Abstract
Bivariate Mittag-Leffler (ML) functions are a substantial generalization of the univariate ML functions, which are widely recognized for their significance in fractional calculus. In the present paper, our initial focus is to investigate the fractional calculus properties of the integral and derivative operators with kernels including the Bivariate ML functions. Further, certain fractional Cauchy-type problems including these operators are considered. Also the numerical approximations of the Caputo type derivative operator are investigated. The theoretical results are justified by applications on examples. Furthermore, the theory of applying the same operators with respect to arbitrary monotonic functions is analyzed in this research.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
Fractional calculus is of great importance in applied mathematics and mathematical analysis, with substantial connections to ML functions. The function is defined by the following series when the real part of \(\alpha \) is strictly positive, [1]
where \(\Gamma \) is the gamma function. For \(\alpha =1\) the ML function reduces to the exponential function and ML functions were used to define the solutions to fractional differential equations. It is well known that fractional differential equations are among the strongest tools of mathematical modeling and are successfully employed to model complex physical and biological phenomena. Such as impulsive neutral Hilfer fractional evolation equations [2, 3], fractional order Zika virus model [4] and fractional Lagevin equation [5].
The bivariate ML function is a generalization of the classical ML function applied to two variables and has been employed to solve fundamental fractional differential equations involving two independent fractional orders. Recent studies include but not limited with the coupled-Laplacian fractional differential equations with nonlinear boundary conditions [6], singular multi-order fractional differential equations of Lane–Emden type [7]. In last quarter century there has been a growing interest in studying different variants of bivariate ML functions and fractional calculus operators with these functions in the kernel (see [8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26]).
This work focuses on the bivariate ML functions \(E_{\alpha ,\beta ,\gamma }^{\delta }(x,y)\), introduced in [13] as a double series:
where \( \alpha , \beta ,\gamma ,\delta \) are complex parameters with \( Re(\alpha )>0\) and \( Re(\beta )>0\), and the notation \((a)_s\) is used for the Pochhammer symbol \(\frac{\Gamma (a+s)}{\Gamma (a)}\). The given series converges absolutely and locally uniformly for \( Re(\alpha )>0\) and \( Re(\beta )>0\). The authers of [13] gave the proof by using the convergence conditions studied in [27] for the generalised Lauricella series in two variables. If all parameters are 1 in Eq. (1) we recover the double exponential function which is the natural analogue of the fact that the ML function reduces to exponential function.
For \(x>a\) the following fractional integral operator was also defined in [13], with the kernel containing (1)
where \(\psi (x) \in L^1(a,b)\), being the space of absolutely integrable functions. As shown in Theorem 8 of [13], the integral operator is bounded on the space \(L^1(a,b)\). The extra restriction \(Re(\gamma )>0\) is to avoid non-integrable singularity at the end point \(t=x\). In the case \(\gamma =0\), the fractional integral operator \({}_{{a}}I_{\alpha ,\beta ,\gamma }^{\delta ,w_1,w_2}\) reduces to the Riemann - Liouville fractional integral operator which is written as follows:
where \(x>a,Re(\mu )>0.\)
Additionally the fractional derivatives are given by,
Distinguishing itself from the Riemann–Liouville fractional derivative, the Caputo fractional derivative is defined based on the Riemann–Liouville fractional integral,
In [13], the function in (1) and the fractional integral operator (2) were introduced and their properties were studied, including the crucial semigroup property of the fractional integral operator for any summable function \(\psi \in L^1(a,b)\):
By using the semigroup property the corresponding left inverse operator to the fractional integral operator (2) was defined in [13] as follows,
and the corresponding Caputo fractional derivative was defined by,
In the present study, we undertake additional analysis of the integral operator (2) and derivative operators (4), (5), naturally extending the results [13]. In Sect. 2 we give Laplace transforms of the operators and the product rule for the operator (2). In Sect. 3 we investigate the eigenvalue problem for the Caputo version derivative and the solution of nonhomogenuous Cauchy problem involving Riemann–Liouville derivative and integral operator (2). Section 4 is devoted to the numerical approximation of the Caputo type derivative operator through the use of linear and quadratic univariate Lagrange interpolation. In Sect. 5, we extend and generalize the theory by using fractional calculus operators with respect to functions. In Sect. 6, we conclude the paper.
2 Fractional-calculus of the operators with bivariate ML kernel
In this section, we remember some properties of the fractional calculus operators from [13] and we give some new results for these operators. The following theoretical results are crucial for the derivation of the theoretical findings in the sebsequent sections.
For \( \alpha , \beta ,\gamma , \delta , w_1, w_2 \in \mathbb {C}, Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0\) and for any function \(\psi \in L^1( a,b)\), the fractional integral operator (2) can be written as [13, Theorem 7]
where the series on the right hand side is locally uniformly convergent. To demonstrate the effectiveness of (6) we give the following resullts which show the compositon of the fractional operators \({}^{{RL}}_{{a}}I_{x}^{\sigma }\) and \({}^{{RL}}_{{a}}D_{x}^{\sigma }\) with the operator (2). The relation [13, Corollary 4]
holds true for any summable function \(\psi \in L^1[a,b]\) and any \( \alpha , \beta ,\gamma , \delta , w_1, w_2 \in \mathbb {C}\), \(Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0.\) Also the relation
holds true for any continuous function \(\psi \in C[a,b]\) and any \( \alpha , \beta ,\gamma , \delta , w_1, w_2 \in \mathbb {C}\), \(Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0.\)
In order to highlight the usefullness of the series formula in (6), we will now illustrate a straightforward approach for computing the outcome of applying the integral operator (2) to an elementary power function. This method offers a simplified alternative for using the original definition (2).
Proposition 1
The integral operator \({}_{{a}}{I}_{\alpha ,\beta ,\gamma }^{\delta ,w_1,w_2}\) on a power function is given as:
Proof
According to Riemann–Liouville the fractional integral of a power function can be presented as
By combining this established result with the series formula (6), we derive the following relation for \(Re(\delta )>-1\)
\(\square \)
Example 1
In this example we calculate the integral operator \(_{0}I_{\alpha ,\beta ,\gamma }^{\delta ,w_{1},w_{2}}\left[ x^{\sigma }\right] \) given in (7) for \(w_{1}=w_{2}=1,\) over the interval \(\Omega =\left[ 0,1\right] \) at the mesh points
For \(N_{1}=64\) and \(\alpha =0.3,\beta =0.9\) and \(\delta =\sigma =-0.7\) the integral \(_{0}I_{0.3,0.9,\gamma }^{-0.7,1,1}\left[ x^{-0.7}\right] \) is presented in Fig. 1 for various values of \(\gamma \) as 0.1, 0.5, 0.9. Further, the integral \(_{0}I_{0.3,0.9,0.6}^{-0.7,1,1}\left[ x^{\sigma }\right] \) for \(\sigma =0.5,0.9,1.9\) is illustrated in Fig. 2.
Proposition 2
The integral operator \({}_{{a}}{I}_{\alpha ,\beta ,\gamma }^{\delta ,w_1,w_2}\) on an exponential function is given as:
Proof
The fractional integral of an exponential function according to Riemann–Liouville can be presented as
Using this result with the series formula (6), we establish the following for \(Re(\delta )>0\)
\(\square \)
Subsequently, our attention is directed towards the Laplace transform, with specific reference to [13], which briefly discussed the Laplace transform of the bivariate ML function (1). Because the gamma function on the denominator involves both s and r together, the property of formula (1) inables computation of the double Laplace transform with respect to x and y. Instead of taking the double Laplace transform they considered the univariate version of (1) and calculated the Laplace transform with respect to one variable t.
It was obtained in [13] that
where \( Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0,Re(p)>0\) and \(\mathbb {L}[f(x)](p)\) is the univariate Laplace transform operator defined by
Now, we obtain the Laplace transforms of the fractional integral (2), Riemann–Liouville type and Caputo type fractional derivatives corresponding to the bivariate ML functions.
Theorem 3
The Laplace transform of integral operator (2) can be expressed by
where \(Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0\) and p is the Laplace transformed parameter satisfying \(Re(p)>0.\)
Proof
By using series formula (6) we get,
The series manipulation is same as in the proof of Proposition 2 and by using the fact [26] that Laplace transforms of Riemann–Liouville integrals,
we obtain (11). \(\square \)
Theorem 4
The Laplace transform of Riemann–Liouville type fractional derivative operator \({}^{{RL}}_{{0}}{D}_{\alpha ,\beta ,\gamma }^{\delta ,w_1,w_2}\) can be expressed as below
where \(Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0\) and p is the Laplace transform parameter satisfying \(Re(p)>0\) and \(n-1 \ne Re(\gamma )< n\), \(n \in \mathbb {N}.\)
Proof
By using the series representation for the operator \({}^{{RL}}_{{0}}{D}_{\alpha ,\beta ,\gamma }^{\delta ,w_1,w_2}\psi (x)\), we get
Based on the locally uniformly convergent property of the series and applying the known formula for the Laplace transform of the \(n-\)th derivative of the function, we obtain
\(\square \)
Theorem 5
The Laplace transform of the Caputo type fractional derivative \({}^{{C}}_{{0}}{D}_{\alpha ,\beta ,\gamma }^{\delta ,w_1,w_2}\) can be given as follows
where the conditions on the parameters are same as in Theorem 4.
Proof
By using Theroem 3 and applying formula for the Laplace transform of the \(n-\)th derivative of the function, we obtain
which gives the desired result. \(\square \)
In the literature the product rule is extensively studied and generalized to the fractional scenarios [28]. For the integral operator (2) using the series formula (6) we give a version of the product rule.
Theorem 6
Let f and g be complex functions such that f,g and f(x)g(x) are all in the form \(x^\eta \Delta (x)\) with \(Re(\eta )>0\) and \(\Delta \) holomorphic on a complex domain \(U \subset \mathbb {C}\). Under the conditions \( Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0\), the integral operator (2) satisfies the following product rule
Proof
By using series formula (6) with the known results [29, 30] on the product rule for Riemann–Liouville fractional differintegrals we get the following
given (12). \(\square \)
Example 2
To verify the results of Theorem 6, we take the functions \(f(x)=e^{cx}\) and \(g(x)=x\) and the constant of differintegration \(a=i\infty \). Consequently, the outer series in (12) has only two non-trivial terms. Thus;
hence we have computed the integral \({}_{{i\infty }}{I}_{\alpha ,\beta ,\gamma }^{\delta ,w_1,w_2}\big (xe^{cx}\big )\).
The chain rule in classical calculus can be generalized to certain models of fractional calculus. In the following theorem for this particular calculus model we use the proposed series formula (6) to derive a result of the chain rule.
Theorem 7
Let f and g be complex functions such that g is smooth and f(g(x)) is a function of the form \(x^\eta \Delta (x)\) with \(Re(\eta )>0\) and \(\Delta \) holomorphic on a complex domain \(U \subset \mathbb {C}\). Under the conditions \( Re(\beta )>0, Re(\alpha )>0, Re(\kappa )>0\), the integral operator (2) satisfies the following chain rule.
where the summation \((P_1,\dots ,P_m)\) is over the set
Proof
Applying Theorem 6 to the product of \(\big (f(g(x))\big )\) and \(I(x)=1\), where the Riemann–Liouville fractional differintegrals of I are well known, yields (12),
We use the reflection formula in order to eliminate some of the gamma functions in this expression,
By use of classical Faa di Bruno formula we get the result (13). \(\square \)
3 Eigenvalue problem and Cauchy problem for fractional integro-differential equation
Our focus now shifts to the eigenvalue problem corresponding to the Caputo fractional derivative (5). To initiate our investigation, we analyze the following Cauchy problem
where \( \alpha , \beta ,\gamma , \delta \in \mathbb {C}, Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0\) and \(n-1\le Re(\gamma )<n\) \((n \in \mathbb {N})\). \(C_1, \delta _0 \dots ,\delta _{n-1} \in \mathbb {C}\). Applying Laplace transform to both sides of (14), we obtain
This leads to
given that \(\bigg |C_1p^{-\gamma }\bigg (1-\frac{w_1}{s^\alpha }-\frac{w_2}{s^\beta }\bigg )^{-\delta }\bigg | < 1.\) After applying inverse Laplace transform to both sides and making use of Eq. (10), we arrive at
which is the eigenfunction of the \({}^{{C}}_{{0}}{D}_{\alpha ,\beta ,\gamma }^{\delta ,w_1,w_2}\) corresponding to the eigenvalue \(C_1\).
Our attention now turns to the following Cauchy-type problem:
where \(k-1 <\sigma \le k\) (\(k \in \mathbb {N}\)) and \(\delta ,\alpha ,\beta ,\gamma , w_1,w_2 \in \mathbb {R}\) with \(\alpha ,\beta >0\).
Theorem 8
The problem (15) and (16) has solution \(\psi \in L_1(0,\infty )\) as:
Proof
Through the application of the Laplace transform to both sides of Eq. (15), we obtain
Consequently,
Applying inverse Laplace transform yields the result (17). \(\square \)
Example 3
Consider the following nonhomogenous Cauchy-type problem:
where \(k-1 <\sigma \le k\) (\(k \in \mathbb {N}\)) and \(\delta , \alpha ,\beta ,\mu ,\gamma , w_1,w_2 \in \mathbb {R}\) with \(\alpha ,\beta , \mu >0\). Considering (2) and the identity given in Theorem 6 of [13] gives,
From Theorem 8 it follows that the problem (18) has solution belonging to \(L_1(0,\infty )\) given by
4 Approximation of Caputo type derivative
Numerical approximations of Caputo–Prabhakar derivative including 3 parameters ML function was given in [31]. In this section, we give numerical approximations to Caputo operator containing bivariate ML in the kernel.
For any \( t\in \mathbb {R}^+\) and \(\mu \in \mathbb {C}\),
For any \( n \in \mathbb {N}\),
The interval [0, T] is divided into \(N_1\) subintervals, each of length \(\zeta =\frac{T}{N_1}\) with points \(0=t_0<t_1<t_2<\dots <t_{N_1}=T\), where \(t_i=i\zeta \), \(i=0,1,\dots ,N_1\). We approximate the function g(t) by using first degree Lagrange interpolation function. For two points \((t_{i-1},g(t_{i-1}))\) and \((t_{i},f(t_{i}))\) linear interpolation function \(P_1(t)\) is defined as,
Also,
and
where \(\epsilon _i\in (t_{i-1},t_i)\).
For \(0<\gamma <1\) and \(0<t<T\) by choosing \(n=1\), the Caputo-type derivative (5) is defined as,
at points \(t=t_l\), \(l=1,\dots ,N_1\),
where
In the next theorem, we determine error bound for the given approximation.
Theorem 9
Let \(g(t) \in C^2[0,T]\) for any \(0<\gamma <1\), the truncation error \( R(g,\zeta ,\gamma )\) satiesfies the following inequality,
where \(K_1\) is a positive constant given as
Proof
From (23) it follows that
\(\square \)
For the numerical computations we used Matlab and the ML function \(E_{\alpha ,\beta ,\gamma }^{\delta }\) as defined in (1) is approximated by truncating the series with 50 terms. We define the absolute error function
Example 4
We consider the functions \(f\left( t\right) =t^{\mu }\) \(\mu \ge 1\) for \(t\in \Omega =\left[ 0,1\right] \) and the approximation of Caputo type derivative \( \big (_{0}^{C}D_{\alpha ,\beta ,\gamma ,w_{1},w_{2}}^{\delta ,\left( 1,N_{1}\right) }f\big )(t)\) by using linear Lagrange interpolation function \(P_1(t)\) as defined in (21) and the exact derivative \(\big (_{0}^{C}D_{\alpha ,\beta ,\gamma ,w_{1},w_{2}}^{\delta }f\big )(t)\) are considered. We take the set of mesh points \(\Omega ^{\zeta }\) as given in (8), also, \( w_{1}=w_{2}=1\) and \(\alpha =0.3,\beta =0.9\) and \(\gamma =0.6\). Figure 3, presents \(\big (_{0}^{C}D_{0.3,0.9,0.6,1,1}^{-0.5,\left( 1,64\right) }f\big )(t)\) and the exact derivative \(\big ( _{0}^{C}D_{0.3,0.9,0.6,1,1}^{-0.5}f\big )(t)\) denoted by ap and ex respectively for \(\mu =1\). Further, Fig. 4, presents \(\big ( _{0}^{C}D_{0.3,0.9,0.6,1,1}^{-0.5,\left( 1,64\right) }f\big )(t)\) and the exact derivative \(\big (_{0}^{C}D_{0.3,0.9,0.6,1,1}^{-0.5}f\big )(t)\) for \(\mu =4\). It can be viewed from Fig. 3 that the approximation is almost exact in double precision for \(f\left( t\right) =t.\) Furthermore, Fig. 5 shows the error function \(\big ( Er_{0.3,0.9,0.6,1,1}^{-0.5,(1,64)}f\big )(t)\) for \(f\left( t\right) =t^{4}.\)
Now we construct the quadratic Lagrange interpolation polynomial for the nodes \(t_{i-2}\), \(t_{i-1}\) and \(t_{i}\).
Further,
and
where \(\epsilon _i\in (t_{i-2},t_i)\).
For \(1<\gamma <2\) and \(0<t<T\) by choosing \(n=2\), the Caputo type derivative (5) is defined as
at points \(t=t_l=l\frac{T}{N_1}\), \(l=1,\dots ,N_1\),
where
In the next theorem, we determine error bound for the given approximation.
Theorem 10
Let \(g(t) \in C^3[0,T]\) for any \(1<\gamma <2\), the truncation error \( R_2(g,\zeta ,\gamma )\) is given as follows
where \(K_2\) is a positive constant
Proof
\(\square \)
Example 5
We take the function \(f\left( t\right) =t^{4 }\) for \(t\in \Omega =\left[ 0,1\right] \) and the approximation of Caputo type derivative \( \big (_{0}^{C}D_{\alpha ,\beta ,\gamma ,w_{1},w_{2}}^{\delta ,\left( 2,N_{1}\right) }f\big )(t)\) by using quadratic Lagrange interpolation polynomial \(P_2(t)\) given in (26) and the exact derivative \(\big (_{0}^{C}D_{\alpha ,\beta ,\gamma ,w_{1},w_{2}}^{\delta }f\big )(t)\) are considered. We take the set of mesh points \(\Omega ^{\zeta }\) as given in (8), also, \( w_{1}=w_{2}=1\) and \(\alpha =0.3,\beta =0.9\) and \(\gamma =1.7\). Figure 6 shows the error function \(\big ( Er_{0.3,0.9,1.7,1,1}^{-0.5,(2,128)}f\big )(t)\) for \(f\left( t\right) =t^{4}\).
5 Considering the operators with respect to functions
The concept of applying fractional integration and differentiation operators to function \(\psi (x)\) with respect to a monotonic function h(x) was studied [32, 33] and Osler’s 1970 paper marked the introduction of the complete generality of Riemann–Liouville fractional calculus with respect to functions. Within this section, we aim to generalize the operators with bivariate ML kernel with respect to an arbitrary monotonic function.
Definition 1
Let \(\psi \in L^1[a,b]\) and \(h \in C^1[a,b]\) be two functions defined on a real interval [a, b] additionally let h be positive and monotonically increasing. For \( \alpha , \beta ,\gamma , \delta , w_1, w_2 \in \mathbb {C}, Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0\) the fractional integral operator with bivariate ML kernel of the function \(\psi \) with respect to the function h is defined by
Definition 2
Let \(h \in C^1[a,b]\) be a positive and monotonically increasing function on a real interval [a, b]. For \( \alpha , \beta ,\gamma , \delta , w_1, w_2 \in \mathbb {C}, Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0\) and let n be a natural number satisfying the condition \(n-1<Re(\gamma )<n\). The fractional derivative operator with bivariate ML kernel of the function \(\psi \in C^n[a,b]\) with respect to the function h is defined by
Theorem 11
The operators (27) and (28) can be expressed as conjugates of the operators with respect to x
where the operator \(Q_h\) is defined as,
Proof
The proof can be given in a similar manner as the classical Riemann–Liouville fractional calculus with respect to functions. It is well established that the classical derivative satisfies the corresponding operational identity
We first select a function \(g \in L^1[a,b]\) and then follow the subsequent steps:
Based on definition (2), we have
where we perform the substitution of \(v=h^{-1}(s)\) into the integral. Finally, by substituting x with h(x) throughout this expression, we obtain
this establishes (29). The result in (30) for fractional derivatives directly follows from the composition of (29) with (31) repeated n times. \(\square \)
Theorem 12
The following relation holds true for the operators (27)
for any function \(\psi \in L^1[a,b]\) and for any monotonic function \(h\in C^1[a,b]\) where \(\alpha , \beta ,\gamma , \delta , w_1, w_2 \in \mathbb {C}, Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0, Re(\mu )>0.\)
Proof
This is a result of Corollary 4 [13], utilizing the conjugation expressions for all operators. These expressions are provided by Theorem 11 for operators with bivariate ML kernels and by the classical result for Riemann–Liouville integrals
\(\square \)
Theorem 13
The following relation holds true for the operators (27)
for any function \(\psi \in C^k[a,b]\), \(k=\lceil \mu \rceil \), any monotonic function \(h \in C^1[a,b]\) and \( \alpha , \beta ,\gamma , \delta , w_1, w_2 \in \mathbb {C}, Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0, Re(\mu )>0.\)
Proof
This is a result of Corollary 4 [13], using results of Theorem 11 and
\(\square \)
Theorem 14
The operator (27) satisfies the following semigroup property given by
for any function \(\psi \in L^1[a,b]\) and for any monotonic function \(h \in C^1[a,b]\), where \( \alpha , \beta ,\gamma _1, \gamma _2, \delta _1,\delta _2,w_1, w_2 \in \mathbb {C}\), \(Re(\beta )>0, Re(\alpha )>0, Re(\gamma _1)>0, Re(\gamma _2)>0.\)
Proof
This is a result of Theorem 9 [13], in combination with results of Theorem 11. \(\square \)
Theorem 15
The application of the bivariate ML integral operator to another power function can be demonstrated in the following manner
where \( \alpha , \beta ,\gamma , \delta , w_1, w_2 \in \mathbb {C}, Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0, \mu>0, \zeta >0\).
Proof
From the assumption \(\mu >0\) the function \(h(x)=(x-a)^\mu \) is a monotonically increasing function on any interval [a, c]. Further, the function to which we apply the operator (27) is \((x-a)^\zeta =h(x)^{\frac{\zeta }{\mu }}\). Thus, utilizing the result of Proposition 1 we have
thus (32) is established. \(\square \)
Theorem 16
The application of the bivariate ML integral operator with respect to a logarithm function to a power function can be given as
where \( \alpha , \beta ,\gamma , \delta , w_1, w_2 \in \mathbb {C}, Re(\beta )>0, Re(\alpha )>0, Re(\gamma )>0, \zeta >0\).
Proof
The function \(h(x)=log(x-a)\) is monotonically increasing function on any interval (a, c] with \(h(x)\rightarrow -\infty \) as \(x\rightarrow a^+\), and we apply the operator (27) to \((x-a)^\zeta =e^{\zeta h(x)}\). Using the Proposition 2 it follows that
hence we get (33). \(\square \)
Example 6
In this example we consider the integral operators given in ( 32) and (33) over the interval \(\Omega =\left[ 0,1\right] \) at the mesh points (8) for \(N_{1}=64\). Taking \(\alpha =0.3,\beta =0.9, \gamma =0.6\) and \(\delta =0.7\) the integral \(_{0}I_{0.3,0.9,0.6;x^ {\mu }}^{0.7,1,1}\left[ x^{\zeta }\right] \) is presented in Fig. 7 for various values of \(\zeta \) as 0.5, 0.9, 1.9 when \(\mu =1.2\) is fixed and for various values of \(\mu \) as 0.5, 0.9, 1.9 when \(\zeta =0.8\) is fixed. Further, Taking \(\alpha =0.3,\beta =0.9\) and \(\delta =0.7\) the integral \(_{0}I_{0.3,0.9,0.8;log(x)}^{0.7,1,1}\left[ x^{\zeta }\right] \) is presented in Fig. 8 for various values of \(\zeta \) as 0.5, 0.9, 1.9 when \(\gamma =0.8\) is fixed and for various values of \(\gamma \) as 0.5, 0.9, 1.9 for \(\zeta =0.8\).
6 Conclusion
In this paper, we have extended the work of [13] by developing a fully formed theory of fractional calculus from integral operators with bivariate ML kernels. Our approach naturally leads to many important results concerning these integral operators, such as the Laplace transform, product rule, and chain rule. We have also extended this model of fractional calculus to a higher level of generality by applying the operators with respect to functions as well as with respect to the independent variable. The composition, semigroup, and inverse properties extend naturally into this more general setting.
As applications of the work in this paper, we considered the solution of an eigenvalue problem involving Caputo type derivative operators with bivariate ML kernels, as well as a Cauchy problem involving both Riemann–Liouville derivatives and the fractional integrals with bivariate ML kernels. We discussed linear and quadratic approximations to approximate the Caputo type derivative operators with bivariate ML kernels.
As a future work we are intending to give numerical approximations for the bivariate and multivariate franctional calculus operators (see [24, 34]) by considering different approaches. Another direction of research will be to give new approximation methods for the solution of fractional differential equations.
Data Availability
No data used.
References
Mittag-Leffler, M.G.: Sur la nouvelle function \(E(x)\). C. R. Acad. Sci. 137, 554–558 (1903)
Bedi, P., Kumar, A., Abdeljawad, T., Khan, A.: Existence of mild solutions for impulsive neutral Hilfer fractional evolution equations. Adv. Differ. Equ. 1, 1–16 (2020)
Bedi, P., Kumar, A., Abdeljawad, T., Khan, Z.A., Khan, A.: Existence and approximate controllability of Hilfer fractional evolution equations with almost sectorial operators. Adv. Differ. Equ. 1, 1–15 (2020)
Begum, R., Tunç, O., Khan, H., Gulzar, H., Khan, A.: A fractional order Zika virus model with Mittag-Leffler kernel. Chaos Solitons Fractals 146, 110898 (2021)
Devi, A., Kumar, A., Abdeljawad, T., Khan, A.: Stability analysis of solutions and existence theory of fractional Lagevin equation. Alex. Eng. J. 60(4), 3641–3647 (2021)
Khan, A., Li, Y., Shah, K., Khan, T.S.: On coupled-Laplacian fractional differential equations with nonlinear boundary conditions. Complexity (2017)
Uwaheren, O.A., Adebisi, A.F., Taiwo, O.A. Perturbed collocation method for solving singular multi-order fractional differential equations of Lane-Emden type. J. Niger. Soc. Phys. Sci. 141–148 (2020)
Gorenflo, R., Kilbas, A.A., Mainardi, F., Rogosin, S.V.: Mittag-Leffler Functions, Related Topics and Applications, 2nd edn. Springer, Berlin (2020)
Kiryakova, V.S., Luchko, Y.F.: The Multi-index Mittag–Leffler functions and their applications for solving fractional order problems in applied analysis. In: AIP Conference Proceedings. American Institute of Physics, vol. 1301, pp. 597 - 613 (2010)
Luchko, Y.: The four-parameters Wright function of the second kind and its applications in FC. Mathematics 8, 970 (2020)
Prabhakar, T.R.: A singular integral equation with a generalized Mittag–Leffler function in the kernel. Yokohama Math. J. 19, 7–15 (1971)
Kilbas, A.A., Saigo, M., Saxena, R.K.: Generalized Mittag–Leffler function and generalized fractional calculus operators. Integral Transform. Spec. Funct. 15, 31–49 (2004)
Fernandez, A., Kürt, C., Özarslan, M.A.: A naturally emerging bivariate Mittag–Leffler function and associated fractional-calculus operators. Comput. Appl. Math. 39, 1–27 (2020)
Ng, Y.X., Phang, C., Loh, J.R., Isah, A., Malaysia, S.: Analytical solutions of incommensurate fractional differential equation systems with fractional order \(1< \gamma , \beta < 2\) via bivariate Mittag-Leffler functions. Mathematics 13, 14 (2022)
Garg, M., Manohar, P., Kalla, S.L.: A Mittag-Leffler-type function of two variables. Integral Transform. Spec. Funct. 24, 934–944 (2013)
Saigo, M., Kilbas, A.A.: On Mittag-Leffler type function and applications. Integral Transform. Spec. Funct. 7, 97–112 (1998)
Kilbas, A.A., Saigo, M.: On Mittag-Leffler type function, fractional calculas operators and solutions of integral equations. Integral Transform. Spec. Funct. 4, 355–370 (1996)
Shukla, A.K., Prajapati, J.C.: On a generalization of Mittag-Leffler function and its properties. J. Math. Anal. Appl. 336, 797–811 (2007)
Kilbas, A.A., Srivastava, H.M., Trujillo, J.J.: Theory and Applications of Fractional Differential Equations. Elsevier, Amsterdam (2006)
Saxena, R.K., Kalla, S.L., Saxena, R.: Multivariate analogue of generalized Mittag-Leffler function. Int. Trans. Spec. Funct. 22, 533–548 (2011)
Özarslan, M.A.: On a singular integral equation including a set of multivariate polynomials suggested by Laguerre polynomials. Appl. Math. Comput. 229, 350–358 (2014)
Özarslan, M.A., Fernandez, A.: On the fractional calculus of multivariate Mittag-Leffler functions. Int. J. Comput. Math. 99, 247–273 (2022)
Kürt, C., Fernandez, A., Özarslan, M.A.: Two unified families of bivariate Mittag-Leffler functions. Appl. Math. Comput. 443, 127785 (2023)
Kürt, C., Özarslan, M.A., Fernandez, A.: On a certain bivariate Mittag-Leffler function analysed from a fractional-calculus point of view. Math. Methods Appl. Sci. 44, 2600–2620 (2021)
Özarslan, M.A., Fernandez, A.: On a five-parameter Mittag-Leffler function and the corresponding bivariate fractional operators. Fractal Fract. 5, 45 (2021)
Samko, S.G., Kilbas, A.A., Marichev, O.I.: Fractional Integrals and Derivatives: Theory and Applications. Taylor and Francis, London (2002)
Srivastava, H.M., Daust, M.C.: A note on the convergence of Kampê de fêriet’s double hypergeometric series. Math. Nachr. 53, 151–159 (1972)
Isah, S.S., Fernandez, A., Özarslan, M.A.: On bivariate fractional calculus with general univariate analytic kernels. Chaos Solitons Fractals 171, 113495 (2023)
Constantine, G.M., Savits, T.H.: A multivariate Faa di Bruno formula with applications. Trans. Am. Math. Soc. 348(2), 503–520 (1996)
Oldham, K., Spanier, J.: The Fractional Calculus Theory and Applications of Differentiation and Integration to Arbitrary Order. Elsevier, New York (1974)
Singh, D., Sultana, F., Pandey, R.K.: Approximation of Caputo–Prabhakar derivative with application in solving time fractional advection-diffusion equation. Int. J. Numer. Methods Fluids 94(7), 896–919 (2022)
Erdélyi, A.: An integral equation involving Legendre functions. J. Soc. Ind. Appl. Math. 12(1), 15–30 (1964)
Osler, T.J.: Leibniz rule for fractional derivatives generalized and an application to infinite series. SIAM J. Appl. Math. 18(3), 658–674 (1970)
Özarslan, M.A., Kürt, C.: Bivariate Mittag-Leffler functions arising in the solutions of convolution integral equation with 2D-Laguerre Konhauser polynomials in the kernel. Appl. Math. Comput. 347, 631–644 (2019)
Funding
Open access funding provided by the Scientific and Technological Research Council of Türkiye (TÜBİTAK).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
Thye authers declare no conflict of interest.
Informed constent
Not applicable.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Elidemir, İ.O., Özarslan, M.A. & Buranay, S.C. On the analysis of fractional calculus operators with bivariate Mittag Leffler function in the kernel. J. Appl. Math. Comput. 70, 1295–1323 (2024). https://doi.org/10.1007/s12190-024-02004-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12190-024-02004-8
Keywords
- Fractional integrals and derivative
- Bivariate Mittag Leffler functions
- Caputo–Prabhakar derivative
- Lagrange interpolation