Research article Special Issues

Orthonormal Euler wavelets method for time-fractional Cattaneo equation with Caputo-Fabrizio derivative

  • In this paper, a new orthonormal wavelets based on the orthonormal Euler polynomials (OEPs) is constructed to approximate the numerical solution of time-fractional Cattaneo equation with Caputo-Fabrizio derivative. By applying the Gram-Schmidt orthonormalization process on sets of Euler polynomials of various degrees, an explicit representation of OEPs is obtained. The convergence analysis and error estimate of the orthonormal Euler wavelets expansion are studied. The exact formula of Caputo-Fabrizio fractional integral of orthonormal Euler wavelets are derived using Laplace transform. The applicability and validity of the proposed method are verified by some numerical examples.

    Citation: Xiaoyong Xu, Fengying Zhou. Orthonormal Euler wavelets method for time-fractional Cattaneo equation with Caputo-Fabrizio derivative[J]. AIMS Mathematics, 2023, 8(2): 2736-2762. doi: 10.3934/math.2023144

    Related Papers:

    [1] Yudhveer Singh, Devendra Kumar, Kanak Modi, Vinod Gill . A new approach to solve Cattaneo-Hristov diffusion model and fractional diffusion equations with Hilfer-Prabhakar derivative. AIMS Mathematics, 2020, 5(2): 843-855. doi: 10.3934/math.2020057
    [2] Aslı Alkan, Halil Anaç . A new study on the Newell-Whitehead-Segel equation with Caputo-Fabrizio fractional derivative. AIMS Mathematics, 2024, 9(10): 27979-27997. doi: 10.3934/math.20241358
    [3] Zeliha Korpinar, Mustafa Inc, Dumitru Baleanu . On the fractional model of Fokker-Planck equations with two different operator. AIMS Mathematics, 2020, 5(1): 236-248. doi: 10.3934/math.2020015
    [4] Sadia Arshad, Iram Saleem, Ali Akgül, Jianfei Huang, Yifa Tang, Sayed M Eldin . A novel numerical method for solving the Caputo-Fabrizio fractional differential equation. AIMS Mathematics, 2023, 8(4): 9535-9556. doi: 10.3934/math.2023481
    [5] Manoj Singh, Ahmed Hussein, Msmali, Mohammad Tamsir, Abdullah Ali H. Ahmadini . An analytical approach of multi-dimensional Navier-Stokes equation in the framework of natural transform. AIMS Mathematics, 2024, 9(4): 8776-8802. doi: 10.3934/math.2024426
    [6] Gulalai, Shabir Ahmad, Fathalla Ali Rihan, Aman Ullah, Qasem M. Al-Mdallal, Ali Akgül . Nonlinear analysis of a nonlinear modified KdV equation under Atangana Baleanu Caputo derivative. AIMS Mathematics, 2022, 7(5): 7847-7865. doi: 10.3934/math.2022439
    [7] Khalid K. Ali, K. R. Raslan, Amira Abd-Elall Ibrahim, Mohamed S. Mohamed . On study the fractional Caputo-Fabrizio integro differential equation including the fractional q-integral of the Riemann-Liouville type. AIMS Mathematics, 2023, 8(8): 18206-18222. doi: 10.3934/math.2023925
    [8] Sami Ul Haq, Saeed Ullah Jan, Syed Inayat Ali Shah, Ilyas Khan, Jagdev Singh . Heat and mass transfer of fractional second grade fluid with slippage and ramped wall temperature using Caputo-Fabrizio fractional derivative approach. AIMS Mathematics, 2020, 5(4): 3056-3088. doi: 10.3934/math.2020198
    [9] Sajjad Ali Khan, Kamal Shah, Poom Kumam, Aly Seadawy, Gul Zaman, Zahir Shah . Study of mathematical model of Hepatitis B under Caputo-Fabrizo derivative. AIMS Mathematics, 2021, 6(1): 195-209. doi: 10.3934/math.2021013
    [10] Khalid Khan, Amir Ali, Manuel De la Sen, Muhammad Irfan . Localized modes in time-fractional modified coupled Korteweg-de Vries equation with singular and non-singular kernels. AIMS Mathematics, 2022, 7(2): 1580-1602. doi: 10.3934/math.2022092
  • In this paper, a new orthonormal wavelets based on the orthonormal Euler polynomials (OEPs) is constructed to approximate the numerical solution of time-fractional Cattaneo equation with Caputo-Fabrizio derivative. By applying the Gram-Schmidt orthonormalization process on sets of Euler polynomials of various degrees, an explicit representation of OEPs is obtained. The convergence analysis and error estimate of the orthonormal Euler wavelets expansion are studied. The exact formula of Caputo-Fabrizio fractional integral of orthonormal Euler wavelets are derived using Laplace transform. The applicability and validity of the proposed method are verified by some numerical examples.



    Over the past few decades, fractional differential equations (FDEs) have gained importance in the modelling of various physical phenomena due to their ability to describe the memory and hereditary properties of different materials and processes. Various fractional models based on Caputo derivative have been successfully established, such as COVID-19 transmission [1], HIV-1 infection of CD4+ T-cells [2], epidemiological MSEIR model [3], model of cancer chemotherapy effect [4], fractional model for Maxwell fluid [5]. Due to the singularity of the definition of Caputo derivatives, it causes some difficulties to the numerical and analytical methods of fractional calculus. Recently, Caputo and Fabrizio[6] proposed a new definition of fractional derivative by replacing the singular kernel in the Caputo derivative with exponential kernel, which was called Caputo-Fabrizio fractional derivative. This new operator has been successfully applied in human liver model [7], groundwater flow model [8,9], electrical circuits model [10], COVID-19 model[11], Korteweg-de Vries-Burgers equation [12], fractional Maxwell fluid[13], pneumococcal pneumonia infection model[14] and transmission dynamics of brucellosis model [15].

    The standard Cattaneo equation is normally obtained by using a generalized form of Fick's law, which describes a diffusion process with finite propagation velocity. Cattaneo equation is a hyperbolic partial differential equation as follows:

    u(x,t)t+2u(x,t)t2=2u(x,t)x2, (1.1)

    where u(x,t) denotes distribution function of diffusing-quantity/temperature [16]. However, the classical Cattaneo equation cannot describe the anomalous diffusion behavior observed in many natural systems. To address this issue, Compte and Metzler [17] extended the classical Cattaneo model to the time-fractional Cattaneo model, and studied the properties of the corresponding fractional Cattaneo equation. Following Compte and Metzler, Kosztolowicz and Lewandowska [18] proposed a theoretical basis for the study of subdiffusion impedance using a hyperbolic equation. Since then, many researchers have numerically considered the fractional Cattaneo equation in the sense of Caputo [19,20,21,22,23,24,25].

    Caputo-Fabrizio derivative has many interesting properties. It can portray substance heterogeneities and configurations with different scales, which cannot be handled by the well-known local theory. In recent years, fractional Cattaneo equation based on Caputo-Fabrizio derivative has attracted much attention from many researchers. Since finding the solutions of these differential equations is a difficult task, it is necessary to find the solutions of partial differential equations(PDEs) by numerical methods. Liu [26] introduced a second order Crank-Nicolson scheme for fractional Cattaneo equation based on Caputo-Fabrizio derivative. Taghipour [27] considered a θ-finite difference scheme using cubic B-spline for the fractional Cattaneo equations. In [28], the spline-based collocation schemes are developed for the numerical solution of the time-fractional Cattaneo equations involving the Caputo-Fabrizio derivative. Li [29] proposed a fully discrete spectral method for fractional Cattaneo equation having Caputo-Fabrizio derivative, in which finite difference method is used in time and Legendre spectral approximation in space. Note that, the above-mentioned papers applied finite difference scheme in time direction, owing to the memory effect of fractional derivative, to compute the solution at the current time level, all previous solutions have to be saved, which would make the storage very expensive if low-order methods are employed for spatial discretization.

    In this paper, we consider the following time-fractional Cattaneo equation with nonhomogeneous term [26]:

    u(x,t)t+γu(x,t)tγ=2u(x,t)x2+f(x,t), (1.2)

    where (x,t)Ω=[0,L]×[0,T], 1<γ<2, f(x,t)C(Ω) with initial conditions

    u(x,0)=ϕ(x),ut|t=0=ψ(x),0xL, (1.3)

    and boundary conditions

    u(0,t)=u(L,t)=0,t>0. (1.4)

    CF0Dγtu(x,t) denotes the Caputo-Fabrizio operator of the function u(x,t) defined as

    CF0Dγtu(x,t)=M(γ)2γt0u(x,s)eσ(ts)ds, (1.5)

    where M(γ) is a normalization function such that M(0)=M(1)=1 and σ=1γ2γ.

    As a global method, spectral method can achieve exponential rates of convergence for smooth problems and requires fewer grid points to produce highly accurate solution, which makes it widely used for solving the numerical solutions of different types of PDEs. Numerical methods based on some orthogonal or semi-orthogonal polynomials [30,31,32] are widely used to solve fractional differential equtions (FDEs). These methods utilize operational matrix of derivatives or integration to transform complicated FDEs into a system of linear or nonlinear algebraic equations. Recently, Genochi polynomial [34], Legendre polynomial [35] and Chebyshev polynomial [36] are applied to solve FDEs with Caputo-Fabrizio derivative. Orthogonal wavelet is a special orthogonal function with compact support, such as Legendre wavelet [37,38], Chebyshev wavelet [39,40] and Jacobi wavelet [41,42], which can approximate discontinuous or rapidly changing functions at different resolutions. The main characteristic of wavelet-based technology is that the coefficient matrix of the discretized algebraic equation is a sparse matrix, which reduces the amount of calculation and accelerates the simulation speed. Since the Caputo-Fabrizio operator is a relatively new operator, as far as we know, there is no paper using orthogonal polynomial wavelets to solve fractional Cattaneo equations with Caputo-Fabrizio derivative. Euler wavelets have good accuracy for solving various differential equations [43,44,45], but do not have orthogonality. It is actually easier to use orthonormal Euler wavelets than Euler wavelets.

    Motivated by the above discussion, our main target is to present an effective numerical method based on orthonormal Euler wavelets to solve time fractional Cattaneo equation with Caputo-Fabrizio derivative. Therefore, we construct orthonormal Euler wavelets by Gram-Schmidt orthogonalization of Euler polynomials. In addition, to obtain more accurate numerical solutions, Laplace transform is utilized to obtain the exact Caputo-Fabrizio fractional integral formula of Euler wavelets. The proposed method has the following advantages:

    (i) Caputo-Fabrizio fractional integral operator for the orthonormal Euler wavelets has been derived directly with no approximations.

    (ii) The present method is convenient for solving this problem, since the boundary and initial conditions are taken into account automatically in the algorithm.

    (iii) This method is applicable to linear and nonlinear Cattaneo equations.

    The paper is organized as follows. Section 2 introduces some fundamental definitions of fractional calculus and Laplace transform. Section 3 is devoted to the construction of orthonormal Euler wavelet. Convergence analysis and error estimate of the orthonormal Euler wavelets expansion are displayed in Section 4. An exact formula for the Caputo-Fabrizio fractional integral of orthonormal Euler wavelets is derived in Section 5. The description of the proposed method is presented in Section 6. Some illustrative examples are provided in Section 7. Finally, conclusion and discussion is given in Section 8.

    In this section, some necessary definitions and mathematical preliminaries are given.

    Definition 2.1. (See [46]) The Laplace transform F(s) of a locally integrable function f(x) is defined by

    L[f(x)]=0esxf(x)dx, (2.1)

    where s is a complex number, and this operator has the following properties:

    (1) L[λ1f1(x)+λ2f2(x)]=λ1L[f1(x)]+λ2L[f2(x)],

    (2) L[f(xa)μ(xa)]=easF(s),

    (3) L[fg]=L[f(x)]L[g(x)],

    where λ1, λ2 and a are constants, μ(x) is the step function and fg is the convolution of two functions f and g.

    Definition 2.2. (See [47]) Let n1 and α[0,1], the Caputo-Fabrizio fractional derivative with order α of a function f(x) is defined by

    CFDn+αaf(x)=M(α)1αxaf(n+1)(s)exp(α(xs)1α)ds. (2.2)

    Definition 2.3. (See [47]) Let n1, α[0,1], and fC1[a,b]. The Caputo-Fabrizio integral of order α+n of a function f is given by

    CFJn+αaf(x)=1M(α)n!xa(xs)n1[α(xs)+n(1α)]f(s)ds, (2.3)

    where M(α) is a normalization function such that M(0)=M(1)=1.

    Lemma 2.1. (See [47]) Let γ(n,n+1), γ=n+α, n=[γ]0, α[0,1]. Assume that f(x)Cn[a,b], then the following statements hold:

    (1)CFJγa(CFDγaf(x))=f(x)ni=0f(i)(a)i!(xa)i,i=0,1,2,,n. (2.4)
    (2)CFDγa(CFJγaf(x))=f(x)f(a)exp(α(xa)1α). (2.5)

    Definition 2.4. (See [48]) The Riemann-Liouville fractional integral of order α(α>0) is defined as

    Iαf(x)=1Γ(α)x0(xs)α1f(s)ds,α>0. (2.6)

    The main goal of this section is to construct the orthonormal Euler wavelets (OEWs). To obtain orthonormal Euler polynomials (OEPs), we apply the Gram-Schmidt orthonormalization process on sets of Euler polynomials. Then, the orthonormal Euler wavelets will be constructed using the orthonormal Euler polynomials and their properties.

    Definition 3.1. (See [49]) The Euler polynomials Em(x) of degree m can be constructed from the following relation:

    mk=0(mk)Ek(x)+Em(x)=2xm,m=0,1,2,. (3.1)

    From Eq (3.1), the first few terms of Euler polynomials are listed below:

    E0(x)=1,E1(x)=x12,E2(x)=x2x,E3(x)=x332x2+14,E4(x)=x42x3+x,E5(x)=x552x4+52x212,. (3.2)

    Euler polynomials has the following properties:

    (1)Em(x)=mEm1(x),m=1,2,,(2)Em(x+1)+Em(x)=2xm,(3)10Em(x)En(x)dx=(1)n1m!(n+1)!(m+n+1)!Em+n+1(0),m,n1.

    Note that Euler polynomials are not mutually orthogonal. To overcome this difficulty, we try to apply the Gram-Schmidt orthogonal normalization process on Em(x) with respect to the weight function w(x)=1. We denote the orthonormal Euler polynomials by OEm(x). For instance, for m=5, we have

    OE0(x)=1,OE1(x)=3(2x1),OE2(x)=5(6x26x+1),
    OE3(x)=7(20x330x2+12x1),
    OE4(x)=9(70x4140x3+90x220x+1),
    OE5(x)=11(252x5630x4+560x3210x2+30x1).

    By analyzing these coefficients, we can get an explicit representation of OEPs.

    Definition 3.2. The explicit form of the OEPs of degree m on the interval [0,1] is given by

    OEm(x)=2m+1mk=0(1)mk(mk)(m+kk)xk,m=0,1,2,. (3.3)

    The orthogonality property for these polynomials is as follows:

    10OEi(x)OEj(x)dx=δi,j,i,j=0,1,2,, (3.4)

    where δi,j is the Kronecker delta function. Furthermore, by analyzing the integrals of these polynomials from 0 to x, we obtain

    x0OE0(t)dt=12OE0(x)+123OE1(x),

    and for m1,

    x0OEm(t)dt=12(2m+1)(2m+3)OEm+1(x)12(2m+1)(2m1)OEm1(x). (3.5)

    Therefore, we also have

    22m+1OEm(x)=12m+3OEm+1(x)12m1OEm1(x),m1. (3.6)

    Wavelets are a set of functions which can be defined from dilation and translation of a mother wavelet function ψ(x). When the dilation and translation parameters vary continuously, we get a series of continuous wavelet functions[50]. The orthonormal Euler wavelets (OEWs) ψn,m(x) have four arguments n=1,2,,2k1 can assume any positive integer, m is the order of the orthonormal Euler polynomials and x is the normalized variable. They are defined on the interval [0,1) by

    ψn,m(x)={2k12OEm(2k1xn+1),n12k1x<n2k1,0,otherwise, (3.7)

    where m=0,1,, and OEm(x) are the OEPs of degree m defined in Eq (3.3). It is easy to verify that the constructed OEWs ψn,m(x), n=1,2,,2k1, m=0,1,, constitutes an orthonormal set on the interval [0,1). Suppose f(x)L2[0,1) can be expressed in terms of OEWs as

    f(x)=2k1n=1m=0cn,mψn,m(x)2k1n=1M1m=0cn,mψn,m(x)=CTΨ(x), (3.8)

    where C and Ψ(x) are 2k1M×1 matrices given by

    C=[c1,0,c1,1,,c1,M1,c2,0,,c2,M1,,c2k1,0,,c2k1,M1]T,
    Ψ(x)=[ψ1,0(x),ψ1,1(x),,ψ1,M1(x),ψ2,0(x),,ψ2,M1(x),,ψ2k1,0(x),,ψ2k1,M1(x)]T, (3.9)

    and

    cn,m=f(x),ψn,m(x)L2[0,1)=10ψn,m(x)f(x)dx. (3.10)

    Two-dimensional orthonormal Euler wavelets can be expressed as product of 1D-OEWs as follows:

    ψn,m,n,m(x,y)={ψn,m(x)ψn,m(y),n12k1x<n2k1,n12k1y<n2k1,0,otherwise, (3.11)

    where n=1,2,,2k1, m=0,1,. It is easy to verify that {ψn,m(x)ψn,m(y)} is an orthonormal set over [0,1)×[0,1). Similarly, an arbitrary function of two variables f(x,y) defined over [0,1)×[0,1) can be expanded into 2D-OEWs basis as

    f(x,y)=2k1n=1m=02k1n=1m=0cn,m,n,mψn,m(x)ψn,m(y)2k1n=1M1m=02k1n=1M1m=0cn,m,n,mψn,m(x)ψn,m(y)=ΨT(x)UΨ(y),

    where cn,m,n,m=ψn,m(x),f(x,y),ψn,m(y) is the coefficient of 2D-OEWs, in which , denotes the inner product and U is a 2k1M×2k1M coefficient matrix.

    In this section, we prove the convergence of OEWs with respect to the L2-norm. To this end, we first derive the upper error bound for the mentioned OEWs expansion.

    Theorem 4.1. Let 2k1n=1m=0cn,mψn,m(x) be the orthonormal Euler wavelet expansion of f(x)L2[0,1) and suppose f(x) has a bounded second-order derivative, say |f(x)|B, then for m2, the following inequality holds:

    |cn,m|23B252k2m+1(2m3)3, (4.1)

    where cn,m=f(x),ψn,m(x)L2[0,1) is given in (3.10).

    Proof. The orthonormal Euler wavelets coefficient is given as follows:

    cn,m=n2k1n12k1f(x)2k12OEm(2k1xn+1)dx.

    By using the substitution t=2k1xn+1 and the relation (3.6), we have

    cn,m=21+k22m+110f(t+n12k1)d[OEm+1(t)2m+3OEm1(t)2m1].

    Taking integration by parts and using the relation (3.6) yield that

    cn,m=213k22m+110f(t+n12k1)d[OEm+2(t)2(2m+3)2m+5OEm(t)2(2m+3)2m+1OEm(t)2(2m1)2m+1+OEm2(t)2(2m1)2m3].

    By using the integration by parts again, we have

    cn,m=235k22m+110f(t+n12k1)[OEm+2(t)2(2m+3)2m+5OEm(t)2(2m+3)2m+1OEm(t)2(2m1)2m+1+OEm2(t)2(2m1)2m3]dt=235k22m+110f(t+n12k1)[(2m+1)2m+5OEm+2(t)(2m+5)2m+1OEm(t)2(2m+1)(2m+3)(2m+5)(2m3)2m+1OEm(t)(2m+1)2m3OEm2(t)2(2m3)(2m1)(2m+1)]dt.

    By applying the Cauchy-Schwarz inequality and using the assumption that |f(x)|B, we obtain

    |10f(t+n12k1)[(2m+1)2m+5OEm+2(t)(2m+5)2m+1OEm(t)2(2m+1)(2m+3)(2m+5)(2m3)2m+1OEm(t)(2m+1)2m3OEm2(t)2(2m3)(2m1)(2m+1)]dt|210|f(t+n12k1)|2|(2m+1)2m+5OEm+2(t)(2m+5)2m+1OEm(t)2(2m+1)(2m+3)(2m+5)(2m3)2m+1OEm(t)(2m+1)2m3OEm2(t)2(2m3)(2m1)(2m+1)|2dtB210[(2m3)2(2m1)2(2m+1)2(2m+5)OE2m+2(t)4(2m3)2(2m1)2(2m+1)2(2m+3)2(2m+5)2+4(2m3)2(2m+1)2(2m+5)2(2m+1)OE2m(t)4(2m3)2(2m1)2(2m+1)2(2m+3)2(2m+5)2+(2m+1)2(2m+3)2(2m+5)2(2m3)OE2m2(t)4(2m3)2(2m1)2(2m+1)2(2m+3)2(2m+5)2]dtB2[14(2m+3)2(2m+5)+44(2m1)2(2m+1)+14(2m1)2(2m3)]3B22(2m1)2(2m3).

    Therefore, for m2,

    |cn,m|23B252k2m+1(2m3)3. (4.2)

    Theorem 4.2. Let f(x)L2[0,1) be a bounded function with second derivative say |f(x)|B. Then the following inequality holds:

    f(x)CTΨ(x)2B22k(2M4)3. (4.3)

    Proof. By the definition of the L2-norm, we have

    f(x)CTΨ(x)22=10(f(x)CTΨ(x))2dx=10(2k1n=1m=0cn,mψn,m(x)2k1n=1M1m=0cn,mψn,m(x))2dx=10(2k1n=1m=Mc2n,mψ2n,m(x))dx=2k1n=1m=Mc2n,m.

    The above equation holds due to the fact that the constructed Euler wavelets are orthonormal. By the Theorem 4.1, we have

    f(x)CTΨ(x)22=2k1n=1m=Mc2n,m3B224k23m=M1(m2)4. (4.4)

    Using the integral inequality yields that

    m=M1(m2)4=m=M21m4M21m4dx=13(M2)3. (4.5)

    Therefore,

    f(x)CTΨ(x)22B224k231(M2)3. (4.6)

    Here, take the square root for both sides of (4.6), then

    f(x)CTΨ(x)2B22k1(2M4)3. (4.7)

    The proof is completed.

    Corollary 4.1. The upper bound of error given in Theorem 4.2 is related to k and M. Obviously, for a fixed k, B22k1(2M4)3 tends to zero as M tends to . Therefore, we conclude that CTΨ(x) converges to f(x) with respect to the L2-norm.

    Theorem 4.3. Let fL2[0,1) be a bounded function with second derivative say |f(x)|B. Then the following inequality holds:

    CFJγaf(x)CFJγafk,M(x)21Γ(η)12η(2η1)B22k(2M3)3, (4.8)

    where γ=η+α, η=[γ], fk,M(x)=CTΨ(x).

    Proof. The definition of Caputo-Fabrizio integral can be further simplified as follows:

    CFJη+αaf(x)=1αM(α)Iηf(x)+αM(α)Iη+1f(x), (4.9)

    using the Cauchy formula [51], where

    Iηf(x)=1Γ(η)xa(xs)η1f(s)ds. (4.10)

    By the Hölder's inequality, we can estimate the difference between CFJγaf(x) and CFJγafk,M(x) at x[0,1], as follows:

    |CFJγaf(x)CFJγafk,M(x)|1αM(α)1Γ(η)xa|f(s)fk,M(s)|(xs)1ηds+αM(α)1Γ(η+1)xa|f(s)fk,M(s)|(xs)ηds1αM(α)1Γ(η)(xads(xs)22η)1/2(xa(f(s)fk,M(s))ds)1/2+αM(α)1Γ(η+1)(xads(xs)2η)1/2(xa(f(s)fk,M(s))ds)1/2(1αM(α)1Γ(η)1(2η1)(xa)η12+αM(α)1Γ(η+1)1(2η+1)(xa)η+12)f(x)CTΨ(x)2. (4.11)

    Hence,

    CFJγafCFJγafk,M2(1αM(α)1Γ(η)1(2η1)(xa)η122+αM(α)1Γ(η+1)1(2η+1)(xa)η+122)f(x)CTΨ(x)2(1αM(α)1Γ(η)12η(2η1)+αM(α)1Γ(η)1(2η+1)(2η+2))B22k1(2M4)31Γ(η)12η(2η1)B22k1(2M4)3. (4.12)

    The proof is completed.

    Theorem 4.4. If a continuous function f(x,y)L2([0,1)×[0,1)) has bounded mixed fourth partial derivative |4f(x,y)x2y2|˜B, then 2D-OEWs expansion of f(x,y) converges uniformly to the function f(x,y), that is

    f(x,y)=2k1n=1m=02k1n=1m=0cn,m,n,mψn,m(x)ψn,m(y), (4.13)

    where cn,m,n,m=ψn,m(x),f(x,y),ψn,m(y).

    Proof.

    cn,m,n,m=1010f(x,y)ψn,m(x)ψn,m(y)dxdy=10ψn,m(y)(n2k1n12k1f(x,y)2k12OEm(2k1xn+1)dx)dy=n2k1n12k12k12OEm(2k1yn+1)(n2k1n12k1f(x,y)2k12OEm(2k1xn+1)dx)dy.

    By change of variable s=2k1xn+1, t=2k1yn+1, dx=12k1ds and dy=12k1dt, we obtain

    cn,m,n,m=21k210OEm(t)(10f(s+n12k1,y)21k2OEm(s)ds)dt. (4.14)

    Similar to the previous steps used in the Theorem 4.1, we have

    |cn,m,n,m|23˜B252k2m+1(2m3)323252k2m+1(2m3)312˜B252(k+k)(2m3)2(2m3)2.

    Since n2k1, n2k1, we get

    |cn,m,n,m|12˜B(2n)52(2m3)2(2n)52(2m3)2. (4.15)

    Therefore, 2k1n=1m=02k1n=1m=0cn,m,n,mψn,m(x)ψn,m(y) is absolutely convergent. If n=n, m=m, then we have

    |cn,m,n,m|12˜B(2n)5(2m3)4.

    In this section, we will derive the exact formula for the fractional integral of orthonormal Euler wavelets ψn,m(x) in the sense of Caputo-Fabrizio. In fact, the αth repeated integral of f(x) is the special case of Riemann-Liouville fractional integral, when α is a positive integer. To derive the Caputo-Fabrizio fractional integral of orthonormal Euler wavelets, first we need to derive the Riemann-Liouville fractional integral of order α of the orthonormal Euler wavelets.

    Theorem 5.1. The Caputo-Fabrizio fractional integral of order γ>0 of orthonormal Euler wavelets ψn,m(x) is given by

    CFJγaψn,m(x)=1βM(β)Inψn,m(x)+βM(β)In+1ψn,m(x), (5.1)

    where γ=n+β, n=[γ]0, β[0,1] and Inψn,m(x) is given by the following theorem. To make the theorem more general, we derive the Riemann-Liouville fractional integral of order α>0 of the orthonormal Euler wavelets.

    Theorem 5.2. The Riemann-Liouville fractional integral of order α>0 of the orthonormal Euler wavelets ψn,m(x) is given by

    Iαψn,m(x)={0,0xn12k1,U(x),n12k1xn2k1,U(x)V(x),n2k1x1, (5.2)

    where

    U(x)=2k122m+1mi=0(1)miCimCim+i2i(k1)Γ(i+1)Γ(i+α+1)(xn12k1)i+α,

    and

    V(x)=2k122m+1mi=0(1)miCimCim+iir=0Cri2r(k1)Γ(r+1)Γ(r+α+1)(xn2k1)r+α.

    Proof. See Appendix.

    For instance, in the case of k=2, M=3, γ=5/2, x=0.65, we obtain

    CFJ5/20Ψ6×1(0.65)=(0.173388475303450.071443450831180.003294039229340.008352698852770.011608284984160.00925321971836),

    where Ψ6×1(x)=(ψ1,0(x),ψ1,1(x),ψ1,2(x),ψ2,0(x),ψ2,1(x),ψ2,2(x))T.

    For simplicity, we consider the following time-fractional Cattaneo equation with T=1, L=1:

    u(x,t)t+γu(x,t)tγ=2u(x,t)x2+f(x,t), (6.1)
    u(x,0)=ϕ(x),ut|t=0=ψ(x),0x1, (6.2)
    u(0,t)=u(1,t)=0,0t1. (6.3)

    To solve the time-fractional Cattaneo eqution (6.1), we assume

    2+γu(x,t)x2tγ=ΨT(x)UΨ(t), (6.4)

    where U=(uij)2k1M×2k1M is an unknown wavelet coefficient matrix and Ψ() is the vector defined in (3.9). By applying the Caputo-Fabrizio fractional integral of order γ with respect to t on both sides of Eq (6.4) and considering the property of Caputo-Fabrizio fractional integral, we obtain

    2u(x,t)x2=ΨT(x)UCFJγ0Ψ(t)+2u(x,t)x2|t=0+tt(2u(x,t)x2)|t=0. (6.5)

    Using the conditions (6.2), we yield

    2u(x,t)x2=ϕ(x)+tψ(x)+ΨT(x)UCFJγ0Ψ(t). (6.6)

    By integrating Eq (6.6) two times with respect to x and combining the conditions (6.2), we have

    u(x,t)=xux|x=0+ϕ(x)ϕ(0)xϕ(0)+t(ψ(x)ψ(0)xψ(0))+(I2Ψ(x))TUCFJγ0Ψ(t). (6.7)

    Putting x=1 in Eq (6.7), we get

    ux|x=0=ϕ(1)ϕ(0)ϕ(0)+t(ψ(1)ψ(0)ψ(0))+(I2Ψ(1))TUCFJγ0Ψ(t). (6.8)

    Therefore, we have

    u(x,t)=(ϕ(x)ϕ(0)xϕ(0))+t(ψ(x)ψ(0)xψ(0))+(I2Ψ(x))TUCFJγ0Ψ(t)x[ϕ(1)ϕ(0)ϕ(0)+t(ψ(1)ψ(0)ψ(0))+(I2Ψ(1))TUCFJγ0Ψ(t)]. (6.9)

    Taking the first derivative of Eq (6.9) with respect to t, we get

    u(x,t)t=ψ(x)ψ(0)xψ(0)+(I2Ψ(x))TUCFJγ10Ψ(t)x[ψ(1)ψ(0)ψ(0)+(I2Ψ(1))TUCFJγ10Ψ(t)]. (6.10)

    Again, taking the Caputo-Fabrizio fractional derivative of order γ of Eq (6.9) with respect to t and considering the Lemma 2.1, we obtain

    γu(x,t)tγ=(I2Ψ(x))TU(Ψ(t)Ψ(0)eαt1α)x(I2Ψ(1))TU(Ψ(t)Ψ(0)eαt1α). (6.11)

    Now, by substituting Eqs (6.6), (6.10) and (6.11) into Eq (6.1) and considering the collocation points xi=2i12kM,tj=2j12kM,i,j=1,2,,2k1M, we obtain the following linear system of algebraic equation:

    (I2Ψ(xi)xiI2Ψ(1))TU(CFJγ10Ψ(tj)+Ψ(tj)Ψ(0)eαtj1α)ΨT(xi)UCFJγ0Ψ(tj)=f(xi,tj)(ψ(xi)ψ(0)xiψ(0))+xi(ψ(1)ψ(0)ψ(0))+(ϕ(xi)+tj(ψ(xi))). (6.12)

    Solving this system, we obtain the unknown Euler wavelet coefficient vector U and thereafter substituting U into Eq (6.9), we obtain the approximate solutions of the given time-fractional Cattaneo equation (6.1).

    In this section, we provide several numerical examples with exact solution to demonstrate the accuracy of the orthonormal Euler wavelets collocation method. All the numerical experiments are performed by Matlab 7.0. We report errors by using

    (i) Mean root square norm (L2):

    L2error=mi=0mj=0(uk,M(xi,tj)u(xi,tj))2m2,

    (ii) Maximum error norm (L):

    Lerror=max1im,1jm|uk,M(xi,tj)u(xi,tj)|,

    where uk,M(xi,tj) is the approximated solution and u(xi,tj) is the exact solution.

    Example 1. Consider the following time-fractional Cattaneo equation:

    {u(x,t)t+γu(x,t)tγ=2u(x,t)x2+f(x,t),0<x<1,0t1,1<γ<2,u(0,t)=0,u(1,t)=0,u(x,0)=0,ut|t=0=0, (7.1)

    where f(x,t)=2(1x2)x163(t+1eσtγ1)+t2(4189x1632089x103). The exact solution is u(x,t)=t2(1x2)x163. Figure 1 shows the numerical solution (left) and absolute error (right) for some different points of [0,1]×[0,1] with γ=5/4, k=5 and M=5. For various γ, Table 1 gives the L, L2 errors with different k and M.

    Figure 1.  The plot of numerical solution and pointwise errors for Example 1 with γ=5/4, k=5 and M=5.
    Table 1.  Error estimate for Example 1 with different k and M.
    γ norm k=3,M=3 k=4,M=4 k=5,M=5 k=6,M=6
    1.2 L 4.4291e-4 4.8747e-6 6.1905e-11 5.2767e-13
    L2 1.2791e-4 1.4450e-6 1.1791e-11 1.0202e-13
    1.4 L 4.3108e-4 4.7223e-6 6.1689e-11 5.2645e-13
    L2 1.2334e-4 1.3840e-6 1.1756e-11 1.0039e-13
    1.6 L 4.2223e-4 4.5005e-6 5.8791e-11 5.2395e-13
    L2 1.2122e-4 1.2886e-6 1.6430e-11 9.8481e-14
    1.8 L 4.6023e-4 3.7309e-6 1.9333e-9 5.5668e-13
    L2 1.7954e-4 9.8998e-7 9.6276e-10 2.8539e-13
    1.9 L 9.6641e-4 2.9868e-5 9.7490e-8 5.9616e-11
    L2 4.9498e-4 1.3332e-5 4.4653e-8 2.7277e-11

     | Show Table
    DownLoad: CSV

    Example 2. Consider the following time-fractional Cattaneo equation with the exact solution u(x,t)=etsin(πx):

    {u(x,t)t+γu(x,t)tγ=2u(x,t)x2+f(x,t),0<x<1,0t1,1<γ<2,u(0,t)=0,u(1,t)=0,u(x,0)=sin(πx),ut|t=0=sin(πx), (7.2)

    where f(x,t)=etsin(πx)+12γ11σ(eteσt)sin(πx)+etπ2sin(πx).

    Figure 2 shows the numerical solution and pointwise errors on [0,1]×[0,1] with γ=1.9, k=5 and M=5. For various γ, Table 2 gives the L, L2 errors with different k and M. In Figure 3, we depict the logarithmic plot for the maximal absolute errors L and mean root square error L2 at selected points with k=4, γ=1.9 and M=3 through 9.

    Figure 2.  The plot of numerical solution and pointwise errors for Example 2 with γ=1.9, k=5 and M=5.
    Table 2.  Error estimate for Example 2 with different k and M.
    γ norm k=3,M=3 k=4,M=4 k=5,M=5 k=6,M=6
    1.2 L 1.0143e-4 8.0837e-7 3.3529e-11 6.0396e-14
    L2 3.6737e-5 2.3647e-7 8.8511e-12 1.9451e-14
    1.4 L 1.0483e-4 7.6592e-7 3.1084e-11 9.9476e-14
    L2 4.0664e-5 2.2115e-7 8.7714e-12 2.7726e-14
    1.6 L 1.6201e-4 5.4781e-7 1.2204e-10 1.6875e-13
    L2 8.8981e-5 1.3852e-7 6.4412e-11 3.6683e-14
    1.8 L 0.0016 1.1935e-5 1.5296e-8 3.9275e-12
    L2 8.7472e-4 6.2400e-6 8.0685e-9 2.0768e-12
    1.9 L 0.0083 2.5393e-4 8.2131e-7 5.0065e-10
    L2 0.0042 1.2783e-4 4.1361e-7 2.5214e-10

     | Show Table
    DownLoad: CSV
    Figure 3.  The logarithmic plot for L and L2 error for Example 2.

    Example 3. Consider the following time-fractional Cattaneo equation with the exact solution u(x,t)=etx2(1x)2:

    {u(x,t)t+γu(x,t)tγ=2u(x,t)x2+f(x,t),0<x<1,0t1,1<γ<2,u(0,t)=0,u(1,t)=0,u(x,0)=x2(1x)2,ut|t=0=x2(1x)2, (7.3)

    where f(x,t)=etx2(1x)2+12γ11σ(eteσt)x2(1x)2et(12x212x+2).

    Figure 4 shows the numerical solution and pointwise errors on [0,1]×[0,1] with γ=1.7, k=5 and M=5. For various γ, Table 3 gives the L, L2 errors with different k and M. In Figure 5, we depict the logarithmic plot for the maximal absolute errors L and mean root square error L2 at selected points with k=3, γ=1.7 and M=3 through 9.

    Figure 4.  The plot of numerical solution and pointwise errors for Example 3 with γ=1.7, k=5 and M=5.
    Table 3.  Error estimate for Example 3 with different k and M.
    γ norm k=3,M=3 k=4,M=4 k=5,M=5 k=6,M=6
    1.2 L 2.0143e-6 3.2792e-9 8.3658e-13 5.7454e-15
    L2 1.1090e-6 1.8354e-9 4.8755e-13 8.2946e-16
    1.4 L 2.4907e-6 2.6650e-9 9.5716e-13 1.2768e-15
    L2 1.4089e-6 1.5146e-9 5.5222e-13 2.1428e-16
    1.6 L 7.6440e-6 1.0909e-8 7.1870e-12 5.4678e-15
    L2 4.2624e-6 6.0637e-9 4.0095e-12 9.1325e-16
    1.8 L 9.1918e-5 6.7269e-7 8.5751e-10 2.1959e-13
    L2 4.8255e-5 3.5286e-7 4.4978e-10 1.1499e-13
    1.9 L 4.6337e-4 1.4125e-5 4.5675e-8 2.7842e-11
    L2 2.3351e-4 7.1249e-6 2.3041e-8 1.4045e-11

     | Show Table
    DownLoad: CSV
    Figure 5.  The logarithmic plot for L and L2 error for Example 3.

    Example 4. Consider the following time-fractional Cattaneo equation with the exact solution u(x,t)=et(1x)x163:

    {u(x,t)t+γu(x,t)tγ=2u(x,t)x2+f(x,t),0<x<1,0t1,1<γ<2,u(0,t)=0,u(1,t)=0,u(x,0)=(1x)x163,ut|t=0=(1x)x163, (7.4)

    where f(x,t)=et(1x)x163+12γ11σ(eteσt)(1x)x163+et(3049x1332089x103).

    Figure 6 displays the numerical solution and pointwise errors on [0,1]×[0,1] with γ=1.5, k=5 and M=5. In Table 4, we list the L, L2 errors with different k, M and γ.

    Figure 6.  The plot of numerical solution and pointwise errors for Example 4 with γ=1.5, k=5 and M=5.
    Table 4.  Error estimate for Example 4 with different k and M.
    γ norm k=3,M=3 k=4,M=4 k=5,M=5 k=6,M=6
    1.2 L 9.4908e-5 9.8307e-7 8.7095e-11 4.4699e-13
    L2 2.6285e-5 2.8646e-7 2.0943e-11 9.9711e-14
    1.4 L 9.1462e-5 9.4324e-7 8.6046e-11 4.4436e-13
    L2 2.5169e-5 2.7294e-7 2.0245e-11 9.7045e-14
    1.6 L 8.8202e-5 8.8560e-7 8.4203e-11 4.4144e-13
    L2 2.4450e-5 2.5215e-7 1.8668e-11 9.3449e-14
    1.8 L 9.8768e-5 6.7160e-7 4.4743e-10 4.4437e-13
    L2 3.8992e-5 1.9644e-7 2.1858e-10 1.2035e-13
    1.9 L 2.5128e-4 7.7518e-6 2.5265e-8 1.5449e-11
    L2 1.2672e-4 3.5255e-6 1.1706e-8 7.1740e-12

     | Show Table
    DownLoad: CSV

    Example 5. Consider the following time-fractional Cattaneo equation with the exact solution u(x,t)=et(sin(πx))2:

    {u(x,t)t+γu(x,t)tγ=2u(x,t)x2+f(x,t),0<x<1,0t1,1<γ<2,u(0,t)=0,u(1,t)=0,u(x,0)=(sin(πx))2,ut|t=0=(sin(πx))2, (7.5)

    where f(x,t)=et((sin(πx))2+12γ11σ(eteσt)(sin(πx))2et2π2(cos(πx))2.

    In Figure 7, we plot the approximate solution and pointwise errors on [0,1]×[0,1] when γ=1.9, k=5 and M=5. For various γ, Table 5 gives the L, L2 errors with different k and M. In Figure 8, we depict the logarithmic plot for the maximal absolute errors L and mean root square error L2 at selected points with k=4, γ=1.9 and M=3 through 9.

    Figure 7.  The plot of numerical solution and pointwise errors for Example 5 with γ=1.9, k=5 and M=5.
    Table 5.  Error estimate for Example 5 with different k and M.
    γ norm k=3,M=3 k=4,M=4 k=5,M=5 k=6,M=6
    1.2 L 0.0014 1.2982e-5 3.2843e-9 6.0574e-12
    L2 3.0838e-4 2.9801e-6 7.6865e-10 1.4442e-12
    1.4 L 0.0014 1.2562e-5 3.1775e-9 5.9392e-12
    L2 2.9784e-4 2.8451e-6 7.3457e-10 1.3902e-12
    1.6 L 0.0014 1.1862e-5 2.9698e-9 5.7323e-12
    L2 3.1058e-4 2.5919e-6 6.5691e-10 1.3124e-12
    1.8 L 0.0018 9.3156e-6 1.2832e-8 6.6569e-12
    L2 8.8981e-4 4.5346e-6 6.5375e-9 2.5842e-12
    1.9 L 0.0072 2.1421e-4 6.9554e-7 4.2455e-10
    L2 0.0037 1.0762e-4 3.5099e-7 2.1461e-10

     | Show Table
    DownLoad: CSV
    Figure 8.  The logarithmic plot for L and L2 error for Example 5.

    Example 6. Consider the following nonlinear time-fractional Cattaneo equation with the exact solution u(x,t)=etx2sin(πx):

    {u(x,t)t+1.5u(x,t)t1.5=2u(x,t)x2+u(1u)+f(x,t),0<x<1,0t1,u(0,t)=0,u(1,t)=0,u(x,0)=x2sin(πx),ut|t=0=x2sin(πx), (7.6)

    where f(x,t)=x2sin(πx)2γ11σ(eteσt)(2sin(πx)+4πxcos(πx)π2x2sin(πx))et+(etx2sin(πx))2. In [36], the authors solved the problem using shifted Chebyshev polynomial collocation method and Caputo-Fabrizio operational matrix of derivative. To compare our numerical findings with those results in [36], we list absolute errors at different spatial points at time t=0.1, and for different value of M with k=1 in Table 6. It can be seen that we obtain more accurate numerical solutions with the same parameters. For k=2 and M=6, the numerical solution and pointwise errors on [0,1]×[0,1] are given in Figure 9. In Figure 10, we depict the logarithmic plot for the maximal absolute errors L and root mean square error L2 at selected points with k=2 and M=3 through 9.

    Table 6.  Absolute errors for M=4 and M=6 in Example 6.
    Method in [36] Present method
    x M=4 M=6 M=4 M=6
    0.2 6.7e-3 3.7e-4 1.7e-5 2.4e-7
    0.4 5.2e-3 8.2e-4 1.0e-5 1.0e-7
    0.6 1.6e-3 3.6e-4 1.2e-5 1.1e-7
    0.8 9.1e-3 4.2e-4 1.9e-5 1.7e-7

     | Show Table
    DownLoad: CSV
    Figure 9.  The plot of numerical solution and pointwise errors for Example 6 with k=2 and M=6.
    Figure 10.  The logarithmic plot for L and L2 error for Example 6.

    In this work, a numerical algorithmn based on orthonormal Euler wavelets together with collocation method is proposed for the time-fractional Cattaneo equation with Caputo-Fabrizio derivative. The orthonormal Euler wavelets are constructed by applying the Gram-Schmidt orthonormalization process on sets of Euler polynomials. The convergence analysis and error estimate of the orthonormal Euler wavelets expansion are investigated. The exact formula of the Caputo-Fabrizio fractional integral of orthogonal Euler wavelets is obtained for the first time. The main characteristic behind the method is that the problem under consideration is converted into a system of algebraic equations, which greatly simplifies the problem. The effectiveness of the proposed method for time-fractional Cattaneo equation with Caputo-Fabrizio derivative is verified by the graphical and tabular demonstrations. It can be seen from the Tables 15 that maximum absolute error L and root mean square error L2 become smaller and smaller with increasing k and M. One can see from the Figures 3, 5, 8 and 10 that L and L2 errors show exponential decay in the coordinate, the error variations are essentially linear versus the polynomial degrees. In addition, we also compare our results in Example 6 with the results of shifted Chebyshev polynomial collocation method. We find that Euler wavelets collocation method in this work provides much better results than the aforementioned numerical method. The outcomes show that the proposed method is suitable for solving time-fractional Cattaneo equations having Caputo-Fabrizio derivative.

    In our future work, we will consider 2D and 3D linear and nonlinear fractional partial differential equations involving Caputo-Fabrizio derivative, such as fractional KdV, KdV-Burgers equations, time distributed-order differential equations.

    This work was supported by the National Natural Science Foundation of China (Grant No. 41962019), the Natural Science Foundation of Jiangxi Province of China (Grant No. 2020BABL201006) and the Doctoral Scientific Research Foundation of East China University of Technology (Grant No. DHBK2019213).

    The authors declare they have no competing interests.

    In this section, we prove Theorem 5.2.

    Proof. According to the definition of orthonormal Euler wavelets ψn,m(x), they are given as follows:

    ψn,m(x)={2k12OEm(2k1xn+1),n12k1x<n2k1,0,otherwise. (A.1)

    Using the unit step function, orthonormal Euler wavelets can be expressed as

    ψn,m(x)=μn12k1(x)2k12OEm(2k1xn+1)μn2k1(x)2k12OEm(2k1xn+1),

    where μc(x) is the unit step function defined as

    μc(x)={1,xc,0,x<c. (A.2)

    Recalling the definition of Caputo-Fabrizio integral of order η=n+β:

    CFJn+βaf(x)=1M(β)n!xa(xs)n1[β(xs)+n(1β)]f(s)ds, (A.3)

    it can be further simplified as follows using the Cauchy formula Inf(x)=1Γ(n)xa(xs)n1f(s)ds and the relation Γ(n)=(n1)!,

    CFJn+βaf(x)=1βM(β)Inf(x)+βM(β)In+1f(x). (A.4)

    Note that Inf(x) is the special case of Riemann-Liuville integral Iαf(x)=1Γ(α)x0(xs)α1f(s)ds. Furthermore, by the aid of convolution operation, Iαf(x) can be express as

    Iαψn,m(x)=1Γ(α)xα1ψn,m(x). (A.5)

    By employing the Laplace transform, we obtain

    L{Iαψn,m(x)}=L{1Γ(α)xα1}L{ψn,m(x)}=1sαL{ψn,m(x)}. (A.6)

    Next, we focus on the derivation of L{ψn,m(x)}. By using the delay property of Laplace transform L{μc(x)f(x)}=ecsL{f(x+c)}, we have

    L{ψn,m(x)}=2k12L{μn12k1(x)OEm(2k1xn+1)}2k12L{μn2k1(x)OEm(2k1xn+1)}=en12k1s2k12L{OEm(2k1(x+n12k1)n+1)}en2k1s2k12L{OEm(2k1(x+n2k1)n+1)}=en12k1s2k12L{OEm(2k1x)}en2k1s2k12L{OEm(2k1x+1)}. (A.7)

    From Eq (3.3), we have

    OEm(2k1x)=2m+1mi=0(1)miCimCim+i2i(k1)xi, (A.8)

    and

    OEm(2k1x+1)=2m+1mi=0(1)miCimCim+iir=0Cri2r(k1)xr. (A.9)

    By substituting Eqs (A.8) and (A.9) into Eq (A.7), we obtain

    L{ψn,m(x)}=en12k1s2k122m+1mi=0(1)miCimCim+i2i(k1)L{xi}en2k1s2k122m+1mi=0(1)miCimCim+iir=0Cri2r(k1)L{xr}.

    Since L{xr}=Γ(r+1)sr+1, we have

    L{ψn,m(x)}=2k122m+1mi=0(1)miCimCim+i2i(k1)Γ(i+1)en12k1ssi+12k122m+1mi=0(1)miCimCim+iir=0Cri2r(k1)Γ(r+1)en2k1ssr+1.

    From the definition of Iαψn,m(x), we get

    L{Iαψn,m(x)}=L{1Γ(α)xα1}L{ψn,m(x)}=1sαL{ψn,m(x)}=2k122m+1mi=0(1)miCimCim+i2i(k1)Γ(i+1)en12k1ssi+α+12k122m+1mi=0(1)miCimCim+iir=0Cri2r(k1)Γ(r+1)en2k1ssr+α+1.

    By taking the inverse Laplace transform and using the delay property of Laplace transform power function L{(xa)nμ(xa)}=easΓ(n+1)sa+1, we obtain

    Iαψn,m(x)=2k122m+1mi=0(1)miCimCim+i2i(k1)Γ(i+1)Γ(i+α+1)(xn12k1)i+αμn12k1(x)2k122m+1mi=0(1)miCimCim+iir=0Cri2r(k1)Γ(r+1)Γ(r+α+1)(xn2k1)r+αμn2k1(x).

    The theorem then follows.



    [1] N. H. Tuan, H. Mohammadi, S. Rezapour, A mathematical model for COVID-19 transmission by using the Caputo fractional derivative, Chaos Solitons Fract., 140 (2020), 110107. https://doi.org/10.1016/j.chaos.2020.110107 doi: 10.1016/j.chaos.2020.110107
    [2] S. Kumar, R. Kumar, J. Singh, K. S. Nisar, D. Kumar, An efficient numerical scheme for fractional model of HIV-1 infection of CD4+ T-cells with the effect of antiviral drug therapy, Alex. Eng. J., 59 (2020), 2053–2064. https://doi.org/10.1016/j.aej.2019.12.046 doi: 10.1016/j.aej.2019.12.046
    [3] R. Almeida, A. M. C. B. da Cruz, N. Martins, M. T. T. Monteiro, An epidemiological MSEIR model described by the Caputo fractional derivative, Int. J. Dyn. Control, 7 (2019), 776–784. https://doi.org/10.1007/s40435-018-0492-1 doi: 10.1007/s40435-018-0492-1
    [4] P. Veeresha, D. G. Prakasha, H. M. Baskonus, New numerical surfaces to the mathematical model of cancer chemotherapy effect in Caputo fractional derivatives, Chaos, 29 (2019), 013119. https://doi.org/10.1063/1.5074099 doi: 10.1063/1.5074099
    [5] R. Ali, M. I. Asjad, A. Aldalbahi, M. Rahimi-Gorji, M. Rahaman, Convective flow of a Maxwell hybrid nanofluid due to pressure gradient in a channel, J. Therm. Anal. Calorim., 143 (2021), 1319–1329. https://doi.org/10.1007/s10973-020-10304-x doi: 10.1007/s10973-020-10304-x
    [6] M. Caputo, M. Fabrizio, A new definition of fractional derivative without singular kernel, Progr. Fract. Differ. Appl., 1 (2015), 73–85.
    [7] D. Baleanu, A. Jajarmi, H. Mohammadi, S. Rezapour, A new study on the mathematical modelling of human liver with Caputo-Fabrizio fractional derivative, Chaos Solitons Fract., 134 (2020), 109705. https://doi.org/10.1016/j.chaos.2020.109705 doi: 10.1016/j.chaos.2020.109705
    [8] A. Atangana, B. S. T. Alkahtani, New model of groundwater flowing within a confine aquifer: application of Caputo-Fabrizio derivative, Arab. J. Geosci. 9 (2016), 1–6. https://doi.org/10.1007/s12517-015-2060-8 doi: 10.1007/s12517-015-2060-8
    [9] A. Atangana, D. Baleanu, Caputo-Fabrizio derivative applied to groundwater flow within confined aquifer, J. Eng. Mech., 143 (2017), D4016005. https://doi.org/10.1061/(ASCE)EM.1943-7889.0001091 doi: 10.1061/(ASCE)EM.1943-7889.0001091
    [10] A. Alshabanat, M. Jleli, S. Kumar, B. Samet, Generalization of Caputo-Fabrizio fractional derivative and applications to electrical circuits, Front. Phys., 8 (2020), 64. https://doi.org/doi:10.3389/fphy.2020.00064 doi: 10.3389/fphy.2020.00064
    [11] M. ur Rahman, S. Ahmad, R. T. Matoog, N. A. Alshehri, T. Khan, Study on the mathematical modelling of COVID-19 with Caputo-Fabrizio operator, Chaos Solitons Fract., 150 (2021), 111121. https://doi.org/10.1016/j.chaos.2021.111121 doi: 10.1016/j.chaos.2021.111121
    [12] E. F. Doungmo Goufo, Application of the Caputo-Fabrizio fractional derivative without singular kernel to Korteweg-de Vries-Burgers equation, Math. Model. Anal., 21 (2016), 188–198. https://doi.org/10.3846/13926292.2016.1145607 doi: 10.3846/13926292.2016.1145607
    [13] F. Gao, X. J. Yang, Fractional Maxwell fluid with fractional derivative without singular kernel, Therm. Sci., 20 (2016), 871–877. https://doi.org/10.2298/TSCI16S3871G doi: 10.2298/TSCI16S3871G
    [14] O. J. Peter, A. Yusuf, K. Oshinubi, F. A. Oguntolu, J. O. Lawal, A. I. Abioye, et al., Fractional order of pneumococcal pneumonia infection model with Caputo Fabrizio operator, Results Phys., 29 (2021), 104581. https://doi.org/10.1016/j.rinp.2021.104581 doi: 10.1016/j.rinp.2021.104581
    [15] O. J. Peter, Transmission dynamics of fractional order brucellosis model using Caputo-Fabrizio operator, Int. J. Differ. Equ., 2020 (2020), 2791380. https://doi.org/10.1155/2020/2791380 doi: 10.1155/2020/2791380
    [16] E. Awad, On the time-fractional Cattaneo equation of distributed order, Phys. A, 518 (2019), 210–233. https://doi.org/10.1016/j.physa.2018.12.005 doi: 10.1016/j.physa.2018.12.005
    [17] A. Compte, R. Metzler, The generalized Cattaneo equation for the description of anomalous transport processes, J. Phys. A Math. Gen., 30 (1997), 7277.
    [18] K. D. Lewandowska, T. Kosztolowicz, Application of generalized Cattaneo equation to model subdiffusion impedance, Acta Phys. Pol. B, 39 (2008), 1211–1220.
    [19] H. R Ghazizadeh, M. Maerefat, A. Azimi, Explicit and implicit finite difference schemes for fractional Cattaneo equation, J. Comput. Phys., 229 (2010), 7042–7057. https://doi.org/10.1016/j.jcp.2010.05.039 doi: 10.1016/j.jcp.2010.05.039
    [20] S. W. Vong, H. K. Pang, X. Q. Jin, A high-order difference scheme for the generalized Cattaneo equation, East Asian J. Appl. Math., 2 (2012), 170–184. https://doi.org/10.4208/eajam.110312.240412a doi: 10.4208/eajam.110312.240412a
    [21] X. Zhao, Z. Z. Sun, Compact Crank-Nicolson schemes for a class of fractional Cattaneo equation in inhomogeneous medium, J. Sci. Comput., 62 (2015), 747–771. https://doi.org/10.1007/s10915-014-9874-5 doi: 10.1007/s10915-014-9874-5
    [22] J. C. Ren, G. H. Gao, Efficient and stable numerical methods for the two-dimensional fractional Cattaneo equation, Numer. Algorithms, 69 (2015), 795–818. https://doi.org/10.1007/s11075-014-9926-9 doi: 10.1007/s11075-014-9926-9
    [23] L. L. Wei, Analysis of a new finite difference/local discontinuous Galerkin method for the fractional Cattaneo equation, Numer. Algorithms, 77 (2018), 675–690. https://doi.org/10.1007/s11075-017-0334-9 doi: 10.1007/s11075-017-0334-9
    [24] Y. M. Wang, A Crank-Nicolson-type compact difference method and its extrapolation for time fractional Cattaneo convection-diffusion equations with smooth solutions, Numer. Algorithms, 81 (2019), 489–527. https://doi.org/10.1007/s11075-018-0558-3 doi: 10.1007/s11075-018-0558-3
    [25] L. J. Nong, Q. Yi, J. X. Cao, A. Chen, Fast compact difference scheme for solving the two-dimensional time-fractional Cattaneo equation, Fractal Fract., 6 (2022), 438. https://doi.org/10.3390/fractalfract6080438 doi: 10.3390/fractalfract6080438
    [26] Z. G. Liu, A. J. Cheng, X. L. Li, A second order Crank-Nicolson scheme for fractional Cattaneo equation based on new fractional derivative, Appl. Math. Comput., 311 (2017), 361–374. https://doi.org/10.1016/j.amc.2017.05.032 doi: 10.1016/j.amc.2017.05.032
    [27] M. Taghipour, H. Aminikhah, A θ-finite difference scheme based on cubic B-spline quasi-interpolation for the time fractional Cattaneo equation with Caputo-Fabrizio operator, J. Differ. Equ. Appl., 27 (2021), 712–738. https://doi.org/10.1080/10236198.2021.1935909 doi: 10.1080/10236198.2021.1935909
    [28] M. Yaseen, Q. U. Nisa Arif, R. George, S. Khan, Comparative numerical study of spline-based numerical techniques for time fractional Cattaneo equation in the sense of Caputo-Fabrizio, Fractal Fract., 6 (2022), 50. https://doi.org/10.3390/fractalfract6020050 doi: 10.3390/fractalfract6020050
    [29] H. N. Li, S. J. Lü, T. Xu, A fully discrete spectral method for fractional Cattaneo equation based on Caputo-Fabrizo derivative, Numer. Methods Partial Differ. Equ., 35 (2019), 936–954. https://doi.org/10.1002/num.22332 doi: 10.1002/num.22332
    [30] E. Hesameddini, M. Shahbazi, Two-dimensional shifted Legendre polynomials operational matrix method for solving the two-dimensional integral equations of fractional order, Appl. Math. Comput., 322 (2018), 40–54. https://doi.org10.1016/j.amc.2017.11.024 doi: 10.1016/j.amc.2017.11.024
    [31] H. Singh, C. S. Singh, A reliable method based on second kind Chebyshev polynomial for the fractional model of Bloch equation, Alex. Eng. J., 57 (2018), 1425–1432. https://doi.org/10.1016/j.aej.2017.07.002 doi: 10.1016/j.aej.2017.07.002
    [32] S. Jaiswal, S. Das, Numerical solution of linear/nonlinear fractional order differential equations using Jacobi operational matrix, Int. J. Appl. Comput. Math., 5 (2019), 1–21. https://doi.org/10.1007/s40819-019-0625-z doi: 10.1007/s40819-019-0625-z
    [33] A. Isah, C. Phang, New operational matrix of derivative for solving non-linear fractional differential equations via Genocchi polynomials, J. King Saud Univ. Sci., 31 (2019), 1–7. https://doi.org/10.1016/j.jksus.2017.02.001 doi: 10.1016/j.jksus.2017.02.001
    [34] S. Sadeghi Roshan, H. Jafari, D. Baleanu, Solving FDEs with Caputo-Fabrizio derivative by operational matrix based on Genocchi polynomials, Math. Methods Appl. Sci., 41 (2018), 9134–9141. https://doi.org/10.1002/mma.5098 doi: 10.1002/mma.5098
    [35] J. R. Loh, A. Isah, C. Phang, Y. T. Toh, On the new properties of Caputo-Fabrizio operator and its application in deriving shifted Legendre operational matrix, Appl. Numer. Math., 132 (2018), 138–153. https://doi.org/10.1016/j.apnum.2018.05.016 doi: 10.1016/j.apnum.2018.05.016
    [36] S. Kumar, J. F. Gómez Aguilar, P. Pandey, Numerical solutions for the reaction-diffusion, diffusion-wave, and Cattaneo equations using a new operational matrix for the Caputo-Fabrizio derivative, Math. Methods Appl. Sci., 43 (2020), 8595–8607. https://doi.org/10.1002/mma.6517 doi: 10.1002/mma.6517
    [37] B. Yuttanan, M. Razzaghi, Legendre wavelets approach for numerical solutions of distributed order fractional differential equations, Appl. Math. Model., 70 (2019), 350–364. https://doi.org/10.1016/j.apm.2019.01.013 doi: 10.1016/j.apm.2019.01.013
    [38] N. Kumar, M. Mehra, Legendre wavelet collocation method for fractional optimal control problems with fractional Bolza cost, Numer. Methods Partial Differ. Equ., 37 (2021), 1693–1724. https://doi.org/10.1002/num.22604 doi: 10.1002/num.22604
    [39] G. Esra Köse, Ö. Oruç A. Esen, An application of Chebyshev wavelet method for the nonlinear time fractional Schröinger equation, Math. Methods Appl. Sci., 45 (2022), 6635–6649. https://doi.org/10.1002/mma.8196 doi: 10.1002/mma.8196
    [40] J. Shahni, R. Singh, A fast numerical algorithm based on Chebyshev-wavelet technique for solving Thomas-Fermi type equation, Eng. Comput., 2021, 1–14. https://doi.org/10.1007/s00366-021-01476-7 doi: 10.1007/s00366-021-01476-7
    [41] A. Secer, M. Cinar, A Jacobi wavelet collocation method for fractional Fisher's equation in time, Therm. Sci., 24 (2020), 119–129. https://doi.org/10.2298/tsci20s1119s doi: 10.2298/tsci20s1119s
    [42] M. R. Eslahchi, M. Kavoosi, The use of Jacobi wavelets for constrained approximation of rational Béier curves, Comput. Appl. Math., 37 (2018), 3951–3966. https://doi.org/10.1007/s40314-017-0552-8 doi: 10.1007/s40314-017-0552-8
    [43] A. T. Turan Dincel, Solution to fractional-order Riccati differential equations using Euler wavelet method, Sci. Iran., 26 (2019), 1608–1616. https://doi.org/10.24200/SCI.2018.51246.2084 doi: 10.24200/SCI.2018.51246.2084
    [44] S. Behera, S. Saha Ray, Euler wavelets method for solving fractional-order linear Volterra-Fredholm integro-differential equations with weakly singular kernels, Comput. Appl. Math., 40 (2021), 1–30. https://doi.org/10.1007/s40314-021-01565-9 doi: 10.1007/s40314-021-01565-9
    [45] R. Aruldoss, R. A. Devi, P. M. Krishna, An expeditious wavelet-based numerical scheme for solving fractional differential equations, Comput. Appl. Math., 40 (2021), 1–14. https://doi.org/10.1007/s40314-020-01387-1 doi: 10.1007/s40314-020-01387-1
    [46] J. L. Schiff, The Laplace transform: theory and applications, Undergraduate Texts in Mathematics, New York: Springer, 1999.
    [47] M. Moumen Bekkouche, H. Guebbai, M. Kurulay, S. Benmahmoud, A new fractional integral associated with the Caputo-Fabrizio fractional derivatives, Rend. Circ. Mat. Palermo (2), 70 (2021), 1277–1288. https://doi.org/10.1007/s12215-020-00557-8 doi: 10.1007/s12215-020-00557-8
    [48] I. Podlubny, Fractional differential equations, San Diego: Academic Press, 1999.
    [49] Y. He, Some new results on products of Apostol-Bernoulli and Apostol-Euler polynomials, J. Math. Anal. Appl., 431 (2015), 34–46. https://doi.org/10.1016/j.jmaa.2015.05.055 doi: 10.1016/j.jmaa.2015.05.055
    [50] F. Mohammadi, A. Ciancio, Wavelet-based numerical method for solving fractional integro-differential equation with a weakly singular kernel, Wavelet Linear Algebra, 4 (2017), 53–73. https://doi.org/10.22072/wala.2017.52567.1091 doi: 10.22072/wala.2017.52567.1091
    [51] M. M. Bekkouche, H. Guebbai, M. Kurulay, On the solvability fractional of a boundary value problem with new fractional integral, J. Appl. Math. Comput., 64 (2020), 551–564. https://doi.org/10.1007/s12190-020-01368-x doi: 10.1007/s12190-020-01368-x
  • This article has been cited by:

    1. Fengying Zhou, Jiakun Zhang, 2D scale-3 fractional Euler wavelets optimization algorithm for fractional-order differential equations, 2024, 83, 18777503, 102459, 10.1016/j.jocs.2024.102459
    2. Akanksha Singh, Ankur Kanaujiya, Jugal Mohapatra, Euler wavelets method for optimal control problems of fractional integro-differential equations, 2025, 454, 03770427, 116178, 10.1016/j.cam.2024.116178
    3. Zafar Hayat Khan, Oluwole Daniel Makinde, Alexander Trounev, Waqar Ahmed Khan, Rashid Ahmad, Entropy generation and heat transfer in Time-Fractional mixed convection of nanofluids in Darcy-Forchheimer porous channel, 2024, 60, 22150986, 101908, 10.1016/j.jestch.2024.101908
  • Reader Comments
  • © 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(1712) PDF downloads(131) Cited by(3)

Figures and Tables

Figures(10)  /  Tables(6)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog