0.917780
242.9420 × 10 9 2.3727 × 10 10 1.1768 × 10 10 1.010576
261.1098 × 10 9 7.7091 × 10 10 4.4393 × 10 11 1.094016
Table 3. The e r 2 , e r , and e r r m s obtained via Stehfest’s method for different values of α , n , and M S at τ = 1 of Example 2.
Table 3. The e r 2 , e r , and e r r m s obtained via Stehfest’s method for different values of α , n , and M S at τ = 1 of Example 2.
α = 0.30 n M S er 2 er er rms CPU(s)
21129.9430 × 10 4 5.2055 × 10 5 4.7348 × 10 5 0.192349
23 1.0940 × 10 3 6.1845 × 10 5 4.7567 × 10 5 0.333385
25 1.1758 × 10 3 6.1712 × 10 5 4.7031 × 10 5 0.303723
25081.0812 × 10 1 4.1584 × 10 3 4.3246 × 10 3 0.271822
101.4554 × 10 3 5.8032 × 10 5 5.8217 × 10 5 0.257647
121.1758 × 10 3 6.1712 × 10 5 4.7031 × 10 5 0.303723
α = 0.50
21129.9254 × 10 4 5.2330 × 10 5 4.7264 × 10 5 0.170534
23 1.0888 × 10 3 5.9124 × 10 5 4.7338 × 10 5 0.257113
25 1.1700 × 10 3 5.4659 × 10 5 4.6800 × 10 5 0.360627
25081.0812 × 10 1 4.1583 × 10 3 4.3246 × 10 3 0.197690
101.4551 × 10 3 5.6736 × 10 5 5.8205 × 10 5 0.383466
121.1700 × 10 3 5.4659 × 10 5 4.6800 × 10 5 0.360627
α = 0.80
21129.9170 × 10 4 5.0410 × 10 5 4.7224 × 10 5 0.178846
23 1.0836 × 10 3 5.0672 × 10 5 4.7115 × 10 5 0.276298
25 1.2056 × 10 3 6.7927 × 10 5 4.8224 × 10 5 0.292158
25081.0812 × 10 1 4.1583 × 10 3 4.3246 × 10 5 0.274343
101.4550 × 10 3 5.6956 × 10 5 5.8199 × 10 5 0.269435
121.2056 × 10 3 6.7927 × 10 5 4.8224 × 10 5 0.292158
Table 4. The e r 2 , e r , and e r r m s obtained using Talbot’s method for different values of α , n , and M T at τ = 1 of Example 2.
Table 4. The e r 2 , e r , and e r r m s obtained using Talbot’s method for different values of α , n , and M T at τ = 1 of Example 2.
α = 0.30 n M T er 2 er er rms CPU(s)
21262.3790 × 10 10 6.3914 × 10 11 1.1328 × 10 11 0.508399
23 2.6664 × 10 10 5.2105 × 10 11 1.1593 × 10 11 0.829189
25 5.6968 × 10 10 3.8928 × 10 10 2.2787 × 10 11 1.151650
25223.6179 × 10 8 1.4491 × 10 9 1.4472 × 10 9 0.977947
242.9202 × 10 9 2.3436 × 10 10 1.1681 × 10 10 1.044065
265.6968 × 10 10 3.8928 × 10 10 2.2787 × 10 11 1.151650
α = 0.50
21262.7669 × 10 10 7.5912 × 10 11 1.3176 × 10 11 0.545475
23 2.7187 × 10 10 6.4024 × 10 11 1.1820 × 10 11 0.806244
25 4.6716 × 10 10 2.0896 × 10 10 1.8687 × 10 11 1.225075
25223.6229 × 10 8 1.4880 × 10 9 1.4492 × 10 9 0.939474
242.8311 × 10 9 1.5034 × 10 10 1.1324 × 10 10 1.027609
264.6716 × 10 10 2.0896 × 10 10 1.8687 × 10 11 1.225075
α = 0.80
21262.4234 × 10 10 4.0678 × 10 11 1.1540 × 10 11 0.512461
23 2.4004 × 10 10 7.5801 × 10 11 1.0437 × 10 11 0.820029
25 4.3673 × 10 10 2.3799 × 10 10 1.7469 × 10 11 1.162087
25223.6188 × 10 8 1.4465 × 10 9 1.4475 × 10 9 0.988509
242.8035 × 10 9 1.4878 × 10 10 1.1214 × 10 10 1.047553
264.3673 × 10 9 2.3799 × 10 10 1.7469 × 10 11 1.162087
Table 5. The e r 2 , e r , and e r r m s obtained via Stehfest’s method for different values of α , n , and M S at τ = 1 of Example 3.
Table 5. The e r 2 , e r , and e r r m s obtained via Stehfest’s method for different values of α , n , and M S at τ = 1 of Example 3.
α = 0.30 n M S er 2 er er rms CPU(s)
21141.9041 × 10 4 3.5446 × 10 5 9.0670 × 10 6 0.142019
23 1.7304 × 10 4 2.7673 × 10 5 7.5234 × 10 6 0.214226
25 2.2300 × 10 4 4.3858 × 10 5 8.9200 × 10 6 0.278411
25103.3997 × 10 3 4.1349 × 10 4 1.3599 × 10 4 0.234626
122.7481 × 10 3 3.3443 × 10 4 1.0992 × 10 4 0.275799
142.2300 × 10 4 4.3858 × 10 5 8.9200 × 10 6 0.334381
α = 0.50
21141.7886 × 10 4 3.4616 × 10 5 8.5172 × 10 6 0.173237
23 1.9875 × 10 4 4.7601 × 10 5 8.6413 × 10 6 0.238699
25 1.8379 × 10 4 2.8417 × 10 5 7.3515 × 10 6 0.321624
25103.3995 × 10 3 4.1349 × 10 4 1.3598 × 10 4 0.246361
122.7522 × 10 3 3.3443 × 10 4 1.1009 × 10 4 0.256780
141.8379 × 10 4 2.8417 × 10 5 7.3515 × 10 6 0.320022
α = 0.80
21142.1292 × 10 4 2.4907 × 10 5 1.0139 × 10 5 0.160703
23 2.8307 × 10 4 4.5577 × 10 5 1.2307 × 10 5 0.228405
25 4.0222 × 10 4 4.9547 × 10 5 1.6089 × 10 5 0.323644
25103.3995 × 10 3 4.1349 × 10 4 1.3598 × 10 4 0.253768
122.7525 × 10 3 3.3443 × 10 4 1.1010 × 10 5 0.318281
144.0222 × 10 4 4.9547 × 10 5 1.6089 × 10 5 0.323644
Table 6. The e r 2 , e r , and e r r m s obtained using Talbot’s method for different values of α , n , and M T at τ = 1 for Example 3.
Table 6. The e r 2 , e r , and e r r m s obtained using Talbot’s method for different values of α , n , and M T at τ = 1 for Example 3.
α = 0.30 n M S er 2 er er rms CPU(s)
21302.4416 × 10 11 8.9271 × 10 12 1.1627 × 10 12 0.623102
23 4.5926 × 10 11 2.4259 × 10 11 1.9968 × 10 12 0.893166
25 1.0575 × 10 10 3.6040 × 10 11 4.2300 × 10 12 1.424708
25265.3511 × 10 10 8.2624 × 10 11 2.1404 × 10 11 1.154428
287.2756 × 10 11 3.0986 × 10 11 2.9102 × 10 12 1.290241
301.0575 × 10 10 3.6040 × 10 11 4.2300r × 10 12 1.304660
α = 0.50
21302.4307 × 10 11 8.8405 × 10 12 1.1575 × 10 12 0.590850
23 4.5760 × 10 11 1.3785 × 10 11 1.9896 × 10 12 0.895766
25 6.7767 × 10 11 3.3908 × 10 11 2.7107 × 10 12 1.342481
25265.0284 × 10 10 7.5070 × 10 11 2.0114 × 10 11 1.118816
288.2316 × 10 11 3.8009 × 10 11 3.2926 × 10 12 1.221575
306.7767 × 10 11 3.3908 × 10 11 2.7107 × 10 12 1.282603
α = 0.80
21302.6976 × 10 11 8.4350 × 10 12 1.2846 × 10 12 0.569593
23 4.9832 × 10 11 1.3129 × 10 11 2.1666 × 10 12 0.901900
25 8.9002 × 10 11 5.0524 × 10 11 3.5601 × 10 12 1.338635
25265.0864 × 10 10 7.6800 × 10 11 2.0346 × 10 11 1.152948
286.4788 × 10 11 2.4306 × 10 11 2.5915 × 10 12 1.197204
308.9002 × 10 11 5.0524 × 10 11 3.5601 × 10 12 1.295479
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

Ali Shah, F.; Kamran; Boulila, W.; Koubaa, A.; Mlaiki, N. Numerical Solution of Advection–Diffusion Equation of Fractional Order Using Chebyshev Collocation Method. Fractal Fract. 2023, 7, 762. https://doi.org/10.3390/fractalfract7100762

AMA Style

Ali Shah F, Kamran, Boulila W, Koubaa A, Mlaiki N. Numerical Solution of Advection–Diffusion Equation of Fractional Order Using Chebyshev Collocation Method. Fractal and Fractional. 2023; 7(10):762. https://doi.org/10.3390/fractalfract7100762

Chicago/Turabian Style

Ali Shah, Farman, Kamran, Wadii Boulila, Anis Koubaa, and Nabil Mlaiki. 2023. "Numerical Solution of Advection–Diffusion Equation of Fractional Order Using Chebyshev Collocation Method" Fractal and Fractional 7, no. 10: 762. https://doi.org/10.3390/fractalfract7100762

Article Metrics

Back to TopTop