
In recent years, the application of variable-order (VO) fractional differential equations for describing complex physical phenomena ranging from biology, hydrology, mechanics and viscoelasticity to fluid dynamics has become one of the most hot topics in the context of scientific modeling. An interesting aspect of VO operators is their capability to address the behavior of scientific and engineering systems with time and spatially varying properties. The VO fractional diffusion equation is a fundamental model that allows transitions among sub-diffusive, diffusive and super-diffusive behaviors without altering the underlying governing equations. In this paper, we considered the two-dimensional fractional diffusion equation with the Caputo time VO derivative, which is essential for describing anomalous diffusion in real-world complex systems. A new Crank-Nicolson (C-N) difference scheme and an efficient explicit decoupled group (EDG) method were proposed to solve the problem under consideration. The proposed EDG method is based on a skewed difference scheme in conjunction with a grouping procedure of the solution grid points. Special attention was devoted to investigating the stability and convergence of the proposed methods. Three numerical examples with known exact analytical solutions were provided to illustrate our considerations. The proposed methods were shown to be stable and convergent theoretically as well as numerically. In addition, a comparative study was done between the EDG method and the C-N difference scheme. It was found that the proposed methods are accurate in simulating the considered problem, while the EDG method is superior to the C-N difference method in terms of Central Processing Unit (CPU) timing, verifying the efficiency of the former method in solving the VO problem.
Citation: Fouad Mohammad Salama, Faisal Fairag. On numerical solution of two-dimensional variable-order fractional diffusion equation arising in transport phenomena[J]. AIMS Mathematics, 2024, 9(1): 340-370. doi: 10.3934/math.2024020
[1] | Ajmal Ali, Tayyaba Akram, Azhar Iqbal, Poom Kumam, Thana Sutthibutpong . A numerical approach for 2D time-fractional diffusion damped wave model. AIMS Mathematics, 2023, 8(4): 8249-8273. doi: 10.3934/math.2023416 |
[2] | Muhammad Asim Khan, Norma Alias, Umair Ali . A new fourth-order grouping iterative method for the time fractional sub-diffusion equation having a weak singularity at initial time. AIMS Mathematics, 2023, 8(6): 13725-13746. doi: 10.3934/math.2023697 |
[3] | Din Prathumwan, Thipsuda Khonwai, Narisara Phoochalong, Inthira Chaiya, Kamonchat Trachoo . An improved approximate method for solving two-dimensional time-fractional-order Black-Scholes model: a finite difference approach. AIMS Mathematics, 2024, 9(7): 17205-17233. doi: 10.3934/math.2024836 |
[4] | Bin Fan . Efficient numerical method for multi-term time-fractional diffusion equations with Caputo-Fabrizio derivatives. AIMS Mathematics, 2024, 9(3): 7293-7320. doi: 10.3934/math.2024354 |
[5] | Zunyuan Hu, Can Li, Shimin Guo . Fast finite difference/Legendre spectral collocation approximations for a tempered time-fractional diffusion equation. AIMS Mathematics, 2024, 9(12): 34647-34673. doi: 10.3934/math.20241650 |
[6] | Emadidin Gahalla Mohmed Elmahdi, Jianfei Huang . Two linearized finite difference schemes for time fractional nonlinear diffusion-wave equations with fourth order derivative. AIMS Mathematics, 2021, 6(6): 6356-6376. doi: 10.3934/math.2021373 |
[7] | Lijuan Nong, An Chen, Qian Yi, Congcong Li . Fast Crank-Nicolson compact difference scheme for the two-dimensional time-fractional mobile/immobile transport equation. AIMS Mathematics, 2021, 6(6): 6242-6254. doi: 10.3934/math.2021366 |
[8] | Fouad Mohammad Salama, Nur Nadiah Abd Hamid, Norhashidah Hj. Mohd Ali, Umair Ali . An efficient modified hybrid explicit group iterative method for the time-fractional diffusion equation in two space dimensions. AIMS Mathematics, 2022, 7(2): 2370-2392. doi: 10.3934/math.2022134 |
[9] | Abdul-Majeed Ayebire, Inderpreet Kaur, Dereje Alemu Alemar, Mukhdeep Singh Manshahia, Shelly Arora . A robust technique of cubic Hermite splines to study the non-linear reaction-diffusion equation with variable coefficients. AIMS Mathematics, 2024, 9(4): 8192-8213. doi: 10.3934/math.2024398 |
[10] | Fouad Mohammad Salama, Nur Nadiah Abd Hamid, Umair Ali, Norhashidah Hj. Mohd Ali . Fast hybrid explicit group methods for solving 2D fractional advection-diffusion equation. AIMS Mathematics, 2022, 7(9): 15854-15880. doi: 10.3934/math.2022868 |
In recent years, the application of variable-order (VO) fractional differential equations for describing complex physical phenomena ranging from biology, hydrology, mechanics and viscoelasticity to fluid dynamics has become one of the most hot topics in the context of scientific modeling. An interesting aspect of VO operators is their capability to address the behavior of scientific and engineering systems with time and spatially varying properties. The VO fractional diffusion equation is a fundamental model that allows transitions among sub-diffusive, diffusive and super-diffusive behaviors without altering the underlying governing equations. In this paper, we considered the two-dimensional fractional diffusion equation with the Caputo time VO derivative, which is essential for describing anomalous diffusion in real-world complex systems. A new Crank-Nicolson (C-N) difference scheme and an efficient explicit decoupled group (EDG) method were proposed to solve the problem under consideration. The proposed EDG method is based on a skewed difference scheme in conjunction with a grouping procedure of the solution grid points. Special attention was devoted to investigating the stability and convergence of the proposed methods. Three numerical examples with known exact analytical solutions were provided to illustrate our considerations. The proposed methods were shown to be stable and convergent theoretically as well as numerically. In addition, a comparative study was done between the EDG method and the C-N difference scheme. It was found that the proposed methods are accurate in simulating the considered problem, while the EDG method is superior to the C-N difference method in terms of Central Processing Unit (CPU) timing, verifying the efficiency of the former method in solving the VO problem.
Nomenclature
VO: Variable-order.
CO: Constant-order.
C-N: Crank-Nicolson.
EDG: Explicit decoupled group.
CPU: Central processing unit.
VO-TFDE: Variable-order time fractional diffusion equation.
Symbols
M1: Number of spatial steps in the x-direction.
M2: Number of spatial steps in the y-direction.
N: Number of temporal steps.
Δx: Length of spatial step along x-direction.
Δy: Length of spatial step along y-direction.
τ: Temporal step size.
T: Final time.
K1: Diffusion coefficient.
K2: Diffusion coefficient.
α: Fractional order.
A plethora of problems in engineering applications can be described by integer-order differential equations [1]. On the other hand, fractional differential equations have numerous applications in many sciences, which makes them one of the most prominent tools of fractional calculus that have attracted the interest of scholars and researchers in recent years. For instance, Lozynskyy et al. [2] applied the Caputo-Fabrizio operator to describe the elastic moment in a long shaft and produced an enhanced mathematical model in the form of a fractional ordinary differential equation for a two-mass system with concentrated parameters. Asjad et al. [3] successfully described the convection heat transfer in clay nanofluid by utilizing a Maxwell model. The authors used the newly developed hybrid fractional derivative in [4] to generalize the classical Maxwell model to the novel fractional one. Liu et al. [5] established a new Green-Ampt model containing a Caputo fractional derivative to describe an important phenomenon in geotechnical engineering; that is, the infiltration process in slopes. The researchers argued that the outcomes of their model were better than those of Chen and Young's model and very close to the actual experimental results. Kumar et al. [6] provided an opportunity for a better understanding of climate change by introducing a fractional-order model that describes the plankton-oxygen dynamics based on the Liouville-Caputo fractional derivative. Mustafa Inc et al. [7] considered a novel Covid-19 model based on the Caputo fractional operator to study the dynamics of the disease spread. The results of the numerical simulations were in good match with the real data and showed the advantages of the fractional derivative. Shen et al. [8] got more insight into the case where a patient has two infections at the same time by considering a fractional co-infection model under the Caputo fractional derivative.
Differential equations with constant-order (CO) fractional derivatives have been considered as powerful instruments for characterizing the long memory of dynamic processes and real-world phenomena. This means modeling the case in which the current state of the system depends on all its historical states rather than only some recent ones. More theoretical and comprehensive discussions of fractional calculus can be found in [9,10,11,12,13,14]. In recent years, many researchers have found systems exhibiting variable memory that cannot be described adequately by CO fractional operators. This fact led to the generalization of the CO fractional calculus to the so-called variable-order (VO) fractional calculus, where the derivatives of fractional operators are functions of time and/or space variables. Sun et al. [15] introduced an important comparative study between constant and VO operators in describing the memory property of systems. It was Samko and Ross [16] who gave the definitions of VO derivatives for the first time in 1993. Later, Lorenzo and Hartley [17,18] and Coimbra [19] constructed the mathematical framework of VO operators and discussed some of their potential applications in mechanics. These works paved the way for the development of new definitions, which served as a springboard for the field of VO fractional calculus. The list of the definitions in the literature includes: Riemann-Liouville, Caputo, Grunwald-Letnikov, Hadamard, Caputo-Fabrizio, Atangana-Baleanu and more. These definitions are reported to describe complex dynamic problems effectively. Consequently, with the extension of CO fractional models to their counterparts, VO models have grown in a remarkable manner. Examples are the diffusion equation [20], reaction-diffusion equation [21], Burgers equation cite{r21}, telegraph equation [23], mobile/immobile equation [24], Schrödinger equation [25] and more. For more details about the VO fractional calculus, the interested reader is referred to [26,27,28].
In the last few years, VO fractional differential equations have achieved tremendous success in modeling physical problems of evolutionary nature-type. The reason for this is the ability of VO operators to adjust the system's order based on its present or even its previous response effectively. For example, Sweilam et al. [29] proposed a VO fractional model to simulate the healing process of the cancer disease in biological systems. The established model involved VO Caputo derivatives with two control variables to reduce the number of cancer cells. Xiang and others [30] made use of the VO Caputo derivative and came up with a novel fractional model for describing the stress responses of glassy polymers. The authors noticed a relationship between the varying order and the behavior of microstructures, which opens up a new avenue for manufacturing new polymers with enhanced properties. Liu et al. [31] suggested a fractional damage creep model to depict the mechanical behavior of creep, i.e., the time-dependent behavior under constant stress. Their model relied on the Caputo derivative of the VO function, which is related to the relaxation time of the rheological model. In another study, Fei et al. [32] developed a Maxwell model with VO fractional derivative to describe the creep phenomenon of a salt rock. A fractional model with a VO time derivative for an economic system was proposed and analyzed in [33]. Fan et al. [34] constructed a VO fractional model and showed its superiority over the Anand model in reporting the mechanical properties of sintered nano-silver particles. To get more insight into the applications of VO fractional calculus, the interested reader is referred to the valuable works [35,36].
Fractional diffusion equations are among the most important models of fractional calculus that are applied to describe the anomalous diffusion phenomena in complex systems. Until date, a plethora of numerical methods have been developed to handle the diffusion models with CO fractional derivatives [37,38,39,40,41,42,43,44,45,46]. However, real data has revealed that the mean square displacement scales as a function of time and/or space variables, which cannot be characterized by CO fractional diffusion models [35]. As a result, fractional diffusion equations with variable orders were introduced to deal with the aforementioned issue. It is worth mentioning that solving variable-order fractional diffusion equations analytically is highly complex and cannot be achieved in many cases due to the variable exponent. This stimulated numerical researchers to design efficient approximation methods for solving these equations, namely, finite difference methods [47,48,49], finite element methods [20,50], wavelet methods [51], collocation methods [52,53], reproducing kernel methods [54] and others.
Due to the nonlocal property of VO operators, standard numerical schemes lead to costly simulations, particularly when a high-dimensional problem or mesh refinement is considered. Explicit group difference methods are combinations of finite difference schemes and grouping strategies on standard and skewed grids. In fact, explicit group difference methods can efficiently reduce the spectral radius of the iteration matrix and increase the convergence rates of the approximation methods. Besides, these methods can be implemented on parallel computers, which makes them very advantageous in practice. In the past few years, several researchers have successfully extended the explicit group difference methods from solving integer-order differential equations to handling various types of CO fractional differential equations [55,56,57,58,59,60]. For more details, the interested reader can refer to the cited studies. However, to the best of our knowledge, there have been no trials to utilize explicit group difference methods to determine numerical solutions of differential equations in the frame of VO fractional derivatives. In this line of reasoning, this article aims to provide a grouping-based numerical method to deal with the two-dimensional VO time fractional diffusion equation (VO-TFDE) of the following form:
{C0Dα(x,y,t)tu(x,y,t)=K1∂2u(x,y,t)∂x2+K2∂2u(x,y,t)∂y2+f(x,y,t),(x,y,t)∈Ω×(0,T],(1.1)u(x,y,t)=h(x,y,t),(x,y,t)∈∂Ω×(0,T],(1.2)u(x,y,0)=g(x,y),(x,y)∈Ω∪∂Ω,(1.3) |
where Ω=[0,L]×[0,L] is a closed subset in R2 representing the solution domain, ∂Ω is its boundary, K1 and K2 are positive constants denoting the diffusion coefficients, f(x,y,t), h(x,y,t) and g(x,y) are known smooth functions and u(x,y,t) is the unknown function. 0<α(x,y,t)<1, C0Dα(x,y,t)tu(x,y,t) is the VO Caputo derivative defined as
C0Dα(x,y,t)tu(x,y,t)={1Γ(1−α(x,y,t))∫t0(t−ξ)−α(x,y,t)∂u(x,y,ξ)∂ξdξ, 0<α(x,y,t)<1,∂u(x,y,t)∂t, α(x,y,t)=1. |
The above definition of the Caputo derivative has received a great deal of interest in the field of applications since it allows the traditional initial and boundary conditions that have solid physical interpretations to be imposed in the fractional models. The aim of this study is to construct a new grouping-based numerical method, namely, the explicit decoupled group (EDG) method for the two-dimensional VO-TFDE (1.1). We start by introducing a Crank-Nicolson (C-N) finite difference method, which serves as a reference scheme to examine the efficiency of the proposed method. Thereafter, the EDG method is developed based on a new difference scheme that is derived by rotating the standard grid and applying the Taylor series expansion. The resulting method is computationally advantageous since it considers only half of the mesh points in the iteration process at each time level, which means the numerical solutions can be computed in a short time. The stability, convergence and numerical implementations of the proposed methods are also presented in this work. As far as we know, explicit group difference methods with stability and convergence analyses for VO fractional models have not appeared in the literature.
The rest of this paper is as follows: The next section presents a C-N finite difference scheme to solve the two-dimensional VO-TFDE (1.1). In section three, we propose the EDG method for the mentioned equation. The stability and convergence analyses are included in sections four and five, respectively. Some numerical experiments are reported in section six to validate the efficiency and accuracy of the presented methods. Finally, a brief conclusion is expressed in section seven.
In this section, we present a C-N numerical scheme for solving the VO-TFDE (1.1). Such a discretization scheme will reduce the considered problem into a system of simultaneous algebraic equations, which can be solved by a proper numerical technique. To this end, we let xi=iΔx, 0≤i≤M1, yj=jΔy, 0≤j≤M2, tk=kτ, 0≤k≤N, where M1, M2 and N are some positive integers. Here, Δx=L/M1, Δy=L/M2 and τ=T/N represent the spatial and temporal step sizes, respectively. For simplicity, define uki,j as the numerical approximation at the point (xi,yj,tk) and f(xi,yj,tk)=fki,j. Then, the discretization of the second order space derivatives is expressed as:
∂2u(xi,yj,tk+1/2)∂x2=12[uk+1i+1,j−2uk+1i,j+uk+1i−1,j(Δx)2+uki+1,j−2uki,j+uki−1,j(Δx)2]+O(τ2+(Δx)2)+(Δy)2), | (2.1) |
∂2u(xi,yj,tk+1/2)∂y2=12[uk+1i,j+1−2uk+1i,j+uk+1i,j−1(Δy)2+uki,j+1−2uki,j+uki,j−1(Δy)2]+O(τ2+(Δx)2)+(Δy)2). | (2.2) |
For the approximation of the Caputo time VO derivative, we utilize the following discretization scheme [61]:
C0Dαi,j,k+1/2tu(xi,yj,tk+1/2)=σi,j,k[Wi,j,k1uki,j−k−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)usi,j−Wi,j,kku0i,j+uk+1i,j−uki,j21−αi,j,k+1/2]+rk+1/2, | (2.3) |
where 0<α(x,y,t)<1, αi,j,k+1/2=α(xi,yj,tk+1/2) and
σi,j,k=1Γ(2−αi,j,k+1/2)ταi,j,k+1/2,Wi,j,ks=(s+1/2)1−αi,j,k+1/2−(s−1/2)1−αi,j,k+1/2. |
The local error term has the following bound,
|rk+1/2|≤Cτ. |
Now, setting Eqs (2.1), (2.2) and (2.3) into (1.1), we get the following,
σi,j,k[Wi,j,k1uki,j−k−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)usi,j−Wi,j,kku0i,j+uk+1i,j−uki,j21−αi,j,k+1/2]=K12[uk+1i+1,j−2uk+1i,j+uk+1i−1,j(Δx)2+uki+1,j−2uki,j+uki−1,j(Δx)2]+K22[uk+1i,j+1−2uk+1i,j+uk+1i,j−1(Δy)2+uki,j+1−2uki,j+uki,j−1(Δy)2]+fk+1/2i,j+O(τ+(Δx)2)+(Δy)2). | (2.4) |
By omitting the local error terms and replacing uki,j with its approximation Uki,j, the desired result of the C-N difference scheme is obtained as
{(Ai,j,k+2B1+2B2)Uk+1i,j=B1(Uk+1i+1,j+Uk+1i−1,j+Uki+1,j+Uki−1,j)+B2(Uk+1i,j+1+Uk+1i,j−1+Uki,j+1+Uki,j−1)+(Ai,j,k−σi,j,kWi,j,k1−2B1−2B2)Uki,j+σi,j,k∑k−1s=1(Wi,j,kk−s−Wi,j,kk−s+1)Usi,j+σi,j,kWi,j,kkU0i,j+fk+1/2i,j,1≤i≤M1−1, 1≤j≤M2−1, 0≤k≤N−1,Ai,j,k=σi,j,k21−αi,j,k+1/2, B1=K12(Δx)2, B2=K22(Δy)2.The initial and boundary conditions areU0i,j=g(xi,yj), 0≤i≤M1, 0≤j≤M2,Uki,j|∂Ω=h(xi,yj,tk), 0≤k≤N. | (2.5) |
Let
Uk=[U1,1,U1,2,…,U1,M2−1,U2,1,U2,2,…,U2,M2−1,…,UM1−1,1,UM1−1,2,…,UM1−1,M2−1]T,fk=[f1,1,f1,2,…,f1,M2−1,f2,1,f2,2,…,f2,M2−1,…,fM1−1,1,fM1−1,2,…,fM1−1,M2−1]T. |
The above difference scheme can be represented in matrix form as,
{AU1=BU0+f1/2, k=0,AUk+1=BUk+σ∑k−1m=1(Wk−mWk−m+1)Um+σWkU0+fk, k≥1. |
Here, A and B are pentadiagonal matrices given by
A=[Mi,j,k…−B1…−B2…0⋮Mi,j,k…−B1…−B2⋮−B1⋱Mi,j,k⋱⋱⋱⋮⋱⋱⋱⋱−B2−B2−B1⋱⋱⋱⋮⋮−B2⋱⋱⋱−B1⋱⋱⋱⋮0…−B2…−B1…Mi,j,k], |
B=[Ni,j,k…B1…B2…0⋮Ni,j,k…B1…B2⋮B1⋱Ni,j,k⋱⋱⋱⋮⋱⋱⋱⋱B2B2B1⋱⋱⋱⋮⋮B2⋱⋱⋱B1⋱⋱⋱⋮0…B2…B1…Ni,j,k], |
where
Mi,j,k=Ai,j,k+2B1+2B2, Ni,j,k=Ai,j,k−σi,j,kWi,j,k1−2B1−2B2. |
From the structure of the above matrices, it's evident that the matrix A is a strictly diagonally dominant matrix. Hence, we deduce that the C-N difference scheme (2.5) is uniquely solvable. In the next section, the EDG method is proposed.
The EDG method was first introduced by Abdullah [62] as an efficient and reliable Poisson solver. The success of the EDG technique in handling the Poisson equation has initiated interest in its application to solving various types of differential equations under the frame of integer-order and CO fractional derivatives. Here, we extend the formulation of the EDG method for solving the two-dimensional VO-TFDE (1.1). In this endeavor, another discretization scheme derived by rorating the standard grid 45 clockwise is necessitated to approximate Eq (1.1). Thus, the rotated C-N finite difference approximation can be written as follows,
σi,j,k[Wi,j,k1uki,j−k−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)usi,j−Wi,j,kku0i,j+uk+1i,j−uki,j21−αi,j,k+1/2]=K12[uk+1i+1,j−1−2uk+1i,j+uk+1i−1,j+12(Δx)2+uki+1,j−1−2uki,j+uki−1,j+12(Δx)2]+K22[uk+1i+1,j+1−2uk+1i,j+uk+1i−1,j−12(Δy)2+uki+1,j+1−2uki,j+uki−1,j−12(Δy)2]+fk+1/2i,j+O(τ+(Δx)2)+(Δy)2). | (3.1) |
By dropping the local error terms and replacing uki,j with its approximation Uki,j, the following expression is obtained,
(Ai,j,k+2C1+2C2)Uk+1i,j=C1(Uk+1i+1,j−1+Uk+1i−1,j+1+Uki+1,j−1+Uki−1,j+1)+C2(Uk+1i+1,j+1+Uk+1i−1,j−1+Uki+1,j+1+Uki−1,j−1)+(Ai,j,k−σi,j,kWi,j,k1−2C1−2C2)Uki,j+σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)Usi,j+σi,j,kWi,j,kkU0i,j+fk+1/2i,j, | (3.2) |
where C1=K1/(4(Δx)2) and C2=K2/(4(Δy)2).
Consider the location points (i,j), (i+1,j+1), (i+1,j) and (i,j+1). The application of Eq (3.2) to any group of these four points will lead to the (4×4) system of equation represented in the following matrix form (see [62] for extra details),
[D1−C200−C2D20000D3−C100−C1D4][Uk+1i,jUk+1i+1,j+1Uk+1i+1,jUk+1i,j+1]=[rhsi,jrhsi+1,j+1rhsi+1,jrhsi,j+1], | (3.3) |
where
D1=Ai,j,k+2C1+2C2, D2=Ai+1,j+1,k+2C1+2C2,D3=Ai+1,j,k+2C1+2C2, D4=Ai,j+1,k+2C1+2C2, |
and
rhsi,j=C1(Uk+1i+1,j−1+Uk+1i−1,j+1+Uki+1,j−1+Uki−1,j+1)+C2(Uk+1i−1,j−1+Uki+1,j+1+Uki−1,j−1)+(Ai,j,k−σi,j,kWi,j,k1−2C1−2C2)Uki,j+σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)Usi,j+σi,j,kWi,j,kkU0i,j+fk+1/2i,j,rhsi+1,j+1=C1(Uk+1i+2,j+Uk+1i,j+2+Uki+2,j+Uki,j+2)+C2(Uk+1i+2,j+2+Uki+2,j+2+Uki,j)+(Ai+1,j+1,k−σi,j,kWi+1,j+1,k1−2C1−2C2)Uki+1,j+1+σi+1,j+1,kk−1∑s=1(Wi+1,j+1,kk−s−Wi+1,j+1,kk−s+1)Usi+1,j+1+σi+1,j+1,kWi+1,j+1,kkU0i+1,j+1+fk+1/2i+1,j+1,rhsi+1,j=C1(Uk+1i+2,j−1+Uki+2,j−1+Uki,j+1)+C2(Uk+1i+2,j+1+Uk+1i,j−1+Uki+2,j+1+Uki,j−1)+(Ai+1,j,k−σi+1,j,kWi+1,j,k1−2C1−2C2)Uki+1,j+σi+1,j,kk−1∑s=1(Wi+1,j,kk−s−Wi+1,j,kk−s+1)Usi+1,j+σi+1,j,kWi+1,j,kkU0i+1,j+fk+1/2i+1,j,rhsi,j+1=C1(Uk+1i−1,j+2+Uki+1,j+Uki−1,j+2)+C2(Uk+1i+1,j+2+Uk+1i−1,j+Uki+1,j+2+Uki−1,j)+(Ai,j+1,k−σi,j+1,kWi,j+1,k1−2C1−2C2)Uki,j+1+σi,j+1,kk−1∑s=1(Wi,j+1,kk−s−Wi,j+1,kk−s+1)Usi,j+1+σi,j+1,kWi,j+1,kkU0i,j+1+fk+1/2i,j+1. |
The last (4×4) system of equations can be decoupled into two (2×2) systems whose explicit forms are as follows,
[Uk+1i,jUk+1i+1,j+1]=L1[−D2−C2−C2−D1][rhsi,jrhsi+1,j+1] | (3.4) |
and
[Uk+1i+1,jUk+1i,j+1]=L2[−D4−C1−C1−D3][rhsi+1,jrhsi,j+1], | (3.5) |
where
L1=1C22−D1D2, L2=1C21−D3D4. |
Figure 1 sheds the light on the grid shape of the EDG method. From the figure, the grid points of the discretized solution domain are branched into two distinct types, namely, black filled diamonds and white filled diamonds. One can easily verify that the implementation of Eq (3.4) involves only points of type ⧫, while the implementation of Eq (3.5) requires only points of type ◊. As a result, the computations of Eqs (3.4) and (3.5) can be performed independently.
By keeping these observations into consideration, the EDG method can be completed in two steps. First, the solution values on one type of points are evaluated iteratively by utilizing either Eq (3.4) or Eq (3.5). Second, after convergence is achieved in the first step, the solution values on the residual grid points are computed directly once using Eq (2.5). In this way, the computational cost is reduced effectively as only half of the grid points are involved in the iterative process, which ultimately reduces the execution time of the whole computation process. The application of the EDG method for solving the VO-TFDE will result in a linear system of algebraic equations. Following a similar manner in section two, it can be easily verified that the coefficient matrix is a strictly diagonally dominant matrix and, hence, the resulting system is uniquely solvable.
In this section, we analyze the stability of the proposed finite difference approximations via the von-Neumann method. The stability of numerical schemes is one of the most important issues in the realm of numerical analysis that must be addressed. We start by introducing the following lemma that will be utilized in the stability analysis,
Lemma 1. The coefficients Wi,j,ks, 0≤i≤M1−1, 0≤j≤M2−1 defined in Eqs (2.5) and (3.2) satisfy the following properties [61],
(i) Wi,j,ks≥Wi,j,ks+1, 1≤s≤k=1,2,…,N−1.
(ii) ∑k−1s=1(Wi,j,kk−s−Wi,j,kk−s+1)=Wi,j,k1−Wi,j,kk.
Let ˆUki,j and ˉUki,j represent the numerical solutions of Eqs (2.5) and (3.2), respectively, then the errors can be defined as:
ξki,j=Uki,j−ˆUki,j, 0≤i≤M1, 0≤j≤M2, 1≤k≤N, | (4.1) |
εki,j=Uki,j−ˉUki,j, 0≤i≤M1, 0≤j≤M2, 1≤k≤N. | (4.2) |
Next, by setting (4.1) into (2.5), we obtain
(Ai,j,k+2B1+2B2)ξk+1i,j−B1(ξk+1i+1,j+ξk+1i−1,j)−B2(ξk+1i,j+1+ξk+1i,j−1)=B1(ξki+1,j+ξki−1,j)+B2(ξki,j+1+ξki,j−1)+(Ai,j,k−σi,j,kWi,j,k1−2B1−2B2)ξki,j+σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)ξsi,j+σi,j,kWi,j,kkξ0i,j. | (4.3) |
Similarly, replacing (4.2) into (3.2), we have
(Ai,j,k+2C1+2C2)εk+1i,j−C1(εk+1i+1,j−1+εk+1i−1,j+1)−C2(εk+1i+1,j+1+εk+1i−1,j−1)=C1(εki+1,j−1+εki−1,j+1)+C2(εki+1,j+1+εki−1,j−1)+(Ai,j,k−σi,j,kWi,j,k1−2C1−2C2)εki,j+σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)εsi,j+σi,j,kWi,j,kkε0i,j. | (4.4) |
Then, ξk(x,y) and εk(x,y) can be expanded in the Fourier series as
ξk(x,y)=∞∑Z1=−∞∞∑Z2=−∞ηk(Z1,Z2)e2πI(Z1x/L+Z2y/L),εk(x,y)=∞∑Z1=−∞∞∑Z2=−∞μk(Z1,Z2)e2πI(Z1x/L+Z2y/L), | (4.5) |
where I=√−1 and ηk and μk are given by
ηk(Z1,Z2)=1L2∫L0∫L0ξk(x,y)e−2πI(Z1x/L+Z2y/L)dxdy, | (4.6) |
μk(Z1,Z2)=1L2∫L0∫L0εk(x,y)e−2πI(Z1x/L+Z2y/L)dxdy. | (4.7) |
Based on the definition of l2 norm and the Pareval equality, we obtain
‖ξk‖2=(∞∑Z2=−∞∞∑Z1=−∞|ηk(Z1,Z2)|2)1/2,‖εk‖2=(∞∑Z2=−∞∞∑Z1=−∞|μk(Z1,Z2)|2)1/2. | (4.8) |
Suppose that the solutions of Eqs (4.3) and (4.4) can be written as
ξki,j=ηkeI(γ1iΔx+γ2jΔy),εki,j=μkeI(γ1iΔx+γ2jΔy), | (4.9) |
in which γ1=2πZ1/L and γ2=2πZ2/L.
Substituting ξki,j=ηkeI(γ1iΔx+γ2jΔy) into Eq (16), we get
ηk+1=Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2ηk+1Ai,j,k+ρ1+ρ2[σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)ηs+σi,j,kWi,j,kkη0], | (4.10) |
where
ρ1=4B1sin2(γ1Δx2), ρ2=4B2sin2(γ2Δy2). |
Next, we prove the following result.
Lemma 2. If ηk (1≤k≤N) satisfies Eq (4.10) and 2≥31−αi,j,k, then, |ηk|≤|η0|.
Proof. We utilize mathematical induction to complete the proof. First, let k=0, and we get
|η1|=|Ai,j,k−ρ1−ρ2Ai,j,k+ρ1+ρ2||η0|. |
Since Ai,j,k, ρ1 and ρ2 are nonnegative numbers, then,
|η1|≤|η0|. |
Now, suppose that |ηm+1|≤|η0|, m=0,1,2,…,k−1.We show it holds for m=k. Using Eq (4.10) and Lemma 1, we get
|ηk+1|≤|Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2||ηk|+|1Ai,j,k+ρ1+ρ2|[σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)|ηs|+σi,j,kWi,j,kk|η0|]≤|Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2||η0|+1Ai,j,k+ρ1+ρ2[σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)|η0|+σi,j,kWi,j,kk|η0|]=|Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1|+σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2|η0|. |
If Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1>0, then
|ηk+1|≤Ai,j,k−ρ1−ρ2Ai,j,k+ρ1+ρ2|η0|<|η0|. |
If Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1<0, then
|ηk+1|≤−Ai,j,k+ρ1+ρ2+2σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2|η0|. |
Here,
|ηk+1|≤|η0|⇔−Ai,j,k+ρ1+ρ2+2σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2≤1⇔−Ai,j,k+ρ1+ρ2+2σi,j,kWi,j,k1≤Ai,j,k+ρ1+ρ2⇔2≥31−αi,j,k. |
Now, we prove the next theorem.
Theorem 1. If 2≥31−αi,j,k, then the finite difference scheme defined by Eq (2.5) is stable.
Proof. From Lemma 2 and the Parseval equality, we have
‖ξk‖22=∞∑Z1=−∞∞∑Z2=−∞|ηk(Z1,Z2)|2≤∞∑Z1=−∞∞∑Z2=−∞|η0(Z1,Z2)|2=‖ξ0‖22, |
from which we get,
‖ξk‖≤‖ξ0‖. |
Setting εki,j=μkeI(γ1iΔx+γ2jΔy) into Eq (4.4) gives
μk+1=Ai,j,k−κ1−κ2−σi,j,kWi,j,k1Ai,j,k+κ1+κ2μk+1Ai,j,k+κ1+κ2[σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)μs+σi,j,kWi,j,kkμ0], | (4.11) |
in which
κ1=4C1sin2(γ1Δx−γ2Δy2), κ2=4C2sin2(γ1Δx+γ2Δy2). |
Lemma 3. If μk (1≤k≤N) satisfies Eq (4.11) and 2≥31−αi,j,k, then, |μk|≤|μ0|.
Proof. First, putting k=0 in Eq (4.11), we obtain
|μ1|=|Ai,j,k−κ1−κ2Ai,j,k+κ1+κ2||μ0|<|μ0|. |
Next, assume that |μm+1|≤|μ0|, m=0,1,2,…,k−1. We prove it is true for m=k. With the help of Eq (4.11) and Lemma 1, we obtain
|μk+1|≤|Ai,j,k−κ1−κ2−σi,j,kWi,j,k1Ai,j,k+κ1+κ2||μk|+|1Ai,j,k+κ1+κ2|[σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)|μs|+σi,j,kWi,j,kk|μ0|]≤|Ai,j,k−κ1−κ2−σi,j,kWi,j,k1Ai,j,k+κ1+κ2||μ0|+1Ai,j,k+κ1+κ2[σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)|μ0|+σi,j,kWi,j,kk|μ0|]=|Ai,j,k−κ1−κ2−σi,j,kWi,j,k1|+σi,j,kWi,j,k1Ai,j,k+κ1+κ2|μ0|. |
If Ai,j,k−κ1−κ2−σi,j,kWi,j,k1>0, then
|μk+1|≤Ai,j,k−κ1−κ2Ai,j,k+κ1+κ2|μ0|<|μ0|. |
If Ai,j,k−κ1−κ2−σi,j,kWi,j,k1<0, then
|μk+1|≤−Ai,j,k+κ1+κ2+2σi,j,kWi,j,k1Ai,j,k+κ1+κ2|μ0|. |
Here,
|μk+1|≤|μ0|⇔−Ai,j,k+κ1+κ2+2σi,j,kWi,j,k1Ai,j,k+κ1+κ2≤1⇔−Ai,j,k+κ1+κ2+2σi,j,kWi,j,k1≤Ai,j,k+κ1+κ2⇔2≥31−αi,j,k. |
Now, Lemma 2 is employed to prove the next theorem.
Theorem 2. If 2≥31−αi,j,k, then, the finite difference scheme defined by Eq (3.2) is stable.
Proof. From Lemma 3 and the Parseval equality, we have
‖εk‖22=∞∑Z1=−∞∞∑Z2=−∞|μk(Z1,Z2)|2≤∞∑Z1=−∞∞∑Z2=−∞|μ0(Z1,Z2)|2=‖ε0‖22, |
which gives the desired result
‖εk‖≤‖ε0‖. |
This section is devoted to analyze the convergence of the proposed methods. Let ζki,j be the difference between the exact and numerical solutions at the point (xi,yj,tk), then the error can be written as,
ζki,j=uki,j−Uki,j, 1≤i≤M1−1, 1≤j≤M2−1, 1≤k≤N. | (5.1) |
By setting Eq (5.1) into Eq (2.5), the error equation follows immediately as
(Ai,j,k+2B1+2B2)ζk+1i,j−B1(ζk+1i+1,j+ζk+1i−1,j)−B2(ζk+1i,j+1+ζk+1i,j−1)=B1(ζki+1,j+ζki−1,j)+B2(ζki,j+1+ζki,j−1)+(Ai,j,k−σi,j,kWi,j,k1−2B1−2B2)ζki,j+σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)ζsi,j+σi,j,kWi,j,kkζ0i,j+Rk+1/2i,j, | (5.2) |
where Rk+1/2i,j denotes the truncation error at the point (xi,yj,tk+1/2). From now on, we suppose that C is a constant that may take different values at different locations. From Eq (2.4), we have
|Rk+1/2i,j|≤C(τ+(Δx)2+(Δy)2), 1≤i≤M1−1, 1≤j≤M2−1, 0≤k≤N, | (5.3) |
where
C=max1≤i≤M1−1,1≤j≤M2−1,0≤k≤N{Cki,j}. |
Next, ζk(x,y) and Rk+1/2(x,y) can be expanded in the Fourier series as
ζk(x,y)=∞∑Z2=−∞∞∑Z1=−∞ψk(Z1,Z2)e2πI(Z1x/L+Z2y/L), |
Rk+1/2(x,y)=∞∑Z2=−∞∞∑Z1=−∞ϕk(Z1,Z2)e2πI(Z1x/L+Z2y/L), |
such that
ψk(Z1,Z2)=1L2∫L0∫L0ζk(x,y)e−2πI(Z1x/L+Z2y/L)dxdy, |
ϕk(Z1,Z2)=1L2∫L0∫L0Rk+1/2(x,y)e−2πI(Z1x/L+Z2y/L)dxdy. |
Considering the definition of l2 norm and the Parseval equality, we get
‖ζk‖2=(M2−1∑j=1M1−1∑i=1ΔyΔx|ζki,j|2)1/2=(∞∑Z1=−∞∞∑Z2=−∞|ψk(Z1,Z2)|2)1/2, | (5.4) |
‖Rk+1/2‖2=(M2−1∑j=1M1−1∑i=1ΔyΔx|Rk+1/2i,j|2)1/2=(∞∑Z1=−∞∞∑Z2=−∞|ϕk(Z1,Z2)|2)1/2. | (5.5) |
Suppose the solutions of Eq (5.2) are as follows:
ζki,j=ψkeI(γ1iΔx+γ2jΔy), Rk+1/2i,j=ϕk+1/2eI(γ1iΔx+γ2jΔy). | (5.6) |
Substitute Eq (5.6) into Eq (5.2) and simplify to emerge with the following expression,
ψk+1=Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2ψk+1Ai,j,k+ρ1+ρ2[σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)ψs+σi,j,kWi,j,kkψ0+ϕk+1/2], | (5.7) |
where ρ1 and ρ2 are as defined in Eq (4.10).
Lemma 4. If ψk(1≤k≤N) satisfies Eq (5.7), then, we have
|ψk+1|≤C(k+1)τ|ϕ1/2|. |
Proof. From Eq (5.5), there is a positive constant C such that
|ϕk+1/2|≤Cτ|ϕ1/2|, 0≤k≤N−1. |
We complete the proof by mathematical induction. First, setting k=0 in Eq (5.7) and noticing that ψ0=0, we obtain
|ψ1|=1Ai,j,k+ρ1+ρ2|ϕ1/2|≤|ϕ1/2|≤Cτ|ϕ1/2|. |
Next, assume that
|ψm+1|≤C(m+1)τ|ϕ1/2|, m=0,1,2,…,k−1. |
Putting m=k in Eq (5.7) and using Lemma 1, we get
|ψk+1|≤|Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2||ψk|+|1Ai,j,k+ρ1+ρ2|[σi,j,kk−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)|ψs|+σi,j,kWi,j,kk|ψ0|+|ϕk+1/2|]≤|Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1|Ai,j,k+ρ1+ρ2Ckτ|ϕ1/2|+1Ai,j,k+ρ1+ρ2[σi,j,kCkτ|ϕ1/2|(k−1∑s=1(Wi,j,kk−s−Wi,j,kk−s+1)+Wi,j,kk)+Cτ|ϕ1/2|]=|Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1|+σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2Ckτ|ϕ1/2|+1Ai,j,k+ρ1+ρ2Cτ|ϕ1/2|=[|Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1|+σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2k+1Ai,j,k+ρ1+ρ2]Cτ|ϕ1/2|. |
If Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1>0, then
|ψk+1|≤[Ai,j,k−ρ1−ρ2Ai,j,k+ρ1+ρ2k+1Ai,j,k+ρ1+ρ2]Cτ|ϕ1/2|≤C(k+1)τ|ϕ1/2|. |
If Ai,j,k−ρ1−ρ2−σi,j,kWi,j,k1<0, then
|ψk+1|≤[−Ai,j,k+ρ1+ρ2+2σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2k+1Ai,j,k+ρ1+ρ2]Cτ|ϕ1/2|. |
Here,
|ψk+1|≤C(k+1)τ|ϕ1/2|⇔−Ai,j,k+ρ1+ρ2+2σi,j,kWi,j,k1Ai,j,k+ρ1+ρ2≤1⇔−Ai,j,k+ρ1+ρ2+2σi,j,kWi,j,k1≤Ai,j,k+ρ1+ρ2⇔2≥31−αi,j,k. |
Therefore, under the above condition we have
|ψk+1|≤C(k+1)τ|ϕ1/2|. |
We utilize Lemma 4 to prove the next result.
Theorem 3. If 2≥31−αi,j,k, then the finite difference scheme defined by Eq (2.5) is l2 convergent with convergence order O(τ+(Δx)2+(Δy)2).
Proof. With the help of Lemma 4 and Eqs (5.4) and (5.5), we obtain
‖ζk‖22=∞∑Z1=−∞∞∑Z2=−∞|ψk(Z1,Z2)|2≤∞∑Z1=−∞∞∑Z2=−∞C2(k+1)2τ2|ϕ1/2(Z1,Z2)|2=C2(k+1)2τ2‖R1/2‖22. |
Taking Eq (5.3) into consideration leads to
‖ζk‖2≤C(k+1)τ‖R1/2‖2≤C(k+1)τ(τ+(Δx)2+(Δy)2)≤C∗(τ+(Δx)2+(Δy)2), |
where C∗=CT as (k+1)τ≤T.
Theorem 4. If 2≥31−αi,j,k, then the finite difference scheme defined by Eq (3.2) is l2 convergent with convergence order O(τ+(Δx)2+(Δy)2).
Proof. The proof can be established in analogous way to Theorem 3.
Theorems 5.1 and 5.2 indicate the convergence of the numerical schemes (2.5) and (3.2), respectively. This means that as the spatial and temporal step sizes approach zero, the numerical solutions approach to the exact solution of the considered problem. These considerations will be verified in the next section.
After the construction and analysis of the C-N and EDG solution methods in the previous sections, we turn our attention to simulating the two-dimensional VO-TFDE (1.1) using the aforementioned numerical schemes. In this endeavor, three numerical examples are chosen to illustrate the capability, accuracy and efficiency of the presented methods. The exact analytical solution of each example is known and is utilized for comparison with the evaluated numerical solutions. Unless otherwise specified, the solution domain is set to be [0,1]×[0,1]×(0,1]. Furthermore, we assume that Δx=Δy=h throughout numerical experiments. All computations are carried out using the Julia Programming Language and run on a PC with the configuration: Intel(R) Core(TM) i7-8550U CPU 2.00 GHz 8 GB RAM and the Windows 10 (64-bit) operating system. Matlab is also employed for generating some figures. The computational cost and the accuracy of the proposed methods will be reported based on the criteria of CPU time (in seconds) and the maximum absolute error (MAE), respectively. The numerical outcomes of these criteria are attained for different values of mesh sizes and various VO functions. Finally, we include some graphical representations to facilitate the discussion of our results.
The MAE between the exact solution and the numerical solution is given by
MAE=max1≤i≤M1−1,1≤j≤M2−1,1≤k≤Nu(xi,yj,tk)−Uki,j. |
To compute the convergence order (CO), we use
CO=ln(MAE(τ,h1)/MAE(τ,h2))ln(h1/h2). |
Example 1. Consider the following two-dimensional VO-TFDE:
C0Dα(x,y,t)tu(x,y,t)=∂2u(x,y,t)∂x2+∂2u(x,y,t)∂y2+6t3−a(x,y,t)Γ(4−a(x,y,t))(1−x2)2(1−y2)2−4t3((1−y2)2(3x2−1)+(1−x2)2(3y2−1)) |
with the initial and boundary conditions
u(x,y,0)=0,u(x,0,t)=t3(1−x2)2,u(x,1,t)=t3(1−x2)2,u(0,y,t)=t3(1−y2)2,u(1,y,t)=t3(1−y2)2, |
and the exact analytical solution is given by
u(x,y,t)=t3(1−x2)2(1−y2)2. |
Tables 1 gives the CPU times and the numerical errors of the proposed methods for a variety of mesh sizes (1/6, 1/12, 1/18, 1/24) with fixed τ=0.01 and various values of α(x,y,t). In view of this table, it can be seen that the C-N and EDG methods produce accurate results in solving Example 1. Furthermore, as the mesh size increases, the observed maximum errors for all chosen VO derivatives α(x,y,t) decrease. Figure 2 sketches the CPU time of the C-N and EDG methods according to different sizes of space steps with α(x,y,t)=10−(xyt)380 and α(x,y,t)=9−x5+t3120. We can see that the EDG method significantly outperforms the C-N method in terms of CPU timing. The time reduction percentage of the EDG method compared to the C-N method is (52.74-79.44)%. For instance, Table 1 shows that we need to take h=1/24 in order to obtain the maximum error of 2.4740E-4 using the C-N method with α(x,y,t)=10−(xyt)380, which costs 44.8959 seconds. On the other hand, the EDG method with h=1/24 achieves a maximum error of 3.4746E-4 and the corresponding computing time is only 11.4655 seconds. These results and comparisons justify our effort to establish the EDG method. Figure 3 shows a comparison of the numerical solutions of the proposed methods and the exact solution of Example 1 for h=1/24, τ=0.01, x=0.5, α(x,y,t)=11−cos(xt)211 and T=0.25,0.5,0.75,1. One can observe from this figure that the behavior of the numerical solutions is the same as that of the exact solution. Figure 4 displays 3D surface plots of the numerical solutions and the maximum errors for the C-N and EDG methods at h=1/24, τ=0.01 and α(x,y,t)=15+sin(yt)516. It follows that the numerical solutions of the C-N and EDG methods are in good agreement with the exact solutions, which illustrate the accuracy and effectiveness of the proposed methods.
C-N method | EDG method | |||||
α(x,y,t) | h | CPU time | MAE | CPU time | MAE | |
10−(xyt)380 | 1/6 | 0.2499 | 4.1038E-03 | 0.0937 | 7.0379E-03 | |
1/12 | 3.7891 | 1.0223E-03 | 0.8840 | 1.6246E-03 | ||
1/18 | 16.3650 | 3.3934E-04 | 4.0260 | 6.7900E-04 | ||
1/24 | 44.8959 | 2.4740E-04 | 11.4655 | 3.4746E-04 | ||
5yt−3 | 1/6 | 0.4966 | 4.1265E-03 | 0.1284 | 7.0739E-03 | |
1/12 | 4.9944 | 1.0522E-03 | 1.2802 | 1.6364E-03 | ||
1/18 | 18.8372 | 3.6115E-04 | 5.0190 | 6.8719E-04 | ||
1/ 24 | 51.9460 | 2.5265E-04 | 13.7734 | 3.4547E-04 | ||
11−cos(xt)211 | 1/6 | 0.1588 | 3.7833E-03 | 0.0630 | 6.5721E-03 | |
1/12 | 1.7410 | 9.5964E-04 | 0.4935 | 1.5089E-03 | ||
1/18 | 5.9911 | 3.3693E-04 | 1.6087 | 6.3153E-04 | ||
1/24 | 17.5482 | 1.8718E-04 | 4.5734 | 3.0326E-04 | ||
15+sin(yt)516 | 1/6 | 0.2364 | 3.7776E-03 | 0.1117 | 6.5633E-03 | |
1/12 | 2.5489 | 9.5797E-04 | 0.8056 | 1.5115E-03 | ||
1/18 | 12.0167 | 3.3706E-04 | 3.0428 | 6.4030E-04 | ||
1/24 | 39.0032 | 1.8076E-04 | 8.0181 | 3.0591E-04 | ||
9−x5+t3120 | 1/6 | 0.3455 | 4.1132E-03 | 0.0837 | 7.0528E-03 | |
1/12 | 3.6123 | 1.0282E-03 | 1.1557 | 1.6213E-03 | ||
1/18 | 15.4070 | 3.5013E-04 | 4.1639 | 7.0226E-04 | ||
1/24 | 43.8414 | 2.5438E-04 | 13.8434 | 3.3037E-04 | ||
15−sin(xyt)416 | 1/6 | 0.3609 | 3.7780E-03 | 0.1149 | 6.5638E-03 | |
1/12 | 2.9381 | 9.5677E-04 | 1.0479 | 1.5113E-03 | ||
1/18 | 11.8527 | 3.3380E-04 | 2.9980 | 6.3989E-04 | ||
1/24 | 31.7521 | 1.8077E-04 | 8.2564 | 3.0450E-04 |
Example 2. In this example, we consider the following two-dimensional VO-TFDE:
C0Dα(x,y,t)tu(x,y,t)=∂2u(x,y,t)∂x2+∂2u(x,y,t)∂y2+2t2−a(x,y,t)Γ(3−a(x,y,t))sin(x2+y2)+2t2(2x2sin(x2+y2)+2y2sin(x2+y2)−2cos(x2+y2)), |
subject to the initial and boundary conditions
u(x,y,0)=0,u(x,0,t)=t2sin(x2),u(x,1,t)=t2sin(x2+1),u(0,y,t)=t2sin(y2),u(1,y,t)=t2sin(y2+1), |
with the exact analytical solution written as
u(x,y,t)=t2sin(x2+y2). |
We now utilize the C-N and EDG methods to solve the given problem in Example 2. The CPU times and the maximum errors for several mesh sizes (1/4, 1/10, 1/16, 1/22) with fixed τ=0.01 and various values of α(x,y,t) are shown in Table 2. Figure 5 depicts the CPU times of the proposed methods against different values of spatial step sizes for α(x,y,t)=exyt+cos(xyt)30 and α(x,y,t)=5+(xy)3−(xt)450. From the data in these tables and figures, it's seen that the C-N and EDG methods generate comparable results in terms of accuracy, but the latter method results in better simulations in terms of CPU timing. From Table 2, the improvement percentage in CPU time of the EDG method compared to the C-N method is (40.36-83.56)%. This example clearly lends itself to a comparison similar to that of Example 1. The numerical solutions of the proposed methods along with the exact solution of Example 2 is portrayed in Figure 6 for h=1/22, τ=0.01, x=0.5, α(x,y,t)=16−exyt17 and T=0.25,0.5,0.75,1. Figure 7 highlights the 3D surface plots of the numerical solutions and the maximum errors of the C-N and EDG methods at h=1/22, τ=0.01 and α(x,y,t)=3+(xy)2−(xt)330. These figures demonstrate the effectiveness of the proposed methods as the numerical solutions are found to match well with the exact analytical solution.
C-N method | EDG method | |||||
α(x,y,t) | h | CPU time | MAE | CPU time | MAE | |
1−(xyt)3+cos(xyt)210 | 1/4 | 0.0876 | 1.0398E-02 | 0.0409 | 2.2604E-02 | |
1/10 | 3.2858 | 1.7702E-03 | 0.8188 | 3.3201E-03 | ||
1/16 | 18.4236 | 7.9260E-04 | 4.9788 | 1.3317E-03 | ||
1/22 | 63.9827 | 5.9961E-04 | 13.8950 | 7.6442E-04 | ||
exyt+cos(xyt)30 | 1/4 | 0.2551 | 1.0465E-02 | 0.0497 | 2.2765E-02 | |
1/10 | 5.3375 | 1.7854E-03 | 1.2344 | 3.3386E-03 | ||
1/16 | 25.7313 | 8.1088E-04 | 6.0432 | 1.3401E-03 | ||
1/22 | 72.5582 | 6.1617E-04 | 22.3957 | 7.9394E-04 | ||
16−exyt17 | 1/4 | 0.0493 | 1.0000E-02 | 0.0294 | 2.1580E-02 | |
1/10 | 0.9723 | 1.7134E-03 | 0.3137 | 3.1995E-03 | ||
1/16 | 5.8175 | 7.8225E-04 | 1.4358 | 1.2959E-03 | ||
1/22 | 17.8491 | 5.7985E-04 | 4.3844 | 7.4330E-04 | ||
3+(xy)2−(xt)330 | 1/4 | 0.0642 | 1.0449E-02 | 0.0227 | 2.2735E-02 | |
1/10 | 2.3931 | 1.7715E-03 | 0.8921 | 3.3315E-03 | ||
1/16 | 14.1170 | 8.2971E-04 | 2.8375 | 1.3347E-03 | ||
1/22 | 55.0142 | 6.1130E-04 | 9.0424 | 7.7265E-04 | ||
5+(xy)3−(xt)450 | 1/4 | 0.0682 | 1.0448E-02 | 0.0291 | 2.2733E-02 | |
1/10 | 2.7203 | 1.7685E-03 | 0.7045 | 3.3373E-03 | ||
1/16 | 15.1587 | 8.2983E-04 | 3.7406 | 1.3408E-03 | ||
1/22 | 50.2882 | 5.8791E-04 | 11.7378 | 7.7962E-04 | ||
1−xyt+sin(xyt)10 | 1/4 | 0.0845 | 1.0447E-02 | 0.0325 | 2.2731E-02 | |
1/10 | 2.9924 | 1.7737E-03 | 0.7802 | 3.3370E-03 | ||
1/16 | 17.1431 | 8.0822E-04 | 4.2529 | 1.3406E-03 | ||
1/22 | 54.7490 | 5.8871E-04 | 13.4222 | 7.7789E-04 |
Example 3. Here, the following two-dimensional VO-TFDE is considered:
C0Dα(x,y,t)tu(x,y,t)=∂2u(x,y,t)∂x2+∂2u(x,y,t)∂y2+(120t5−a(x,y,t)Γ(6−a(x,y,t))+π2t5)(sin(πx)+sin(πy))2, |
with the initial and boundary conditions
u(x,y,0)=0,u(x,0,t)=12t5sin(πx),u(x,1,t)=12t5sin(πx),u(0,y,t)=12t5sin(πy),u(1,y,t)=12t5sin(πy), |
and the exact analytical solution of the following form:
u(x,y,t)=12t5(sin(πx)+sin(πy)). |
Table 3 highlights the numerical results of solving Example 3 using CN and EDG methods with T=1, τ=0.01, h−1=12,24 and different VO functions. From the data in this table, it can be seen that the numerical solutions converge to the exact solution where maximum errors decrease as mesh sizes increase. A reduction of CPU timing and, hence, computational burden is also noted in using the EDG method over the CN method for solving the considered problem. This can be beneficial for capturing the long-term behavior of physical phenomena that is governed by VO fractional models. From the previous discussion, it follows that the proposed methods are accurate and the EDG method is superior to the C-N method in terms of CPU timing. Finally, Table 4 highlights the computational orders of the proposed methods for solving Examples 1–3. From this table, the second order convergence of the proposed methods can be anticipated, which is in line with the theoretical considerations.
h−1=12 | h−1=24 | ||||||||||
C-N | EDG | CN | EDG | ||||||||
α(x,y,t) | CPU time | MAE | CPU time | MAE | CPU time | MAE | CPU time | MAE | |||
10−(xyt)380 | 4.2446 | 2.9811E-03 | 0.9564 | 3.0869E-03 | 57.7629 | 3.7745E-04 | 11.7092 | 5.3877E-04 | |||
5yt−3 | 5.2871 | 3.0868E-03 | 1.2871 | 3.1129E-03 | 69.0271 | 5.4651E-04 | 16.3071 | 6.8800E-04 | |||
11−cos(xt)211 | 1.4539 | 2.4022E-03 | 0.4342 | 2.4534E-03 | 16.7202 | 3.2210E-04 | 5.8109 | 3.1297E-04 | |||
15+sin(yt)516 | 2.6672 | 2.4036E-03 | 0.8971 | 2.4604E-03 | 44.4374 | 3.1177E-04 | 8.8679 | 3.1733E-04 | |||
9−x5+t3120 | 4.0845 | 3.0183E-03 | 1.3774 | 3.1207E-03 | 64.1056 | 4.3822E-04 | 14.0114 | 5.7422E-04 | |||
15−sin(xyt)416 | 2.7664 | 2.3978E-03 | 9.95E-01 | 2.4571E-03 | 36.6685 | 3.1184E-04 | 10.6867 | 3.0756E-04 | |||
1−(xyt)3+cos(xyt)210 | 6.5047 | 2.9637E-03 | 1.931 | 3.0515E-03 | 96.8315 | 3.8151E-04 | 24.4635 | 5.2750E-04 | |||
exyt+cos(xyt)30 | 8.3275 | 3.0230E-03 | 2.3265 | 3.1152E-03 | 125.3629 | 4.5041E-04 | 33.0581 | 5.6177E-04 | |||
16−exyt17 | 1.8139 | 2.3908E-03 | 0.5174 | 2.4595E-03 | 28.0809 | 3.1432E-04 | 6.7689 | 2.5614E-04 | |||
3+(xy)2−(xt)330 | 4.2048 | 3.0191E-03 | 1.0323 | 3.0820E-03 | 69.2202 | 4.1416E-04 | 14.2108 | 5.7308E-04 | |||
5+(xy)3−(xt)450 | 5.9701 | 3.0114E-03 | 1.5368 | 3.0890E-03 | 86.0504 | 4.0849E-04 | 20.5851 | 5.6393E-04 | |||
1−xyt+sin(xyt)10 | 6.4259 | 3.0110E-03 | 1.4328 | 3.0912E-03 | 74.6506 | 4.0501E-04 | 18.0687 | 5.6381E-04 |
C-N | EDG | ||||||
Example | α(x,y,t) | h | MAE | CO | MAE | CO | |
1 | 10−(xyt)380 | 1/4 | 9.0598E-03 | 1.8596E-02 | |||
1/8 | 2.4085E-03 | 1.91 | 3.7500E-03 | 2.31 | |||
1/16 | 5.0287E-04 | 2.25 | 8.7401E-04 | 2.10 | |||
2 | 16−exyt17 | 1/4 | 1.0000E-02 | 2.1580E-02 | |||
1/8 | 2.6132E-03 | 1.93 | 5.0291E-03 | 2.10 | |||
1/16 | 7.8225E-04 | 1.74 | 1.2959E-03 | 1.95 | |||
3 | 11−cos(xt)211 | 1/4 | 2.3825E-02 | 2.6705E-02 | |||
1/8 | 5.8325E-03 | 2.03 | 6.0067E-03 | 2.15 | |||
1/16 | 1.1573E-03 | 2.33 | 1.2173E-03 | 2.30 |
In the previous subsection, three numerical examples were solved using the proposed methods, and the results were illustrated in Tables 1–3 along with Figures 2–7. In view of the said tables, it can be seen that the numerical solutions of the proposed methods approach the exact solution as the spatial step size decreases. This is in line with our theoretical analysis. In addition, the EDG method was observed to be computationally efficient compared to the C-N method in solving the test problems, where the former required less CPU time compared to the latter. Such reduction in computational burden is attributed to the smaller number of mesh points comprised in the iterative process as discussed earlier. The experimental results are in line with the considerations of previous sections and demonstrate the feasibility, accuracy and efficiency of the proposed methods.
The VO fractional diffusion equation and its numerical solution are of practical importance for describing complex diffusion processes in which the system memory changes as a function of time and/or space variables. In this paper, the two-dimensional VO-TFDE (1.1) has been considered to account for its numerical solution. A C-N difference scheme along with an EDG method constructed based on a skewed finite difference approximation on the rotated grid have been developed for solving the VO fractional problem. The stability and convergence analyses have been investigated in detail. Numerical test problems were provided to illustrate the capability and performance of the proposed methods.
Based on the numerical simulations and the discussion of experimental results, some conclusions are drawn.
First, both C-N and EDG methods perform well for the VO-TFDE as their numerical solutions are found to be in good agreement with the exact analytical solutions. See Figures 3, 4, 6 and 7.
Second, the proposed methods are stable and convergent, and the maximum absolute errors are observed to decrease as the mesh sizes increase. See Tables 1 and 2.
Third, the EDG method reduces the computational burden of solving the VO-TFDE and results in faster simulations compared to the C-N difference method. See Figures 2 and 5.
Finally, the current work has successfully developed the C-N and EDG numerical methods to solve the fractional diffusion equation involving the VO time derivative defined in section one. The extension of the proposed methods to other classes of linear and nonlinear fractional differential equations with time-invariant type VO derivative [63] is an interesting line of future research. The construction of the proposed methods based on higher-order difference approximations for enhancing the accuracy is another direction for further study.
The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.
The authors would like to acknowledge the support provided by the Deanship of Research Oversight and Coordination (DROC) at King Fahd University of Petroleum & Minerals (KFUPM) for funding this work through project (No. EC221011). We are thankful for the anonymous reviewers for their valuable comments and suggestions.
The authors declare no conflicts of interest.
[1] |
O. Nave, Analysis of the two-dimensional polydisperse liquid sprays in a laminar boundary layer flow using the similarity transformation method, Adv. Model. and Simul. in Eng. Sci., 2 (2015), 20. https://doi.org/10.1186/s40323-015-0042-8 doi: 10.1186/s40323-015-0042-8
![]() |
[2] |
A. Lozynskyy, A. Chaban, T. Perzynski, A. Szafraniec, L. Kasha, Application of fractional-order calculus to improve the mathematical model of a two-mass system with a long shaft, Energies, 14 (2021), 1854. https://doi.org/10.3390/en14071854 doi: 10.3390/en14071854
![]() |
[3] |
M. I. Asjad, R. Ali, A. Iqbal, T. Muhammad, Y. M. Chu, Application of water based drilling clay-nanoparticles in heat transfer of fractional maxwell fluid over an infinite flat surface, Sci. Rep., 11 (2021), 18833. https://doi.org/10.1038/s41598-021-98066-w doi: 10.1038/s41598-021-98066-w
![]() |
[4] |
D. Baleanu, A. Fernandez, A. Akgl, On a fractional operator combining proportional and classical differintegrals, Mathematics, 8 (2020), 360. https://doi.org/10.3390/math8030360 doi: 10.3390/math8030360
![]() |
[5] |
G. Liu, S. Li, J. Wang, New green-ampt model based on fractional derivative and its application in 3d slope stability analysis, J. Hydrol., 603 (2021), 127084. https://doi.org/10.1016/j.jhydrol.2021.127084 doi: 10.1016/j.jhydrol.2021.127084
![]() |
[6] |
P. Kumar, V. S. Erturk, R. Banerjee, M. Yavuz, V. Govindaraj, Fractional modeling of plankton-oxygen dynamics under climate change by the application of a recent numerical algorithm, Phys. Scr., 96 (2021), 124044. https://doi.org/10.1088/1402-4896/ac2da7 doi: 10.1088/1402-4896/ac2da7
![]() |
[7] |
M. Inc, B. Acay, H. W. Berhe, A. Yusuf, A. Khan, S. W. Yao, Analysis of novel fractional COVID-19 model with real-life data application, Results Phys., 23 (2021), 103968. https://doi.org/10.1016/j.rinp.2021.103968 doi: 10.1016/j.rinp.2021.103968
![]() |
[8] |
W. Y. Shen, Y. M. Chu, M. ur Rahman, I. Mahariq, A. Zeb, Mathematical analysis of hbv and hcv co-infection model under nonsingular fractional order derivative, Results Phys., 28 (2021), 104582. https://doi.org/10.1016/j.rinp.2021.104582 doi: 10.1016/j.rinp.2021.104582
![]() |
[9] | I. Podlubny, Fractional differential equations: an introduction to fractional derivatives, fractional differential equations, to methods of their solution and some of their applications, New York: Academic Press, 1999. |
[10] | A. A. Kilbas, H. M. Srivastava, J. J. Trujillo, Theory and applications of fractional differential equations, Amsterdam: Elsevier Science, 2006. |
[11] | C. Li, F. Zeng, Numerical methods for fractional calculus, Boca Raton: CRC Press, 2015. |
[12] |
G. S. Teodoro, J. A. T. Machado, E. C. de Oliveira, A review of definitions of fractional derivatives and other operators, J. Comput. Phys., 388 (2019), 195–208. https://doi.org/10.1016/j.jcp.2019.03.008 doi: 10.1016/j.jcp.2019.03.008
![]() |
[13] |
M. D. Ortigueira, J. A. T. Machado, What is a fractional derivative, J. Comput. Phys., 293 (2015), 4–13. https://doi.org/10.1016/j.jcp.2014.07.019 doi: 10.1016/j.jcp.2014.07.019
![]() |
[14] |
V. E. Tarasov, No nonlocality. no fractional derivative, Commun. Nonlinear Sci., 62 (2018), 157–163. https://doi.org/10.1016/j.cnsns.2018.02.019 doi: 10.1016/j.cnsns.2018.02.019
![]() |
[15] |
H. G. Sun, W. Chen, H. Wei, Y. Q. Chen, A comparative study of constant-order and variable-order fractional models in characterizing memory property of systems, Eur. Phys. J. Spec. Top., 193 (2011), 185–192. https://doi.org/10.1140/epjst/e2011-01390-6 doi: 10.1140/epjst/e2011-01390-6
![]() |
[16] |
S. G. Samko, B. Ross, Integration and differentiation to a variable fractional order, Integr. Transf. Spec. F., 1 (1993), 277–300. https://doi.org/10.1080/10652469308819027 doi: 10.1080/10652469308819027
![]() |
[17] |
C. F. Lorenzo, T. T. Hartley, Initialization, conceptualization, and application in the generalized (fractional) calculus, Crit. Rev. Biomed. Eng., 35 (2007), 447–553. https://doi.org/10.1615/CritRevBiomedEng.v35.i6.10 doi: 10.1615/CritRevBiomedEng.v35.i6.10
![]() |
[18] |
C. F. Lorenzo, T. T. Hartley, Variable order and distributed order fractional operators, Nonlinear Dynamics, 29 (2002), 57–98. https://doi.org/10.1023/A:1016586905654 doi: 10.1023/A:1016586905654
![]() |
[19] |
C. F. Coimbra, Mechanics with variable-order differential operators, Ann. Phys., 515 (2003), 692–703. https://doi.org/10.1002/andp.200351511-1203 doi: 10.1002/andp.200351511-1203
![]() |
[20] |
X. Zheng, H. Wang, Optimal-order error estimates of finite element approximations to variable-order time-fractional diffusion equations without regularity assumptions of the true solution, IMA J. Numer. Anal., 41 (2021), 1522–1545. https://doi.org/10.1093/imanum/draa013 doi: 10.1093/imanum/draa013
![]() |
[21] |
P. Pandey, J. F. Gomez-Aguilar, On solution of a class of nonlinear variable order fractional reaction-diffusion equation with mittag-effler kernel, Numer. Meth. Part. D. E., 37 (2021), 998–1011. https://doi.org/10.1002/num.22563 doi: 10.1002/num.22563
![]() |
[22] |
M. Hosseininia, M. H. Heydari, F. M. M. Ghaini, A numerical method for variable-order fractional version of the coupled 2d burgers equations by the 2D chelyshkov polynomials, Math. Method. Appl. Sci., 44 (2021), 6482–6499. https://doi.org/10.1002/mma.7199 doi: 10.1002/mma.7199
![]() |
[23] |
H. Hassani, Z. Avazzadeh, J. A. T. Machado, Numerical approach for solving variable-order space-time fractional telegraph equation using transcendental Bernstein series, Engineering with Computers, 36 (2020), 867–878. https://doi.org/10.1007/s00366-019-00736-x doi: 10.1007/s00366-019-00736-x
![]() |
[24] |
K. Sadri, H. Aminikhah, An efficient numerical method for solving a class of variable-order fractional mobile-immobile advection-dispersion equations and its convergence analysis, Chaos Soliton. Fract., 146 (2021), 110896. https://doi.org/10.1016/j.chaos.2021.110896 doi: 10.1016/j.chaos.2021.110896
![]() |
[25] |
M. Hosseininia, M. H. Heydari, C. Cattani, A wavelet method for non-linear variable-order time fractional 2D Schrodinger equation, Discrete Cont. Dyn.-S, 14 (2021), 2273–2295. https://doi.org/10.3934/dcdss.2020295 doi: 10.3934/dcdss.2020295
![]() |
[26] |
M. D. Ortigueira, D. Valerio, J. T. Machado, Variable order fractional systems, Commun. Nonlinear Sci., 71 (2019), 231–243. https://doi.org/10.1016/j.cnsns.2018.12.003 doi: 10.1016/j.cnsns.2018.12.003
![]() |
[27] |
H. Wang, X. Zheng, Wellposedness and regularity of the variable-order time-fractional diffusion equations, J. Math. Anal. Appl., 475 (2019), 1778–1802. https://doi.org/10.1016/j.jmaa.2019.03.052 doi: 10.1016/j.jmaa.2019.03.052
![]() |
[28] | R. Almeida, D. Tavares, D. F. M. Torres, The variable-order fractional calculus of variations, Heidelberg: Springer, 2019. https://doi.org/10.1007/978-3-319-94006-9 |
[29] |
N. H. Sweilam, S. M. Al-Mekhlafi, A. O. Albalawi, J. A. T. Machado, Optimal control of variable-order fractional model for delay cancer treatments, Appl. Math. Model., 89 (2021), 1557–1574. https://doi.org/10.1016/j.apm.2020.08.012 doi: 10.1016/j.apm.2020.08.012
![]() |
[30] |
G. Xiang, D. Yin, R. Meng, C. Cao, Predictive model for stress relaxation behavior of glassy polymers based on variable-order fractional calculus, Polym. Advan. Technol., 32 (2021), 703–713. https://doi.org/10.1002/pat.5123 doi: 10.1002/pat.5123
![]() |
[31] |
X. Liu, D. Li, C. Han, Y. Shao, A Caputo variable-order fractional damage creep model for sandstone considering effect of relaxation time, Acta Geotech., 17 (2022), 153–167. https://doi.org/10.1007/s11440-021-01230-9 doi: 10.1007/s11440-021-01230-9
![]() |
[32] |
W. Fei, L. Jie, Z. Quanle, L. Cunbao, C. Jie, G. Renbo, A triaxial creep model for salt rocks based on variable-order fractional derivative, Mech. Time-Depend. Mater., 25 (2021), 101–118. https://doi.org/10.1007/s11043-020-09470-0 doi: 10.1007/s11043-020-09470-0
![]() |
[33] |
H. Jahanshahi, S. S. Sajjadi, S. Bekiros, A. A. Aly, On the development of variable-order fractional hyperchaotic economic system with a nonlinear model predictive controller, Chaos Soliton. Fract., 144 (2021), 110698. https://doi.org/10.1016/j.chaos.2021.110698 doi: 10.1016/j.chaos.2021.110698
![]() |
[34] | J. Fan, T. Gu, P. Wang, W. Cai, X. Fan, G. Zhang, Constitutive modeling of sintered nano-silver particles: A variable-order fractional model versus an anand model, 2021 22nd International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE), St. Julian, Malta, 2021, 1–4 https://doi.org/10.1109/EuroSimE52062.2021.9410807 |
[35] |
S. Patnaik, J. P. Hollkamp, F. Semperlotti, Applications of variable-order fractional operators: a review, P. Royal Soc. A-Math. Phy., 476 (2020), 20190498. https://doi.org/10.1098/rspa.2019.0498 doi: 10.1098/rspa.2019.0498
![]() |
[36] |
H. Sun, A. Chang, Y. Zhang, W. Chen, A review on variable-order fractional differential equations: mathematical foundations, physical models, numerical methods and applications, Fract. Calc. Appl. Anal., 22 (2019), 27–59. https://doi.org/10.1515/fca-2019-0003 doi: 10.1515/fca-2019-0003
![]() |
[37] |
M. R. S. Ammi, I. Jamiai, D. F. Torres, Finite element approximation for a class of Caputo time-fractional diffusion equations, Comput. Math. Appl., 78 (2019), 1334–1344. https://doi.org/10.1016/j.camwa.2019.05.031 doi: 10.1016/j.camwa.2019.05.031
![]() |
[38] |
B. Li, H. Luo, X. Xie, Analysis of a time-stepping scheme for time fractional diffusion problems with nonsmooth data, SIAM J. Numer. Anal., 57 (2019), 779–798. https://doi.org/10.1137/18M118414X doi: 10.1137/18M118414X
![]() |
[39] |
U. Ali, M. Sohail, M. Usman, F. A. Abdullah, I. Khan, K. S. Nisar, Fourth-order difference approximation for time-fractional modified sub-diffusion equation, Symmetry, 12 (2020), 691. https://doi.org/10.3390/sym12050691 doi: 10.3390/sym12050691
![]() |
[40] |
N. H. Tuan, Y. E. Aghdam, H. Jafari, H. Mesgarani, A novel numerical manner for two‐dimensional space fractional diffusion equation arising in transport phenomena, Numer. Meth. Part. D. E., 37 (2021), 1397–1406. https://doi.org/10.1002/num.22586 doi: 10.1002/num.22586
![]() |
[41] |
F. M. Salama, N. H. M. Ali, Computationally efficient hybrid method for the numerical solution of the 2D time fractional advection-diffusion equation, Int. J. Math. Eng. Manag., 5 (2020), 432–446. https://doi.org/10.33889/IJMEMS.2020.5.3.036 doi: 10.33889/IJMEMS.2020.5.3.036
![]() |
[42] |
N. Dhiman, M. Huntul, M. Tamsir, A modified trigonometric cubic b-spline collocation technique for solving the time-fractional diffusion equation, Eng. Computation., 38 (2021), 2921–2936. https://doi.org/10.1108/EC-06-2020-0327 doi: 10.1108/EC-06-2020-0327
![]() |
[43] |
F. M. Salama, N. H. M. Ali, N. N. A. Hamid, Fast O(N) hybrid Laplace transform-finite difference method in solving 2D time fractional diffusion equation, J. Math. Comput. Sci., 23 (2021), 110–123. https://doi.org/10.22436/jmcs.023.02.04 doi: 10.22436/jmcs.023.02.04
![]() |
[44] |
F. M. Salama, N. N. A. Hamid, N. H. M. Ali, U. Ali, An efficient modified hybrid explicit group iterative method for the time-fractional diffusion equation in two space dimension, AIMS Mathematics, 7 (2022), 2370–2392. https://doi.org/10.3934/math.2022134 doi: 10.3934/math.2022134
![]() |
[45] |
F. M. Salama, N. N. A. Hamid, U. Ali, N. H. M. Ali, Fast hybrid explicit group methods for solving 2d fractional advection-diffusion equation, AIMS Mathematics, 7 (2022), 15854–15880. https://doi.org/10.3934/math.2022868 doi: 10.3934/math.2022868
![]() |
[46] |
M. A. Khan, N. Alias, U. Ali, A new fourth-order grouping iterative method for the time fractional sub-diffusion equation having a weak singularity at initial time, AIMS Mathematics, 8 (2023), 3725–13746. https://doi.org/10.3934/math.2023697 doi: 10.3934/math.2023697
![]() |
[47] |
U. Ali, M. Sohail, F. A. Abdullah, An efficient numerical scheme for variable-order fractional sub-diffusion equation, Symmetry, 12 (2020), 1437. https://doi.org/10.3390/sym12091437 doi: 10.3390/sym12091437
![]() |
[48] |
F. R. Lin, Q. Y. Wang, X. Q. Jin, Crank-Nicolson-Weighted-Shifted-Grunwald-difference schemes for space riesz variable-order fractional diffusion equations, Numer. Algor., 87 (2021), 601–631. https://doi.org/10.1007/s11075-020-00980-z doi: 10.1007/s11075-020-00980-z
![]() |
[49] |
U. Ali, M. Naeem, F. A. Abdullah, M. K. Wang, F. M. Salama, Analysis and implementation of numerical scheme for the variable-order fractional modified sub-diffusion equation, Fractals, 30 (2022), 2240253. https://doi.org/10.1142/S0218348X22402538 doi: 10.1142/S0218348X22402538
![]() |
[50] |
J. Jia, H. Wang, X. Zheng, A preconditioned fast finite element approximation to variable-order time-fractional diffusion equations in multiple space dimension, Appl. Numer. Math., 163 (2021), 15–29. https://doi.org/10.1016/j.apnum.2021.01.001 doi: 10.1016/j.apnum.2021.01.001
![]() |
[51] |
M. H. Heydari, Z. Avazzadeh, M. F. Haromi, A wavelet approach for solving multi-term variable-order time fractional diffusion-wave equation, Appl. Math. Comput., 341 (2019), 215–228. https://doi.org/10.1016/j.amc.2018.08.034 doi: 10.1016/j.amc.2018.08.034
![]() |
[52] |
S. Wei, W. Chen, Y. Zhang, H. Wei, R. M. Garrard, A local radial basis function collocation method to solve the variable-order time fractional diffusion equation in a two-dimensional irregular domain, Numer. Meth. Part. D. E., 34 (2018), 1209–1223. https://doi.org/10.1002/num.22253 doi: 10.1002/num.22253
![]() |
[53] |
J. Jia, H. Wang, X. Zheng, A fast collocation approximation to a two-sided variable-order space-fractional diffusion equation and its analysis, J. Comput. Appl. Math., 388 (2021), 13234. https://doi.org/10.1016/j.cam.2020.113234 doi: 10.1016/j.cam.2020.113234
![]() |
[54] |
X. Y. Li, B. Y. Wu, Iterative reproducing kernel method for nonlinear variable-order space fractional diffusion equations, Int. J. Comput. Math., 95 (2018), 1210–1221. https://doi.org/10.1080/00207160.2017.1398325 doi: 10.1080/00207160.2017.1398325
![]() |
[55] |
F. M. Salama, N. H. M. Ali, N. N. A. Hamid, Efficient hybrid group iterative methods in the solution of two-dimensional time fractional cable equation, Adv. Differ. Equ., 2020 (2020), 257. https://doi.org/10.1186/s13662-020-02717-7 doi: 10.1186/s13662-020-02717-7
![]() |
[56] |
M. A. Khan, N. H. M. Ali, N. N. A. Hamid, A new fourth-order explicit group method in the solution of two-dimensional fractional rayleigh-stokes problem for a heated generalized second-grade fluid, Adv. Differ. Equ., 2020 (2020), 598. https://doi.org/10.1186/s13662-020-03061-6 doi: 10.1186/s13662-020-03061-6
![]() |
[57] |
A. Ali, T. Abdeljawad, A. Iqbal, T. Akram, M. Abbas, On unconditionally stable new modified fractional group iterative scheme for the solution of 2D time-fractional telegraph model, Symmetry, 13 (2021), 2078. https://doi.org/10.3390/sym13112078 doi: 10.3390/sym13112078
![]() |
[58] |
F. M. Salama, A. T. Balasim, U. Ali, M. A. Khan, Efficient numerical simulations based on an explicit group approach for the time fractional advection-diffusion reaction equation, Comput. Appl. Math., 42 (2023), 157. https://doi.org/10.1007/s40314-023-02278-x doi: 10.1007/s40314-023-02278-x
![]() |
[59] |
N. Abdi, H. Aminikhah, A. H. R. Sheikhani, On rotated grid point iterative method for solving 2d fractional reaction-subdiffusion equation with Caputo-Fabrizio operator, J. Differ. Equ. Appl., 27 (2021), 1134–1160. https://doi.org/10.1080/10236198.2021.1965592 doi: 10.1080/10236198.2021.1965592
![]() |
[60] |
F. M. Salama, U. Ali, A. Ali, Numerical solution of two-dimensional time fractional mobile/immobile equation using explicit group methods, Int. J. Appl. Comput. Math., 8 (2022), 188. https://doi.org/10.1007/s40819-022-01408-z doi: 10.1007/s40819-022-01408-z
![]() |
[61] |
Z. Liu, X. Li, A Crank-Nicolson difference scheme for the time variable fractional mobile-immobile advection-dispersion equation, J. Appl. Math. Comput., 56 (2018), 391–410. https://doi.org/10.1007/s12190-016-1079-7 doi: 10.1007/s12190-016-1079-7
![]() |
[62] |
A. R. Abdullah, The four point explicit decoupled group (EDG) method: A fast Poisson solver, Int. J. Comput. Math., 38 (1991), 61–70. https://doi.org/10.1080/00207169108803958 doi: 10.1080/00207169108803958
![]() |
[63] |
X. M. Gu, H. W. Sun, Y. L. Zhao, X. Zheng, An implicit difference scheme for time-fractional diffusion equations with a time-invariant type variable order, Appl. Math. Lett., 120 (2021), 107270. https://doi.org/10.1016/j.aml.2021.107270 doi: 10.1016/j.aml.2021.107270
![]() |
1. | Fouad Mohammad Salama, On Numerical Simulations of Variable-Order Fractional Cable Equation Arising in Neuronal Dynamics, 2024, 8, 2504-3110, 282, 10.3390/fractalfract8050282 | |
2. | S Naveen, V Parthiban, Variable‐order Caputo derivative of LC and RC circuits system with numerical analysis, 2024, 0098-9886, 10.1002/cta.4240 | |
3. | Mahdi Ahmadinia, Mokhtar Abbasi, Parisa Hadi, An intelligent non-uniform mesh to improve errors of a stable numerical method for time-tempered fractional advection–diffusion equation with weakly singular solution, 2024, 80, 0920-8542, 26280, 10.1007/s11227-024-06442-w |
C-N method | EDG method | |||||
α(x,y,t) | h | CPU time | MAE | CPU time | MAE | |
10−(xyt)380 | 1/6 | 0.2499 | 4.1038E-03 | 0.0937 | 7.0379E-03 | |
1/12 | 3.7891 | 1.0223E-03 | 0.8840 | 1.6246E-03 | ||
1/18 | 16.3650 | 3.3934E-04 | 4.0260 | 6.7900E-04 | ||
1/24 | 44.8959 | 2.4740E-04 | 11.4655 | 3.4746E-04 | ||
5yt−3 | 1/6 | 0.4966 | 4.1265E-03 | 0.1284 | 7.0739E-03 | |
1/12 | 4.9944 | 1.0522E-03 | 1.2802 | 1.6364E-03 | ||
1/18 | 18.8372 | 3.6115E-04 | 5.0190 | 6.8719E-04 | ||
1/ 24 | 51.9460 | 2.5265E-04 | 13.7734 | 3.4547E-04 | ||
11−cos(xt)211 | 1/6 | 0.1588 | 3.7833E-03 | 0.0630 | 6.5721E-03 | |
1/12 | 1.7410 | 9.5964E-04 | 0.4935 | 1.5089E-03 | ||
1/18 | 5.9911 | 3.3693E-04 | 1.6087 | 6.3153E-04 | ||
1/24 | 17.5482 | 1.8718E-04 | 4.5734 | 3.0326E-04 | ||
15+sin(yt)516 | 1/6 | 0.2364 | 3.7776E-03 | 0.1117 | 6.5633E-03 | |
1/12 | 2.5489 | 9.5797E-04 | 0.8056 | 1.5115E-03 | ||
1/18 | 12.0167 | 3.3706E-04 | 3.0428 | 6.4030E-04 | ||
1/24 | 39.0032 | 1.8076E-04 | 8.0181 | 3.0591E-04 | ||
9−x5+t3120 | 1/6 | 0.3455 | 4.1132E-03 | 0.0837 | 7.0528E-03 | |
1/12 | 3.6123 | 1.0282E-03 | 1.1557 | 1.6213E-03 | ||
1/18 | 15.4070 | 3.5013E-04 | 4.1639 | 7.0226E-04 | ||
1/24 | 43.8414 | 2.5438E-04 | 13.8434 | 3.3037E-04 | ||
15−sin(xyt)416 | 1/6 | 0.3609 | 3.7780E-03 | 0.1149 | 6.5638E-03 | |
1/12 | 2.9381 | 9.5677E-04 | 1.0479 | 1.5113E-03 | ||
1/18 | 11.8527 | 3.3380E-04 | 2.9980 | 6.3989E-04 | ||
1/24 | 31.7521 | 1.8077E-04 | 8.2564 | 3.0450E-04 |
C-N method | EDG method | |||||
α(x,y,t) | h | CPU time | MAE | CPU time | MAE | |
1−(xyt)3+cos(xyt)210 | 1/4 | 0.0876 | 1.0398E-02 | 0.0409 | 2.2604E-02 | |
1/10 | 3.2858 | 1.7702E-03 | 0.8188 | 3.3201E-03 | ||
1/16 | 18.4236 | 7.9260E-04 | 4.9788 | 1.3317E-03 | ||
1/22 | 63.9827 | 5.9961E-04 | 13.8950 | 7.6442E-04 | ||
exyt+cos(xyt)30 | 1/4 | 0.2551 | 1.0465E-02 | 0.0497 | 2.2765E-02 | |
1/10 | 5.3375 | 1.7854E-03 | 1.2344 | 3.3386E-03 | ||
1/16 | 25.7313 | 8.1088E-04 | 6.0432 | 1.3401E-03 | ||
1/22 | 72.5582 | 6.1617E-04 | 22.3957 | 7.9394E-04 | ||
16−exyt17 | 1/4 | 0.0493 | 1.0000E-02 | 0.0294 | 2.1580E-02 | |
1/10 | 0.9723 | 1.7134E-03 | 0.3137 | 3.1995E-03 | ||
1/16 | 5.8175 | 7.8225E-04 | 1.4358 | 1.2959E-03 | ||
1/22 | 17.8491 | 5.7985E-04 | 4.3844 | 7.4330E-04 | ||
3+(xy)2−(xt)330 | 1/4 | 0.0642 | 1.0449E-02 | 0.0227 | 2.2735E-02 | |
1/10 | 2.3931 | 1.7715E-03 | 0.8921 | 3.3315E-03 | ||
1/16 | 14.1170 | 8.2971E-04 | 2.8375 | 1.3347E-03 | ||
1/22 | 55.0142 | 6.1130E-04 | 9.0424 | 7.7265E-04 | ||
5+(xy)3−(xt)450 | 1/4 | 0.0682 | 1.0448E-02 | 0.0291 | 2.2733E-02 | |
1/10 | 2.7203 | 1.7685E-03 | 0.7045 | 3.3373E-03 | ||
1/16 | 15.1587 | 8.2983E-04 | 3.7406 | 1.3408E-03 | ||
1/22 | 50.2882 | 5.8791E-04 | 11.7378 | 7.7962E-04 | ||
1−xyt+sin(xyt)10 | 1/4 | 0.0845 | 1.0447E-02 | 0.0325 | 2.2731E-02 | |
1/10 | 2.9924 | 1.7737E-03 | 0.7802 | 3.3370E-03 | ||
1/16 | 17.1431 | 8.0822E-04 | 4.2529 | 1.3406E-03 | ||
1/22 | 54.7490 | 5.8871E-04 | 13.4222 | 7.7789E-04 |
h−1=12 | h−1=24 | ||||||||||
C-N | EDG | CN | EDG | ||||||||
α(x,y,t) | CPU time | MAE | CPU time | MAE | CPU time | MAE | CPU time | MAE | |||
10−(xyt)380 | 4.2446 | 2.9811E-03 | 0.9564 | 3.0869E-03 | 57.7629 | 3.7745E-04 | 11.7092 | 5.3877E-04 | |||
5yt−3 | 5.2871 | 3.0868E-03 | 1.2871 | 3.1129E-03 | 69.0271 | 5.4651E-04 | 16.3071 | 6.8800E-04 | |||
11−cos(xt)211 | 1.4539 | 2.4022E-03 | 0.4342 | 2.4534E-03 | 16.7202 | 3.2210E-04 | 5.8109 | 3.1297E-04 | |||
15+sin(yt)516 | 2.6672 | 2.4036E-03 | 0.8971 | 2.4604E-03 | 44.4374 | 3.1177E-04 | 8.8679 | 3.1733E-04 | |||
9−x5+t3120 | 4.0845 | 3.0183E-03 | 1.3774 | 3.1207E-03 | 64.1056 | 4.3822E-04 | 14.0114 | 5.7422E-04 | |||
15−sin(xyt)416 | 2.7664 | 2.3978E-03 | 9.95E-01 | 2.4571E-03 | 36.6685 | 3.1184E-04 | 10.6867 | 3.0756E-04 | |||
1−(xyt)3+cos(xyt)210 | 6.5047 | 2.9637E-03 | 1.931 | 3.0515E-03 | 96.8315 | 3.8151E-04 | 24.4635 | 5.2750E-04 | |||
exyt+cos(xyt)30 | 8.3275 | 3.0230E-03 | 2.3265 | 3.1152E-03 | 125.3629 | 4.5041E-04 | 33.0581 | 5.6177E-04 | |||
16−exyt17 | 1.8139 | 2.3908E-03 | 0.5174 | 2.4595E-03 | 28.0809 | 3.1432E-04 | 6.7689 | 2.5614E-04 | |||
3+(xy)2−(xt)330 | 4.2048 | 3.0191E-03 | 1.0323 | 3.0820E-03 | 69.2202 | 4.1416E-04 | 14.2108 | 5.7308E-04 | |||
5+(xy)3−(xt)450 | 5.9701 | 3.0114E-03 | 1.5368 | 3.0890E-03 | 86.0504 | 4.0849E-04 | 20.5851 | 5.6393E-04 | |||
1−xyt+sin(xyt)10 | 6.4259 | 3.0110E-03 | 1.4328 | 3.0912E-03 | 74.6506 | 4.0501E-04 | 18.0687 | 5.6381E-04 |
C-N | EDG | ||||||
Example | α(x,y,t) | h | MAE | CO | MAE | CO | |
1 | 10−(xyt)380 | 1/4 | 9.0598E-03 | 1.8596E-02 | |||
1/8 | 2.4085E-03 | 1.91 | 3.7500E-03 | 2.31 | |||
1/16 | 5.0287E-04 | 2.25 | 8.7401E-04 | 2.10 | |||
2 | 16−exyt17 | 1/4 | 1.0000E-02 | 2.1580E-02 | |||
1/8 | 2.6132E-03 | 1.93 | 5.0291E-03 | 2.10 | |||
1/16 | 7.8225E-04 | 1.74 | 1.2959E-03 | 1.95 | |||
3 | 11−cos(xt)211 | 1/4 | 2.3825E-02 | 2.6705E-02 | |||
1/8 | 5.8325E-03 | 2.03 | 6.0067E-03 | 2.15 | |||
1/16 | 1.1573E-03 | 2.33 | 1.2173E-03 | 2.30 |
C-N method | EDG method | |||||
α(x,y,t) | h | CPU time | MAE | CPU time | MAE | |
10−(xyt)380 | 1/6 | 0.2499 | 4.1038E-03 | 0.0937 | 7.0379E-03 | |
1/12 | 3.7891 | 1.0223E-03 | 0.8840 | 1.6246E-03 | ||
1/18 | 16.3650 | 3.3934E-04 | 4.0260 | 6.7900E-04 | ||
1/24 | 44.8959 | 2.4740E-04 | 11.4655 | 3.4746E-04 | ||
5yt−3 | 1/6 | 0.4966 | 4.1265E-03 | 0.1284 | 7.0739E-03 | |
1/12 | 4.9944 | 1.0522E-03 | 1.2802 | 1.6364E-03 | ||
1/18 | 18.8372 | 3.6115E-04 | 5.0190 | 6.8719E-04 | ||
1/ 24 | 51.9460 | 2.5265E-04 | 13.7734 | 3.4547E-04 | ||
11−cos(xt)211 | 1/6 | 0.1588 | 3.7833E-03 | 0.0630 | 6.5721E-03 | |
1/12 | 1.7410 | 9.5964E-04 | 0.4935 | 1.5089E-03 | ||
1/18 | 5.9911 | 3.3693E-04 | 1.6087 | 6.3153E-04 | ||
1/24 | 17.5482 | 1.8718E-04 | 4.5734 | 3.0326E-04 | ||
15+sin(yt)516 | 1/6 | 0.2364 | 3.7776E-03 | 0.1117 | 6.5633E-03 | |
1/12 | 2.5489 | 9.5797E-04 | 0.8056 | 1.5115E-03 | ||
1/18 | 12.0167 | 3.3706E-04 | 3.0428 | 6.4030E-04 | ||
1/24 | 39.0032 | 1.8076E-04 | 8.0181 | 3.0591E-04 | ||
9−x5+t3120 | 1/6 | 0.3455 | 4.1132E-03 | 0.0837 | 7.0528E-03 | |
1/12 | 3.6123 | 1.0282E-03 | 1.1557 | 1.6213E-03 | ||
1/18 | 15.4070 | 3.5013E-04 | 4.1639 | 7.0226E-04 | ||
1/24 | 43.8414 | 2.5438E-04 | 13.8434 | 3.3037E-04 | ||
15−sin(xyt)416 | 1/6 | 0.3609 | 3.7780E-03 | 0.1149 | 6.5638E-03 | |
1/12 | 2.9381 | 9.5677E-04 | 1.0479 | 1.5113E-03 | ||
1/18 | 11.8527 | 3.3380E-04 | 2.9980 | 6.3989E-04 | ||
1/24 | 31.7521 | 1.8077E-04 | 8.2564 | 3.0450E-04 |
C-N method | EDG method | |||||
α(x,y,t) | h | CPU time | MAE | CPU time | MAE | |
1−(xyt)3+cos(xyt)210 | 1/4 | 0.0876 | 1.0398E-02 | 0.0409 | 2.2604E-02 | |
1/10 | 3.2858 | 1.7702E-03 | 0.8188 | 3.3201E-03 | ||
1/16 | 18.4236 | 7.9260E-04 | 4.9788 | 1.3317E-03 | ||
1/22 | 63.9827 | 5.9961E-04 | 13.8950 | 7.6442E-04 | ||
exyt+cos(xyt)30 | 1/4 | 0.2551 | 1.0465E-02 | 0.0497 | 2.2765E-02 | |
1/10 | 5.3375 | 1.7854E-03 | 1.2344 | 3.3386E-03 | ||
1/16 | 25.7313 | 8.1088E-04 | 6.0432 | 1.3401E-03 | ||
1/22 | 72.5582 | 6.1617E-04 | 22.3957 | 7.9394E-04 | ||
16−exyt17 | 1/4 | 0.0493 | 1.0000E-02 | 0.0294 | 2.1580E-02 | |
1/10 | 0.9723 | 1.7134E-03 | 0.3137 | 3.1995E-03 | ||
1/16 | 5.8175 | 7.8225E-04 | 1.4358 | 1.2959E-03 | ||
1/22 | 17.8491 | 5.7985E-04 | 4.3844 | 7.4330E-04 | ||
3+(xy)2−(xt)330 | 1/4 | 0.0642 | 1.0449E-02 | 0.0227 | 2.2735E-02 | |
1/10 | 2.3931 | 1.7715E-03 | 0.8921 | 3.3315E-03 | ||
1/16 | 14.1170 | 8.2971E-04 | 2.8375 | 1.3347E-03 | ||
1/22 | 55.0142 | 6.1130E-04 | 9.0424 | 7.7265E-04 | ||
5+(xy)3−(xt)450 | 1/4 | 0.0682 | 1.0448E-02 | 0.0291 | 2.2733E-02 | |
1/10 | 2.7203 | 1.7685E-03 | 0.7045 | 3.3373E-03 | ||
1/16 | 15.1587 | 8.2983E-04 | 3.7406 | 1.3408E-03 | ||
1/22 | 50.2882 | 5.8791E-04 | 11.7378 | 7.7962E-04 | ||
1−xyt+sin(xyt)10 | 1/4 | 0.0845 | 1.0447E-02 | 0.0325 | 2.2731E-02 | |
1/10 | 2.9924 | 1.7737E-03 | 0.7802 | 3.3370E-03 | ||
1/16 | 17.1431 | 8.0822E-04 | 4.2529 | 1.3406E-03 | ||
1/22 | 54.7490 | 5.8871E-04 | 13.4222 | 7.7789E-04 |
h−1=12 | h−1=24 | ||||||||||
C-N | EDG | CN | EDG | ||||||||
α(x,y,t) | CPU time | MAE | CPU time | MAE | CPU time | MAE | CPU time | MAE | |||
10−(xyt)380 | 4.2446 | 2.9811E-03 | 0.9564 | 3.0869E-03 | 57.7629 | 3.7745E-04 | 11.7092 | 5.3877E-04 | |||
5yt−3 | 5.2871 | 3.0868E-03 | 1.2871 | 3.1129E-03 | 69.0271 | 5.4651E-04 | 16.3071 | 6.8800E-04 | |||
11−cos(xt)211 | 1.4539 | 2.4022E-03 | 0.4342 | 2.4534E-03 | 16.7202 | 3.2210E-04 | 5.8109 | 3.1297E-04 | |||
15+sin(yt)516 | 2.6672 | 2.4036E-03 | 0.8971 | 2.4604E-03 | 44.4374 | 3.1177E-04 | 8.8679 | 3.1733E-04 | |||
9−x5+t3120 | 4.0845 | 3.0183E-03 | 1.3774 | 3.1207E-03 | 64.1056 | 4.3822E-04 | 14.0114 | 5.7422E-04 | |||
15−sin(xyt)416 | 2.7664 | 2.3978E-03 | 9.95E-01 | 2.4571E-03 | 36.6685 | 3.1184E-04 | 10.6867 | 3.0756E-04 | |||
1−(xyt)3+cos(xyt)210 | 6.5047 | 2.9637E-03 | 1.931 | 3.0515E-03 | 96.8315 | 3.8151E-04 | 24.4635 | 5.2750E-04 | |||
exyt+cos(xyt)30 | 8.3275 | 3.0230E-03 | 2.3265 | 3.1152E-03 | 125.3629 | 4.5041E-04 | 33.0581 | 5.6177E-04 | |||
16−exyt17 | 1.8139 | 2.3908E-03 | 0.5174 | 2.4595E-03 | 28.0809 | 3.1432E-04 | 6.7689 | 2.5614E-04 | |||
3+(xy)2−(xt)330 | 4.2048 | 3.0191E-03 | 1.0323 | 3.0820E-03 | 69.2202 | 4.1416E-04 | 14.2108 | 5.7308E-04 | |||
5+(xy)3−(xt)450 | 5.9701 | 3.0114E-03 | 1.5368 | 3.0890E-03 | 86.0504 | 4.0849E-04 | 20.5851 | 5.6393E-04 | |||
1−xyt+sin(xyt)10 | 6.4259 | 3.0110E-03 | 1.4328 | 3.0912E-03 | 74.6506 | 4.0501E-04 | 18.0687 | 5.6381E-04 |
C-N | EDG | ||||||
Example | α(x,y,t) | h | MAE | CO | MAE | CO | |
1 | 10−(xyt)380 | 1/4 | 9.0598E-03 | 1.8596E-02 | |||
1/8 | 2.4085E-03 | 1.91 | 3.7500E-03 | 2.31 | |||
1/16 | 5.0287E-04 | 2.25 | 8.7401E-04 | 2.10 | |||
2 | 16−exyt17 | 1/4 | 1.0000E-02 | 2.1580E-02 | |||
1/8 | 2.6132E-03 | 1.93 | 5.0291E-03 | 2.10 | |||
1/16 | 7.8225E-04 | 1.74 | 1.2959E-03 | 1.95 | |||
3 | 11−cos(xt)211 | 1/4 | 2.3825E-02 | 2.6705E-02 | |||
1/8 | 5.8325E-03 | 2.03 | 6.0067E-03 | 2.15 | |||
1/16 | 1.1573E-03 | 2.33 | 1.2173E-03 | 2.30 |