Calculating differential Galois groups of parametrized differential equations, with applications to hypertranscendence (Q2396234)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Calculating differential Galois groups of parametrized differential equations, with applications to hypertranscendence
scientific article

    Statements

    Calculating differential Galois groups of parametrized differential equations, with applications to hypertranscendence (English)
    0 references
    0 references
    0 references
    0 references
    7 June 2017
    0 references
    The paper is devoted to issues related to the direct problem of the differential Galois theory. In particular, with algebraic independence of solutions of linear differential equations. Based on their previous results on the representation theory of linear differential algebraic groups (see [\textit{A. Minchenko} et al., Int. Math. Res. Not. 2015, No. 7, 1733--1793 (2015; Zbl 1339.12003)]), the authors obtained the following generalization of one result of Kolchin on algebraic independence (see [\textit{E. R. Kolchin}, Am. J. Math. 90, 1151--1164 (1968; Zbl 0169.36701)]). Let \(\Delta =\{\partial,\delta \}\) be a set of two derivations and \(K\) be a \(\Delta\)-field such that \(k= K^{\partial}\) is \(\delta\)-closed. Theorem 4.7. Let \(L\in K[\partial ]\) be an irreducible \(\partial \)-operator such that \(\mathrm{Gal}(L)\) is a quasi-simple linear algebraic group. Denote \(n= \operatorname{ord} L\) and \(m=\dim\mathrm{Gal}(L)\). Suppose that \(m\ne n\). Let \(b\in K^*\) and \(F\) be a \(\Delta\)-field extension of \(K\) such that \(F^\partial = k\) and \(F\) contains \(z\), a solution of \(L(y) = b\), and \(u_1, \ldots, u_n\) \(K\)-linearly independent solutions of \(L(y) = 0\). Then the functions \(v_1, \ldots, v_m, z, \ldots, \partial^{n-1}z\) and all their derivatives with respect to \(\delta \) are algebraically independent over \(K\), where \(\{v_1, \ldots, v_m\}\subset \{u_1, \ldots, \partial^{n-1}u_1, \ldots, u_n, \ldots, \partial^{n-1}u_n\}\) is a maximal algebraically independent over \(K\) subset if and only if the linear differential system \(\partial (B)-\delta (A_L)= A_L B-BA_L\), where \(A_L\) denotes the companion matrix of \(L\), has no solutions \(B\in K^{n\times n}\) and the linear differential equation \(L(y) = b\) has no solutions in \(K\). By applying Theorem 4.7 to the Lommel equation \[ \frac{d^2y}{dx^2}+\frac{1}{x}\frac{dy}{dx}+ \left(1-\frac{\alpha^2}{x^2}\right)y = x^{\mu -1}\quad (\alpha, \mu \in \mathbb{C}),\] the authors showed the algebraic independence of some functions associated with its solutions. A similar consideration of the problems of algebraic independence of solutions of difference equations can be found in [\textit{L. Di Vizio} et al., J. Inst. Math. Jussieu 16, No. 1, 59--119 (2017; Zbl 1390.12006); see also Preprint, \url{arXiv:1310.1289}].
    0 references
    0 references
    parametrized Picard-Vessiot Theory
    0 references
    calculation of differential Galois groups
    0 references
    hypertranscendence
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references