Optimality conditions involving the Mittag-Leffler tempered fractional derivative (Q2118451)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Optimality conditions involving the Mittag-Leffler tempered fractional derivative
scientific article

    Statements

    Optimality conditions involving the Mittag-Leffler tempered fractional derivative (English)
    0 references
    0 references
    0 references
    22 March 2022
    0 references
    The authors study optimization problems where the differential operator is a generalization of the tempered fractional derivative. They define the left and right Mittag-Leffler tempered fractional derivatives of \(x\) and of order \(\alpha \), based on the left and right Caputo fractional derivatives and defined through: \(^{C}\mathbb{D}_{a+}^{\alpha ,\beta ,\lambda }x(t)=\frac{1}{ \Gamma (n-\alpha )E_{\beta }(\lambda t)}\int_{a}^{t}(t-s)^{n-\alpha -1}( \frac{d}{ds})^{n}(E_{\beta }(\lambda s)x(s))ds\), \(^{C}\mathbb{D} _{b-}^{\alpha ,\beta ,\lambda }x(t)=\frac{E_{\beta }(\lambda t)}{\Gamma (n-\alpha )}\int_{t}^{b}(s-t)^{n-\alpha -1}(-\frac{d}{ds})^{n}(\frac{x(s)}{ E_{\beta }(\lambda s)})ds\), where \(x:[a,b]\rightarrow \mathbb{R}\) is an integrable function, \(n=[\alpha ]+1\), if \(\alpha \notin \mathbb{N}\), and \( n=\alpha \) otherwise, \(E_{\beta }\) is a Mittag-Leffler function, and \(\alpha ,\beta >0\), \(\lambda \geq 0\). They also define the left and right Mittag-Leffler tempered fractional derivatives of \(x\) and of order \(\alpha \) , based on the left and right Riemann-Liouville fractional derivatives and defined through: \(\mathbb{D}_{a+}^{\alpha ,\beta ,\lambda }x(t)=\frac{1}{ \Gamma (n-\alpha )E_{\beta }(\lambda t)}(\frac{d}{dt})^{n} \int_{a}^{t}(t-s)^{n-\alpha -1}E_{\beta }(\lambda s)x(s)ds\), \(\mathbb{D} _{b-}^{\alpha ,\beta ,\lambda }x(t)=\frac{E_{\beta }(\lambda t)}{\Gamma (n-\alpha )}(-\frac{d}{ds})^{n}\int_{t}^{b}(s-t)^{n-\alpha -1}(\frac{x(s)}{ E_{\beta }(\lambda s)})ds\).\ They prove integration by parts formulae involving the left and right tempered fractional derivatives. They consider the functional \(J(x)=\int_{a}^{b}L(t,x(t);^{C}\mathbb{D}_{a+}^{\alpha ,\beta ,\lambda }x(t),^{C}\mathbb{D}_{b-}^{\alpha ,\beta ,\lambda }x(t))dt\), for every \(x\in C^{1}[a,b]\), and \(\alpha \in (0,1)\), where \(L:[a,b]\times \mathbb{R}^{3}\rightarrow \mathbb{R}\) is continuously differentiable with respect to the second, third and fourth variables. Boundary conditions may be imposed. The authors derive the Euler-Lagrange equations associated to this problem assuming that the maps \(t\rightarrow \mathbb{D}_{b-}^{\alpha ,\beta ,\lambda }(\partial _{3}L[x^{\ast }](t))\) and \(t\rightarrow \mathbb{D} _{a+}^{\alpha ,\beta ,\lambda }(\partial _{4}L[x^{\ast }](t))\) are continuous on \([a,b]\), where \(x^{\ast }\) is the solution. They prove quite similar results in the cases they impose boundary conditions, of a holonomic problem, when considering higher-order derivatives or for a Herglotz problem, all with boundary conditions. The paper ends with the presentation of a numerical scheme for such problems, the authors using either Matlab or Mathematica software. They present an example.
    0 references
    fractional calculus
    0 references
    calculus of variations
    0 references
    optimization
    0 references
    tempered fractional derivative
    0 references
    Mittag-Leffler function
    0 references
    numerical method
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references