Next Article in Journal
Fejér-Type Fractional Integral Inequalities Involving Mittag-Leffler Function
Previous Article in Journal
Quantitative Analysis of Crack Propagation Behavior in Recycled Concrete Subjected to Axial Compression Using Digital Image Correlation (DIC) Technology and Fractal Theory
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Numerical Simulation Based on Interpolation Technique for Multi-Term Time-Fractional Convection–Diffusion Equations

1
College of Big Data Statistics, Guizhou University of Finance and Economics, Guiyang 550025, China
2
College of General Studies, Xinjiang University of Technology, Hotan 848000, China
3
School of Mathematics and Statistics, Henan University of Technology, Zhengzhou 450001, China
4
Department of Mathematics, National Institute of Technology, Jamshedpur 831014, Jharkhand, India
*
Author to whom correspondence should be addressed.
Fractal Fract. 2024, 8(12), 687; https://doi.org/10.3390/fractalfract8120687
Submission received: 15 October 2024 / Revised: 9 November 2024 / Accepted: 20 November 2024 / Published: 23 November 2024
(This article belongs to the Section Numerical and Computational Methods)

Abstract

This paper introduces a novel approach for solving multi-term time-fractional convection–diffusion equations with the fractional derivatives in the Caputo sense. The proposed highly accurate numerical algorithm is based on the barycentric rational interpolation collocation method (BRICM) in conjunction with the Gauss–Legendre quadrature rule. The discrete scheme constructed in this paper can achieve high computational accuracy with very few interval partitioning points. To verify the effectiveness of the present discrete scheme, some numerical examples are presented and are compared with the other existing method. Numerical results demonstrate the effectiveness of the method and the correctness of the theoretical analysis.

1. Introduction

In this study, we investigate the following multi-term time-fractional convection–diffusion equations with the fractional derivatives in the Caputo sense,
D t β 0 C + j = 1 J λ j D t β j 0 C u ( x , t ) = P ( x , t ) u x x ( x , t ) Q ( x , t ) u x ( x , t ) + f ( x , t ) , ( x , t ) Ω , u ( x , 0 ) = ψ ( x ) , x ( 0 , L ) , u ( 0 , t ) = u ( L , t ) = 0 , t ( 0 , T ] ,
where 0 < β s β 2 β 1 < β < 1 are the fractional orders, λ j > 0 ( 1 j J ) , and P ( x , t ) and Q ( x , t ) are the diffusion coefficient and the convection coefficient, respectively. Ω = [ 0 , L ] × [ 0 , T ] , f ( x , t ) is the forcing function, ψ ( x ) is given sufficiently smooth function, and u ( x , t ) is the unknown function.
If Q ( x , t ) = 0 , then Equation (1) will become the multi-term time-fractional diffusion equation (TFDE). And if J = 0 , then Equation (1) will become the time-fractional convection–diffusion equation, which has been studied by many scholars using various numerical methods, including the finite difference method [1,2,3,4], the finite element method [5,6,7], the finite volume method [8,9,10] and the spectral method [11,12], etc.
Multi-term fractional order differential equations provide a higher degree of flexibility in modelling complex real-world phenomena. They can be used to capture a wider range of memory effects by combining multiple fractional orders, which makes them more effective in modelling complex systems. Therefore, it is necessary to explore the numerical method of multi-term fractional order differential equations. A number of studies on multi-term fractional order differential equations have been conducted recently, in particular, studies such as those of the Hermite wavelets approach [13], the Pseudospectral method [14], Chebyshev polynomials [15,16], the generalized squared remainder minimization method [17], the Haar wavelet collocation method [18], and so on.
The main purpose of this paper is to solve a class of multi-term fractional convection–diffusion equations using the BRICM, where the fractional order derivatives are in turn given by the Caputo definition:
D t β 0 C u ( x , t ) = 1 Γ ( 1 β ) 0 t ( t s ) β u ( x , s ) s d s ,
which is one of the common derivatives of fractional order and has been applied in many areas. Properties and more details about Caputo’s fractional derivative can be found in [19,20,21]. The BRICM is a high-order interpolation algorithm, which can effectively avoid the Runge’s phenomenon and has good robustness to irregular data. In recent years, the BRICM has been applied in solving various differential equations. Additional studies can be found in [22,23,24,25,26,27], among others.
The rest of this paper is organized as follows: In Section 2, the discrete scheme is constructed by using the combination of the BRICM and the Gauss–Legendre quadrature rule, the theoretical analysis is given in Section 3, and the numerical results in Section 4 support the theoretical analysis. Finally, we conclude our results in Section 5.

2. Highly Accurate Numerical Algorithm for Equation (1)

2.1. Background Knowledge of the BRICM

For classical rational interpolation, the existence of poles has a significant negative impact on it. Therefore, Berrut and Mittelmann [28] proposed an interpolation technique to avoid poles and improved the result by using higher-order rational functions. The interpolation can be written in the following barycentric rational form:
u ( x ) i = 0 I w i x x i u i / p = 0 I w p x x p ,
where x i ( i [ I ] , [ I ] = { 0 , 1 , , I } ) are I + 1 different interpolation nodes, the value of u ( x ) at point x i is denoted by u i = u ( x i ) . In [29], Berrut used
w i = ( 1 ) i , i [ I ]
to denote the interpolation weights of barycentric rational interpolation. Let d x ( 0 d x I ) be an arbitrary integer; in [30], Floater and Hormann used the interpolation weights as
w i = z [ O i ] ( 1 ) z θ = z , θ i z + d x 1 x i x θ , [ O i ] = { z [ I ] : i d x z i } ,
and if d x = 0 , we can get that w i = ( 1 ) i ( i [ I ] ) which is the same as that in [29]. Thus, in the following, we just focus on the case of d x 1 .
Let ϕ i ( x ) = w i x x i / p = 0 I w p x x p , then ϕ i ( x ) 1 as x x i , and ϕ i ( x ) = 0 as x x i . By Equation (3), we can obtain the barycentric rational interpolation function (BRIF) of u ( x ) , denoted by u B 1 ( x ) as
u ( x ) u B 1 ( x ) = i = 0 I ϕ i ( x ) u i .
Similarly, by Equation (6), the transcription in the time domain implies that
u ( t ) u B 1 ( t ) = k = 0 K ϑ k ( t ) u k ,
where u k = u ( t k ) and ϑ k ( t ) = ω k t t k / ρ = 0 K ω ρ t t ρ .
Let u i ( t ) = u ( x i , t ) and u i k = u ( x i , t k ) . By Equation (7), the BRIF of u i ( t ) is denoted by u B 1 ( x i , t ) , then we have
u i ( t ) u B 1 ( x i , t ) = k = 0 K ϑ k ( t ) u i k .
Next, we will consider the BRIF of u ( x , t ) at interpolation nodes ( x i , t k ) ( i [ I ] , k [ K ] ) . Similar to Equation (6), we can get the following BRIF of u ( x , t ) denoted by u B 2 ( x , t ) ,
u ( x , t ) i = 0 I ϕ i ( x ) u i ( t ) i = 0 I ϕ i ( x ) u B 1 ( x i , t ) = i = 0 I k = 0 K ϕ i ( x ) ϑ k ( t ) u i k = u B 2 ( x , t ) .
In this paper, the second class of Chebyshev nodes ( x i = cos i π I , i = 0 , 1 , , I ) will be used for analysis and calculation.

2.2. The Differential Matrices

In this subsection, we will consider the differential matrix of barycentric rational interpolation. As in [31], we can obtain the BRIF for the μ -order ( μ N + ) derivative of u ( x , t ) on nodes ( x n , t m ) ( n [ I ] , m [ K ] )
( μ ) u B 2 x μ = i = 0 I k = 0 K ϕ i ( μ ) ( x n ) ϑ k ( t m ) u i k = i = 0 I k = 0 K A n i ( μ ) ϑ k ( t m ) u i k , ( μ ) u B 2 t μ = i = 0 I k = 0 K ϕ i ( x n ) ϑ k ( μ ) ( t m ) u i k = i = 0 I k = 0 K ϕ i ( x n ) B m k ( μ ) u i k ,
where A n i ( μ ) and B m k ( μ ) are the μ -order differential matrices for the corresponding variables, respectively. As in [31], the form of A n i ( μ ) and B m k ( μ ) can be obtained as follows,
A n i ( 1 ) = w i w n 1 x n x i , n i , l = 0 , l n n A n l ( 1 ) , n = i , A n i ( μ ) = μ A n i ( 1 ) A n n ( μ 1 ) A n i ( μ 1 ) x n x i , n i , l = 0 , l n I A n l ( μ ) , n = i ,
B m k ( 1 ) = w k w m 1 t m t k , m k , q = 0 , q m K B m q ( 1 ) , m = k , B m k ( μ ) = μ B m k ( 1 ) B m m ( μ 1 ) B m k ( μ 1 ) t m t k , m k , q = 0 , q m K B m q ( μ ) , m = k .

2.3. Approximate Scheme of the Caputo Derivative

We consider the approximate scheme of the Caputo derivative in this subsection. By Equation (2), we can infer that
D t β 0 C u ( x , t ) = 1 Γ ( 2 β ) u ( x , 0 ) s t 1 β + 1 Γ ( 2 β ) 0 t ( t s ) 1 β 2 u ( x , s ) s 2 d s .
In Equation (11), by using Equation (10) and discretizing the domain by I + 1 ( 0 = x 0 < x 1 < < x I = 1 ) nodes in space and K + 1 ( 0 = t 0 < t 1 < < t K = 1 ) nodes in time, a preliminary discrete scheme for the Caputo derivative can be obtained as
D t β 0 C u ( x , t ) = t 1 β Γ ( 2 β ) i = 0 I k = 0 K ϕ i ( x ) ϑ k ( t 0 ) u i k + 1 Γ ( 2 β ) 0 t ( t s ) 1 β i = 0 I k = 0 K ϕ i ( x ) ϑ k ( s ) u i k d s .
For the second term at the right-hand side of Equation (12), by using the Gauss–Legendre quadrature rule, that is 0 t f ( s ) d s r = 1 R W r f ( s r ) (see [32] for more details), we can get the discrete scheme of it. Denote
D t β 0 C u ( x , t ) = 1 Γ ( 2 β ) i = 0 I k = 0 K t 1 β ϕ i ( x ) ϑ k ( t 0 ) u i k + r = 1 R ( t s r ) 1 β ϕ i ( x ) ϑ k ( s r ) W r u i k .
Then, we can obtain the fully discrete scheme of the Caputo derivative as D t β 0 C u ( x , t ) , that is D t β 0 C u ( x , t ) D t β 0 C u ( x , t ) .

2.4. Discrete Scheme of Equation (1)

In this subsection, the fully discrete scheme of Equation (1) will be given based on the BRICM. Applying Equations (10) and (13), we get
1 Γ ( 2 β ) t 1 β i = 0 I k = 0 K ϕ i ( x ) ϑ k ( t 0 ) u i k + i = 0 I k = 0 K r = 1 R ( t s r ) 1 β ϕ i ( x ) ϑ k ( s r ) W r u i k + j = 1 J λ j 1 Γ ( 2 β j ) i = 0 I k = 0 K t 1 β j ϕ i ( x ) ϑ k ( t 0 ) u i k + r = 1 R ( t s r ) 1 β j ϕ i ( x ) ϑ k ( s r ) W r u i k = i = 0 I k = 0 K P ( x , t ) ϕ i ( x ) Q ( x , t ) ϕ i ( x ) ϑ k ( t ) u i k + f ( x , t ) .
Let L k ( t m ) = r = 1 R ( t m s r ) 1 β ϑ k ( s r ) W r and L j k ( t m ) = r = 1 R ( t m s r ) 1 β j ϑ k ( s r ) W r , combining these results, ( ϑ k ( t 0 ) = B 0 k ( 1 ) , ϕ i ( x n ) = δ i n , ϕ i ( x n ) = A n i ( 1 ) and ϕ i ( x n ) = A n i ( 2 ) ), and by Equation (14), we can get
1 Γ ( 2 β ) t m 1 β i = 0 I k = 0 K δ i n B 0 k ( 1 ) u i k + i = 0 I k = 0 K δ i n L k ( t m ) u i k + j = 1 J λ j 1 Γ ( 2 β j ) i = 0 I k = 0 K t m 1 β j δ i n B 0 k ( 1 ) u i k + δ i n L j k ( t m ) u i k = i = 0 I k = 0 K P ( x n , t m ) A n i ( 2 ) Q ( x n , t m ) A n i ( 1 ) δ k m u i k + f ( x n , t m ) .
Taking all values of n [ I ] and m [ K ] , the fully discrete scheme of Equation (1) can be expressed as
[ 1 Γ ( 2 β ) I N B ( 1 ) + I N L + j = 1 J λ j 1 Γ ( 2 β j ) I N B j ( 1 ) + I N L j P A ( 2 ) I M Q A ( 1 ) I M ] U = F ,
where U = [ u 0 u I ] with u i = [ u i 0 u i K ] , F = [ f 0 f I ] with f i = [ f ( x i , t 0 ) f ( x i , t K ) ] ,
B ( 1 ) = t 0 1 β B 00 ( 1 ) t 0 1 β B 0 K ( 1 ) t K 1 β B 00 ( 1 ) t K 1 β B 0 K ( 1 ) , L = L 0 ( t 0 ) L K ( t 0 ) L 0 ( t K ) L K ( t K ) ,
B j ( 1 ) = t 0 1 β j B 00 ( 1 ) t 0 1 β j B 0 K ( 1 ) t K 1 β j B 00 ( 1 ) t K 1 β j B 0 K ( 1 ) , L j = L j 0 ( t 0 ) L j K ( t 0 ) L j 0 ( t K ) L j K ( t K ) ,
A ( μ ) = A 00 ( μ ) A 0 I ( μ ) A I 0 ( μ ) A I I ( μ ) with μ = 1 , 2 ,
P = d i a g { P 0 , P 1 , , P I } with P i = [ P ( x i , t 0 ) P ( x i , t K ) ] , and Q = d i a g { Q 0 , Q 1 , , Q I } with Q i = [ Q ( x i , t 0 ) Q ( x i , t K ) ] . I M and I N are the identity matrices of order K + 1 and I + 1 , respectively, and the discrete formats of the initial value conditions are
u B 2 ( x , 0 ) = i = 0 I k = 0 K ϕ i ( x ) ϑ k ( 0 ) u i k = ψ ( x ) , u B 2 ( 0 , t ) = i = 0 I k = 0 K ϕ i ( 0 ) ϑ k ( t ) u i k = 0 a n d u B 2 ( L , t ) = i = 0 I k = 0 K ϕ i ( L ) ϑ k ( t ) u i k = 0 .

3. Convergence Analysis

Set h = max { h i } and h i = | x i + 1 x i | with i [ I 1 ] , and set τ = max { τ k } and τ k = | t k + 1 t k | with k [ K 1 ] . Let u i = u ( x i ) , u k = u ( t k ) and u i k = u ( x i , t k ) . Let J 1 be the function space consisting of the interpolated basis functions ϕ i i = 0 I defined by Equation (6), and J 2 be the function space consisting of ϑ k k = 0 K defined by Equation (7). We first give the following definitions and lemmas, which will be used in the following discussion.
Definition 1. 
Suppose u ( x ) C [ 0 , 1 ] and u ( t ) C [ 0 , 1 ] . Let R x , I : C [ 0 , 1 ] J 1 and R t , K : C [ 0 , 1 ] J 2 be the interpolation operators for x and t, respectively. They still satisfy the requirement that
R x , I u ( x ) = i = 0 I ϕ i ( x ) u i , R t , K u ( t ) = k = 0 K ϑ k ( t ) u k .
Similarly, let J = J 1 J 2 ; we can define R x , I R t , K : C ( [ 0 , 1 ] × [ 0 , 1 ] ) J , and it satisfies
R x , I R t , K u ( x , t ) = i = 0 I k = 0 K ϕ i ( x ) ϑ k ( t ) u i k .
It is obvious that u B 2 : = R x , I R t , K u , and R x , I , R t , K , R x , I R t , K are linear operators.
Definition 2 
([33]). (Lebesgue constant) λ I , d x = | | R x , I | | = max x [ 0 , 1 ] i = 0 I | ϕ i ( x ) | , λ K , d t = | | R t , K | | = max t [ 0 , 1 ] k = 0 K | ϑ k ( t ) | .
Lemma 1 
([34]). For any set of well-spaced interpolation nodes, any μ 1 with 0 μ 1 d x , and u ( x ) C d x + 2 + μ 1 [ 0 , 1 ] , then
| u ( μ 1 ) ( x ) u B 1 ( μ 1 ) ( x ) | C 1 h d x + 1 μ 1
and more specifically, | u ( μ 1 ) ( x ) u B 1 ( μ 1 ) ( x ) | C 1 h i d x + 1 μ 1 , where C 1 is a constant, u B 1 is the BRIF of u ( x ) , x [ x i , x i + 1 ] and i [ I 1 ] .
Lemma 2 
([35]). When the BRICM at quasi-equidistant nodes ( x 0 , , x I ) R ( I + 1 ) with the basis function of Equation (6), its Lebesgue constant λ I , d x satisfies
λ I , d x ( 2 + H x ln I ) 3 4 H x , i f d x = 0 , 2 d x 1 H x d x , i f d x 1 ,
where H x h / h x , h x = min { | x i + 1 x i | } and i [ I 1 ] .
Lemma 3 
([32]). Let u ( 2 R ) ( t ) C [ 0 , T ] ; there exists γ ( 0 , T ) , and the error estimate of the Gauss–Legendre quadrature rule can be presented as follows
0 T u ( s ) d s r = 1 R W r u ( s r ) = T 2 R + 1 ( R ! ) 4 ( 2 R + 1 ) [ ( 2 R ) ! ] 3 u ( 2 R ) ( γ ) ,
where s r and W r are the integral points and integral weights of the Gauss–Legendre quadrature rule, respectively, and R is the number of points.
By a similar analysis, the following theorems can be obtained, as performed in [34,35].
Theorem 1. 
Let u = u ( x , t ) be the exact solution of Equation (1) and u B 2 = u B 2 ( x , t ) be the numerical solution of Equation (16), and let u x ( d x + 2 ) ( x , t ) , u t ( d t + 2 ) ( x , t ) C ( Ω ) with d x 1 and d t 1 , then
| | u u B 2 | | C 1 h d x + 1 + C 2 ( 2 + H x l n I ) 2 d x 1 H x d x τ d t + 1
where C 1 , C 2 are all constants.
Proof 
Applying triangle inequality, we can assert that
| | u u B 2 | | = | | u R x , I R t , K u | | = | | u R x , I u + R x , I u R x , I R t , K u | | | | u R x , I u | | + | | R x , I u R x , I R t , K u | | .
From Lemma 1, it follows that
| | u R x , I u | | C 1 h d x + 1 a n d | | u R t , K u | | C 2 τ d t + 1 .
Since R x , I and R t , K are linear operators, Definition 2 and Equation (17) show that
| | R x , I u R x , n R t , K u | | | | R x , I ( u R t , K u ) | | | | R x , I | | | | u R t , K u | | C 2 λ I , d x τ d t + 1 .
According to the above remark and Lemma 2, we have
| | R x , K u R x , I R t , K u | | C 2 ( 2 + H x l n I ) 2 d x 1 H x d x τ d t + 1 .
By Equation (17) and Equation (18), we conclude that
| | u u B 2 | | C 1 h d x + 1 + C 2 ( 2 + H x l n I ) 2 d x 1 H x d x τ d t + 1 .
This completes the proof. □
Theorem 2. 
Let D t β 0 C u ( x , t ) be the fully discrete scheme of D t β 0 C u ( x , t ) as in Equation (13) and let u x ( d x + 2 ) ( x , t ) , u t ( d t + 2 ) ( x , t ) C ( [ 0 , 1 ] × [ 0 , 1 ] ) with d x 1 and d t 1 . Then, the following holds:
| | 0 C D t β u ( x , t ) D t β 0 C u ( x , t ) | | C 3 h d x + 1 + C 4 ( 2 + H x l n I ) 2 d x 1 H x d x τ d t 1 + C 5 ( R ! ) 4 ( 2 R + 1 ) [ ( 2 R ) ! ] 3 ,
where C 3 , C 4 and C 5 are constants.
Proof. 
Applying triangle inequality, we deduce that
| | 0 C D t β u ( x , t ) D t β 0 C u ( x , t ) | | | | D t β 0 C u ( x , t ) D t β 0 C u ( x , t ) | | + | | D t β 0 C u ( x , t ) D t β 0 C u ( x , t ) | | .
By Theorem 1, subtracting Equation (12) from Equation (11) yields
| | 0 C D t β u ( x , t ) 0 C D t β u ( x , t ) | | = 1 Γ ( 2 β ) u ( x , 0 ) s R x , I R t , K u ( x , 0 ) s t 1 β + 0 t ( t s ) 1 β 2 u ( x , s ) s 2 2 R x , I R t , K u ( x , s ) s 2 d s 1 Γ ( 2 β ) ( u ( x , 0 ) s R x , I R t , K u ( x , 0 ) s + 0 t ( t s ) 1 β 2 u ( x , s ) s 2 2 R x , I R t , K u ( x , s ) s 2 d s )
Similarly to Theorem 1, by Lemma 1 and R t , K u ( x , 0 ) = u ( x , 0 ) , we can deduce that
u ( x , 0 ) s R x , I R t , K u ( x , 0 ) s u ( x , 0 ) s R x , I u ( x , 0 ) s + R x , I u ( x , 0 ) s R x , I R t , K u ( x , 0 ) s = u ( x , 0 ) s R x , I u ( x , 0 ) s C 1 h d x + 1
and
2 u ( x , s ) s 2 2 R x , I R t , K u ( x , s ) s 2 2 u ( x , s ) s 2 2 R x , I u ( x , s ) s 2 + 2 R x , I u ( x , s ) s 2 2 R x , I R t , K u ( x , s ) s 2 2 u ( x , s ) s 2 R x , I 2 u ( x , s ) s 2 + R x , I 2 u ( x , s ) s 2 2 R t , K u ( x , s ) s 2 C 1 h d x + 1 + C 2 ( 2 + H x l n I ) 2 d x 1 H x d x τ d t 1 .
Combining Equation (20) and Equation (21), we have
| | 0 C D t β u ( x , t ) 0 C D t β u ( x , t ) | | 1 Γ ( 2 β ) C 1 t 1 β h d x + 1 + t 2 β 2 β ( C 1 h d x + 1 + C 2 ( 2 + H x l n I ) 2 d x 1 H x d x τ d t 1 ) C 3 h d x + 1 + C 4 ( 2 + H x l n I ) 2 d x 1 H x d x τ d t 1 .
By Lemma 3, if we subtract Equation (13) from Equation (12), then it holds that
0 C D t β u ( x , t ) D t β 0 C u ( x , t ) = 1 Γ ( 2 β ) i = 0 I k = 0 K ϕ i ( x ) 0 t ( t s ) 1 β ϑ k ( s ) u i k d s r = 1 R W r ( t s r ) 1 β ϑ k ( s r ) u i k 1 Γ ( 2 β ) ( R ! ) 4 ( 2 R + 1 ) [ ( 2 R ) ! ] 3 i = 0 I k = 0 K ϕ i ( x ) u i k [ ( 1 s ) 1 β ϑ k ( s ) ] ( 2 R ) C 5 ( R ! ) 4 ( 2 R + 1 ) [ ( 2 R ) ! ] 3 ,
where C 5 is a constant and [ ( 1 s ) 1 β ϑ k ( s ) ] ( 2 R ) is the 2 p -order derivative of ( 1 s ) 1 β ϑ k ( s ) with respect to s. According to the above remark and Equation (19), we have
| | D t β 0 C u ( x , t ) D t β 0 C u ( x , t ) | | C 3 h d x + 1 + C 4 ( 2 + H x l n I ) 2 d x 1 H x d x τ d t + 1 + C 5 ( R ! ) 4 ( 2 R + 1 ) [ ( 2 R ) ! ] 3 .
This proves the theorem. □

4. Numerical Examples

This section demonstrates the effectiveness of the BRICM in solving multi-term time-fractional diffusion problems through four examples. All numerical results are implemented on a AMD Ryzen 5 5600H Windows 10 system by using MATLAB R2022b. The absolute errors E a and relative errors E r in all Tables are defined as
E a = max | u B 2 ( x i , t k ) u ( x i , t k ) | , E r = max | u B 2 ( x i , t k ) u ( x i , t k ) | max | u ( x i , t k ) | ,
and the absolute errors in all Figures are denoted by
| u B 2 ( x i , t k ) u ( x i , t k ) | ,
where u ( x i , t k ) is the exact solution and u B 2 ( x i , t k ) is the numerical solution, respectively. The convergence order is defined by log ( E 1 / E 2 ) / ( log ( I 2 × K 2 I 1 × K 1 ) ) , where E 2 is the current error and E 1 is the previous error, I 2 and K 2 are the numbers of current nodes, I 1 and K 1 are the numbers of previous nodes.
Example 1. 
Consider the following one-term time-fractional convection–diffusion equation:
D t 0.1 0 C u ( x , t ) = P ( x , t ) 2 u ( x , t ) x 2 Q ( x , t ) u ( x , t ) x + f ( x , t ) , ( x , y ) [ 0 , 1 ] × [ 0 , 1 ] ,
with u ( x , 0 ) = 0 , P ( x , t ) = 128 t 0.2 Γ ( 1.9 ) , Q ( x , t ) = 64 t 0.1 cos ( π x ) Γ ( 1.9 ) and
f ( x , t ) = sin ( π x ) Γ ( 1.9 ) F 2 1 1 ; [ 0.95 1.45 ] ; π 2 t 2 4 π t 0.9 + 128 π 2 t 0.2 sin ( π t ) + 64 π t 0.1 Γ ( 1.9 ) cos 2 ( π x ) sin ( π t ) ,
where F 2 1 ( 1 ; [ 0.95 1.45 ] ; π 2 t 2 4 ) represents the generalized hypergeometric function. The exact solution of this example is u ( x , t ) = sin ( π t ) sin ( π x ) .
Table 1 shows the E a and convergence order with d x = d t = I = K , in which 1000 Gaussian nodes are used. Table 2 shows the comparison of results for Example 1 at different nodes (the second class of Chebyshev nodes and the equidistant nodes), in which d x = d t = I = K and the number of Gaussian nodes is 1000. We perceive from these tables that the present scheme maintains the high-order accuracy, which fits well with the theoretical analysis.
Example 2. 
Consider the following two-term time-fractional diffusion equation [36]:
D t β 0 C + D t β 1 0 C u ( x , t ) = 2 u ( x , t ) x 2 + f ( x , t ) , ( x , y ) [ 0 , 1 ] × [ 0 , 1 ] ,
with u ( x , 0 ) = 0 and
f ( x , t ) = 2 t 2 β Γ ( 3 β ) sin ( 2 π x ) + 2 t 2 β 1 Γ ( 3 β 1 ) sin ( 2 π x ) + 4 π 2 t 2 sin ( 2 π x ) .
The exact solution of this example is u ( x , t ) = t 2 sin ( 2 π x ) .
In the second example, 1100 Gaussian nodes are used for numerical calculations. Table 3 shows the absolute errors for Example 2 and compares the present results with the results obtained by the method in [36]. We perceive from Table 3 that the results obtained by the proposed method are more accurate than the results in [36]. Table 4 shows that the second class of Chebyshev nodes is more suitable for this study than the equidistant nodes. Absolute errors and corresponding convergence orders with different fractional orders are listed in Table 5 for Example 2 with d x = d t = I = K . For the given β , Table 6 shows the relative errors for Example 2 at various time levels with d x = d t = I = K = 14 and β 1 = 0.1 .
In Figure 1, Figure 2 and Figure 3, we solve Example 2 by the present method with d x = d t = I = K = 14 and p = 1100 . The exact solution, numerical solution, absolute error, and the contour plot of absolute error for β = 0.2 , β 1 = 0.1 and β = 0.8 , β 1 = 0.4 are displayed in Figure 1 and Figure 2, respectively. For β = 0.9 and β 1 = 0.5 , the numerical solution and the situation of solutions at various time levels are shown in Figure 3. We perceive from Figure 1, Figure 2 and Figure 3 that the numerical solution agrees with the exact solution. Numerical results of the second example show the efficiency and applicability of the present method.
Example 3. 
Consider the following three-term time-fractional convection–diffusion equation without an exact solution:
D t β 0 C + j = 1 2 λ j D t β j 0 C u ( x , t ) = P ( x , t ) 2 u ( x , t ) x 2 Q ( x , t ) u ( x , t ) x + f ( x , t ) ,
with ( x , t ) [ 0 , 1 ] × [ 0 , 1 ] , u ( x , 0 ) = sin ( π x ) , λ 1 = λ 2 = 1 , P ( x , t ) = 16 t 0.2 β Γ ( 2 β ) sin ( π x ) ,   Q ( x , t ) = 8 t 0.1 β Γ ( 2 β ) cos ( π x ) , and f ( x , t ) is a given function.
In the third example, 900 Gaussian nodes and the equidistant nodes are used for numerical calculations. This example is used to test the case where the exact solution is unknown. We will use the solution on the fine grid ( I = K = 16 ) as the exact solution. The solutions on coarse grids are used as numerical solutions. For different β , Table 7 and Table 8 show the errors for Example 3 with d x = d t = I = K , β 1 = 0.2 , and β 2 = 0.1 . The calculation results show that the numerical method in our paper has high numerical calculation accuracy.
Example 4. 
Consider the following four-term time-fractional convection–diffusion equation under the hexagonal region:
D t β 0 C + j = 1 3 λ j D t β j 0 C u ( x , t ) = 2 u ( x , t ) x 2 u ( x , t ) x + f ( x , t ) ,
with u ( x , 0 ) = 0 , λ 1 = 1 , λ 2 = 2 , λ 3 = 0.5 . The exact solution of this example is u ( x , t ) = ( t 3 + t 2 ) sin ( π x ) and the corresponding forcing term f ( x , t ) can be obtained by the given conditions.
In the fourth example, 1330 Gaussian nodes and the second class of Chebyshev nodes are used for numerical calculations. The region where the nodes are located is as follows (see Figure 4):
The results of the fourth example are displayed in Table 9 and Table 10 and Figure 5. For different I, K, and β values, Table 9 shows the absolute errors and corresponding convergence orders for Example 4 with β 1 = 0.3 , β 2 = 0.2 , and β 3 = 0.1 . Table 10 shows the relative errors at various time levels. We perceive from these tables that the present method has high computational accuracy and a fast convergence speed. For different β values, the contour plots of absolute error with I = K = 10 , β 1 = 0.3 , β 2 = 0.2 , and β 3 = 0.1 are displayed in Figure 5. Numerical results of this example also show the efficiency and applicability of the present method.

5. Conclusions

In this paper, we give a fully discrete scheme for multi-term time-fractional convection–diffusion equations. The fully discrete scheme is constructed based on the BRICM and the Gauss–Legendre quadrature rule. We prove the convergence of the proposed scheme. The method proposed in this paper can achieve high computational accuracy with very few nodes. We present some numerical examples to illustrate the effectiveness of the method. A comparison of the obtained results with exact solutions and other existing methods reveals that our method is more accurate and efficient for multi-term time-fractional convection–diffusion equations.

Author Contributions

Conceptualization, X.Z.; software, Y.C. and L.W.; formal analysis, X.Z. and Y.C.; writing—original draft preparation, Y.C.; writing—review and editing, X.Z. and S.K.; visualization, Y.C. and L.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Scientific Research Foundation for Talents Introduction of Guizhou University of Finance and Economics (No. 2023YJ16) and the 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 cyan19981022@163.com (Y.C., the corresponding author).

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. Fang, Z.W.; Sun, H.W.; Wang, H. A fast method for variable-order Caputo fractional derivative with applications to time-fractional diffusion equations. Comput. Math. Appl. 2020, 80, 1443–1458. [Google Scholar] [CrossRef]
  2. Mehandiratta, V.; Mehra, M.; Leugering, G. Optimal control problems driven by time-fractional diffusion equations on metric graphs: Optimality system and finite difference approximation. SIAM J. Control Optim. 2021, 59, 4216–4242. [Google Scholar] [CrossRef]
  3. Gu, X.M.; Sun, H.W.; Zhao, Y.L.; Zheng, X. An implicit difference scheme for time-fractional diffusion equations with a time-invariant type variable order. Appl. Math. Lett. 2021, 120, 107270. [Google Scholar] [CrossRef]
  4. 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]
  5. Huang, C.B.; Chen, H.; An, N. β-robust superconvergent analysis of a finite element method for the distributed order time-fractional diffusion equation. J. Sci. Comput. 2022, 90, 44. [Google Scholar] [CrossRef]
  6. Zheng, X.C.; Wang, H. Optimal-order error estimates of finite element approximations to variable-order time-fractional diffusion equations without regularity assumptions of the true solutions. IMA J. Numer. Anal. 2021, 41, 1522–1545. [Google Scholar] [CrossRef]
  7. Feng, L.B.; Zhuang, P.; Liu, F.; Turner, I.; Gu, Y. Finite element method for space-time fractional diffusion equation. Numer. Algorithms 2016, 72, 749–767. [Google Scholar] [CrossRef]
  8. Liu, F.; Zhuang, P.; Turner, I.; Burrage, K.; Anh, V. A new fractional finite volume method for solving the fractional diffusion equation. Appl. Math. Model. 2014, 38, 3871–3878. [Google Scholar] [CrossRef]
  9. Li, J.; Liu, F.W.; Feng, L.B.; Turner, I. A novel finite volume method for the Riesz space distributed-order diffusion equation. Comput. Math. Appl. 2017, 74, 772–783. [Google Scholar] [CrossRef]
  10. Lipnikov, K.N.; Svyatskiy, D.; Vassilevski, Y.V. Interpolation-free monotone finite volume method for diffusion equations on polygonal meshes. J. Comput. Phys. 2009, 228, 703–716. [Google Scholar] [CrossRef]
  11. Zheng, M.L.; Liu, F.W.; Anh, V.V.; Turner, I. A high-order spectral method for the multi-term time-fractional diffusion equations. Appl. Math. Model. 2016, 40, 4970–4985. [Google Scholar] [CrossRef]
  12. Jafarzadeh, S.; Larios, A.; Bobaru, F. Efficient solutions for nonlocal diffusion problems via boundary-adapted spectral methods. J. Peridynamics Nonlocal Model. 2020, 2, 85–110. [Google Scholar] [CrossRef]
  13. Kumbinarasaiah, S. Hermite wavelets approach for the multi-term fractional differential equations. J. Interdiscip. Math. 2021, 24, 1241–1262. [Google Scholar] [CrossRef]
  14. Delkhosh, M.; Parand, K. A new computational method based on fractional Lagrange functions to solve multi-term fractional differential equations. Numer. Algorithms 2021, 88, 729–766. [Google Scholar] [CrossRef]
  15. Abd-Elhameed, W.M.; Alsuyuti, M.M. Numerical treatment of multi-term fractional differential equations via new kind of generalized Chebyshev polynomials. Fractal Fract. 2023, 7, 74. [Google Scholar] [CrossRef]
  16. Tural-Polat, S.N.; Dincel, A.T. Numerical solution method for multi-term variable order fractional differential equations by shifted Chebyshev polynomials of the third kind. Alex. Eng. J. 2022, 61, 5145–5153. [Google Scholar] [CrossRef]
  17. Hashemi, M.S.; Inc, M.; Hajikhah, S. Generalized squared remainder minimization method for solving multi-term fractional differential equations. Nonlinear Anal. Model. 2021, 26, 57–71. [Google Scholar] [CrossRef]
  18. Shiralashetti, S.C.; Deshi, A.B. An efficient Haar wavelet collocation method for the numerical solution of multi-term fractional differential equations. Nonlinear Dynam. 2016, 83, 293–303. [Google Scholar] [CrossRef]
  19. Guo, B.L.; Pu, X.K.; Huang, F.H. Fractional Partial Differential Equations and Their Numerical Solutions; World Scientific: Singapore, 2015. [Google Scholar]
  20. Hilfer, R. Applications of Fractional Calculus in Physics; World Scientific: Singapore, 2000. [Google Scholar]
  21. Podlubny, I. Fractional Differential Equations; Academic Press: New York, NY, USA, 1999. [Google Scholar]
  22. Torkaman, S.; Heydari, M.; Loghmani, G.B. A combination of the quasilinearization method and linear barycentric rational interpolation to solve nonlinear multi-dimensional Volterra integral equations. Math. Comput. Simula. 2023, 208, 366–397. [Google Scholar]
  23. Liu, H.Y.; Ma, Y.Y.; Li, H.; Zhang, W. Combination of discrete technique on graded meshes with barycentric rational interpolation for solving a class of time-dependent partial integro-differential equations with weakly singular kernels. Comput. Math. Appl. 2023, 141, 159–169. [Google Scholar]
  24. Fahimi-khalilabad, I.; Irandoust-Pakchin, S.; Abdi-Mazraeh, S. High-order finite difference method based on linear barycentric rational interpolation for Caputo type sub-diffusion equation. Math. Comput. Simula. 2022, 199, 60–80. [Google Scholar] [CrossRef]
  25. Li, J. Linear barycentric rational collocation method for solving biharmonic equation. Demonstr. Math. 2022, 55, 587–603. [Google Scholar] [CrossRef]
  26. Yang, M.M.; Ma, W.T.; Ge, Y.B. Barycentric rational interpolation method of the Helmholtz equation with irregular domain. Math. Model. Anal. 2023, 28, 330–351. [Google Scholar] [CrossRef]
  27. Li, J.; Cheng, Y.L. Barycentric rational interpolation method for solving time-dependent fractional convection-diffusion equation. Electron. Res. Arch. 2023, 31, 4034–4056. [Google Scholar] [CrossRef]
  28. Berrut, J.P.; Mittelmann, H.D. Lebesgue constant minimizing linear rational interpolation of continuous functions over the interval. Comput. Math. Appl. 1997, 33, 77–86. [Google Scholar] [CrossRef]
  29. Berrut, J.P. Rational functions for guaranteed and experimentallywell-conditioned global interpolation. Comput. Math. Appl. 1988, 15, 1–16. [Google Scholar] [CrossRef]
  30. Floater, M.S.; Hormann, K. Barycentric rational interpolation with no poles and high rates of approximation. Numer. Math. 2007, 107, 315–331. [Google Scholar] [CrossRef]
  31. Klein, G.; Berrut, J.P. Linear rational finite differences from derivatives of barycentric rational interpolants. SIAM J. Numer. Anal. 2012, 50, 643–656. [Google Scholar] [CrossRef]
  32. Kahaner, D.; Moler, C.; Nash, S. Numerical Methods and Software; Prentice-Hall, Inc.: Hoboken, NJ, USA, 1989. [Google Scholar]
  33. Fornberg, B. A practical Guide to Pseudospectral Methods; Cambridge University Press: Cambridge, UK, 1998. [Google Scholar]
  34. Cirillo, E.; Hormann, K.; Sidon, J. Convergence rates of derivatives of Floater-Hormann interpolants for well-spaced nodes. Appl. Numer. Math. 2017, 116, 108–118. [Google Scholar] [CrossRef]
  35. Hormann, K.; Klein, G.; Marchi, S.D. Barycentric rational interpolation at quasi-equidistant nodes. Dolomit. Res. Notes Approx. 2012, 5, 1–6. [Google Scholar]
  36. Ravi Kanth, A.S.V.; Garg, N. An implicit numerical scheme for a class of multi-term time-fractional diffusion equation. Eur. Phys. J. Plus 2019, 134, 312. [Google Scholar] [CrossRef]
Figure 1. Results of Example 2 with β = 0.2 and β 1 = 0.1 .
Figure 1. Results of Example 2 with β = 0.2 and β 1 = 0.1 .
Fractalfract 08 00687 g001
Figure 2. Results of Example 2 with β = 0.8 and β 1 = 0.4 .
Figure 2. Results of Example 2 with β = 0.8 and β 1 = 0.4 .
Fractalfract 08 00687 g002
Figure 3. Results of Example 2 with β = 0.9 and β 1 = 0.5 . (a) Numerical solution; (b) Numerical solution (special symbols) and exact solution (solid line) at various time levels.
Figure 3. Results of Example 2 with β = 0.9 and β 1 = 0.5 . (a) Numerical solution; (b) Numerical solution (special symbols) and exact solution (solid line) at various time levels.
Fractalfract 08 00687 g003
Figure 4. Distribution of solution area and solution nodes.
Figure 4. Distribution of solution area and solution nodes.
Fractalfract 08 00687 g004
Figure 5. Contour plots of absolute error for Example 4 at different β .
Figure 5. Contour plots of absolute error for Example 4 at different β .
Fractalfract 08 00687 g005
Table 1. E a and convergence order for Example 1 with d x = d t = I = K .
Table 1. E a and convergence order for Example 1 with d x = d t = I = K .
I × K E a Order CPU Time (s)
4 × 4 4.5310 × 10 3 0.1089
6 × 6 2.0196 × 10 5 6.67530.1445
8 × 8 1.0301 × 10 7 9.17400.2182
10 × 10 6.4164 × 10 10 11.37960.2408
Table 2. Comparison of results for Example 1 at different nodes with d x = d t = I = K .
Table 2. Comparison of results for Example 1 at different nodes with d x = d t = I = K .
I × K E a E r
Chebyshev Nodes Equidistant Nodes Chebyshev Nodes Equidistant Nodes
5 × 54.7064 × 10 4 4.0635 × 10 3 6.0160 × 10 4 4.4925 × 10 3
7 × 73.9313 × 10 6 1.0369 × 10 4 4.4536 × 10 6 1.0910 × 10 4
9 × 92.1562 × 10 8 1.6026 × 10 6 2.3250 × 10 8 1.6524 × 10 6
11 × 116.2475 × 10 10 1.6719 × 10 8 6.5704 × 10 10 1.7065 × 10 8
Table 3. Comparison of absolute errors for Example 2 with d x = d t = 4 and K = 10 .
Table 3. Comparison of absolute errors for Example 2 with d x = d t = 4 and K = 10 .
I β = 0.6 , β 1 = 0.4 β = 0.9 , β 1 = 0.1
Present Method Method in [36] Present Method Method in [36]
251.6819 × 10 6 3.5338 × 10 3 1.6817 × 10 6 3.5334 × 10 3
506.0084 × 10 8 8.8537 × 10 4 5.9549 × 10 8 8.8528 × 10 4
1008.6425 × 10 10 2.2146 × 10 4 1.4588 × 10 9 2.2144 × 10 4
2002.0880 × 10 10 5.5372 × 10 5 9.5881 × 10 10 5.5365 × 10 5
Table 4. Comparison of absolute error of Example 2 at different nodes with d x = d t = I = K .
Table 4. Comparison of absolute error of Example 2 at different nodes with d x = d t = I = K .
I × K β = 0.6 , β 1 = 0.4 β = 0.9 , β 1 = 0.1
Chebyshev Nodes Equidistant Nodes Chebyshev Nodes Equidistant Nodes
6 × 64.3316 × 10 3 4.5049 × 10 2 4.3319 × 10 3 4.4997 × 10 2
8 × 81.2194 × 10 4 4.2148 × 10 3 1.2193 × 10 4 4.2113 × 10 3
10 × 102.3307 × 10 6 2.3521 × 10 4 2.3305 × 10 6 2.3506 × 10 4
12 × 123.2037 × 10 8 8.8581 × 10 6 3.2093 × 10 8 8.8541 × 10 6
Table 5. E a and convergence order for Example 2 with d x = d t = I = K .
Table 5. E a and convergence order for Example 2 with d x = d t = I = K .
I × K β = 0.2 , β 1 = 0.1 β = 0.5 , β 1 = 0.25 β = 0.8 , β 1 = 0.4
E a Order E a Order E a Order
6 × 6 4.3384 × 10 3 4.3340 × 10 3 4.3298 × 10 3
8 × 8 1.2217 × 10 4 6.20451.2202 × 10 4 6.20491.2187 × 10 4 6.2053
10 × 10 2.3337 × 10 6 8.86862.3318 × 10 6 8.86772.3298 × 10 6 8.8669
12 × 12 3.2065 × 10 8 11.75793.2047 × 10 8 11.75723.2050 × 10 8 11.7546
Table 6. E r for Example 2 at various time levels with β 1 = 0.1 .
Table 6. E r for Example 2 at various time levels with β 1 = 0.1 .
t = 0.0125 t = 0.2831 t = 0.5 t = 0.7169 t = 1
β = 0.3 3.4117 × 10 10 3.4003 × 10 10 3.3950 × 10 10 3.3972 × 10 10 3.3953 × 10 10
β = 0.4 3.4392 × 10 10 3.3966 × 10 10 3.3948 × 10 10 3.3950 × 10 10 3.3955 × 10 10
β = 0.6 6.2334 × 10 10 3.3713 × 10 10 3.3797 × 10 10 3.3795 × 10 10 3.3803 × 10 10
β = 0.7 6.8880 × 10 10 3.2667 × 10 10 3.3074 × 10 10 3.3288 × 10 10 3.3408 × 10 10
Table 7. E a and convergence order for Example 3 with β 1 = 0.2 and β 2 = 0.1 .
Table 7. E a and convergence order for Example 3 with β 1 = 0.2 and β 2 = 0.1 .
I × K β = 0.3 β = 0.5 β = 0.7
E a Order E a Order E a Order
4 × 4 1.6758 × 10 2 1.6752 × 10 2 1.6778 × 10 2
6 × 6 4.0572 × 10 4 4.58854.0565 × 10 4 4.58834.0809 × 10 4 4.5828
8 × 8 6.4739 × 10 6 7.19186.4784 × 10 6 7.19036.5555 × 10 6 7.1801
10 × 10 6.5814 × 10 8 10.28196.6072 × 10 8 10.27476.7627 × 10 8 10.2491
Table 8. E r for Example 3 with β 1 = 0.2 and β 2 = 0.1 at different β .
Table 8. E r for Example 3 with β 1 = 0.2 and β 2 = 0.1 at different β .
I × K 4 × 4 6 × 6 8 × 8 10 × 10
β = 0.3 8.3788 × 10 3 2.0286 × 10 4 3.2370 × 10 6 3.2907 × 10 8
β = 0.5 8.3758 × 10 3 2.0282 × 10 4 3.2392 × 10 6 3.3036 × 10 8
β = 0.7 8.3890 × 10 3 2.0404 × 10 4 3.2777 × 10 6 3.3813 × 10 8
β = 0.9 8.4041 × 10 3 2.0540 × 10 4 3.3215 × 10 6 3.2910 × 10 8
Table 9. E a and convergence order for Example 4 with β 1 = 0.3 , β 2 = 0.2 and β 3 = 0.1 .
Table 9. E a and convergence order for Example 4 with β 1 = 0.3 , β 2 = 0.2 and β 3 = 0.1 .
I × K β = 0.4 β = 0.6 β = 0.8
E a Order E a Order E a Order
4 × 4 8.0841 × 10 2 9.0868 × 10 2 1.1011 × 10 1
6 × 6 2.5106 × 10 4 4.16552.4138 × 10 4 4.27822.1358 × 10 4 7.7013
8 × 8 2.2096 × 10 6 8.22592.1925 × 10 6 8.17112.1735 × 10 6 7.9735
10 × 10 8.9602 × 10 9 12.34138.6162 × 10 9 12.41161.1141 × 10 8 11.8163
Table 10. E r for Example 4 at various time levels with β 1 = 0.3 , β 2 = 0.2 and β 3 = 0.1 .
Table 10. E r for Example 4 at various time levels with β 1 = 0.3 , β 2 = 0.2 and β 3 = 0.1 .
t = 0.0245 t = 0.2061 t = 0.5 t = 0.7939 t = 1
β = 0.35 2.1999 × 10 8 5.2842 × 10 9 4.2828 × 10 9 4.4389 × 10 9 4.4982 × 10 9
β = 0.55 3.9612 × 10 8 6.4776 × 10 9 4.6389 × 10 9 4.2809 × 10 9 4.3786 × 10 9
β = 0.75 1.2493 × 10 7 1.2395 × 10 8 6.5162 × 10 9 5.2498 × 10 9 4.9184 × 10 9
β = 0.95 1.5758 × 10 6 1.0574 × 10 7 3.1175 × 10 8 1.7277 × 10 8 1.3319 × 10 8
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.; Chen, Y.; Wei, L.; Kumar, S. Numerical Simulation Based on Interpolation Technique for Multi-Term Time-Fractional Convection–Diffusion Equations. Fractal Fract. 2024, 8, 687. https://doi.org/10.3390/fractalfract8120687

AMA Style

Zhang X, Chen Y, Wei L, Kumar S. Numerical Simulation Based on Interpolation Technique for Multi-Term Time-Fractional Convection–Diffusion Equations. Fractal and Fractional. 2024; 8(12):687. https://doi.org/10.3390/fractalfract8120687

Chicago/Turabian Style

Zhang, Xindong, Yan Chen, Leilei Wei, and Sunil Kumar. 2024. "Numerical Simulation Based on Interpolation Technique for Multi-Term Time-Fractional Convection–Diffusion Equations" Fractal and Fractional 8, no. 12: 687. https://doi.org/10.3390/fractalfract8120687

APA Style

Zhang, X., Chen, Y., Wei, L., & Kumar, S. (2024). Numerical Simulation Based on Interpolation Technique for Multi-Term Time-Fractional Convection–Diffusion Equations. Fractal and Fractional, 8(12), 687. https://doi.org/10.3390/fractalfract8120687

Article Metrics

Back to TopTop