


Vol 42, No 2 (2016)
- Year: 2016
- Articles: 8
- URL: https://journal-vniispk.ru/0361-7688/issue/view/10827
Article
Procedures for searching local solutions of linear differential systems with infinite power series in the role of coefficients
Abstract
Construction of Laurent, regular, and formal (exponential–logarithmic) solutions of full-rank linear ordinary differential systems is discussed. The systems may have an arbitrary order, and their coefficients are formal power series given algorithmically. It has been established earlier that the first two problems are algorithmically decidable and the third problem is not decidable. A restricted variant of the third problem was suggested for which the desired algorithm exists. In the paper, a brief survey of algorithms for the abovementioned decidable problems is given. Implementations of these algorithms in the form of Maple procedures with a uniform interface and data representation are suggested.



Parameterization of the discriminant set of a polynomial
Abstract
The discriminant set of a real polynomial is studied. It is shown that this set has a complex hierarchical structure and consists of algebraic varieties of various dimensions. A constructive algorithm for a polynomial parameterization of the discriminant set in the space of the coefficients of the polynomial is proposed. Each variety of a greter dimension can be geometrically considered as a tangent developable surface formed by one-dimensional linear varieties. The role of the directrix is played by the component of the discriminant set with the dimension by one less on which the original polynomial has a single multiple root and the other roots are simple. The relationship between the structure of the discriminant set and the partitioning of natural numbers is revealed. Various algorithms for the calculation of subdiscriminants of polynomials are also discussed. The basic algorithms described in this paper are implemented as a library for Maple.



Using two types of computer algebra systems to solve maxwell optics problems
Abstract
To synthesize Maxwell optics systems, the mathematical apparatus of tensor and vector analysis is generally employed. This mathematical apparatus implies executing a great number of simple stereotyped operations, which are adequately supported by computer algebra systems. In this paper, we distinguish between two stages of working with a mathematical model: model development and model usage. Each of these stages implies its own computer algebra system. As a model problem, we consider the problem of geometrization of Maxwell’s equations. Two computer algebra systems—Cadabra and FORM—are selected for use at different stages of investigation.



Partitions of the set of selected unknowns in linear differential–algebraic systems
Abstract
As was shown earlier, for a linear differential–algebraic system A1y′ + A0y = 0 with a selected part of unknowns (entries of a column vector y), it is possible to construct a differential system ỹ′ = Bỹ, where the column vector ỹ is formed by some entries of y, and a linear algebraic system by means of which the selected entries that are not contained in ỹ can be expressed in terms of the selected entries included in ỹ. In the paper, sizes of the differential and algebraic systems obtained are studied. Conditions are established under the fulfillment of which the size of the algebraic system is determined unambiguously and the size of the differential system is minimal.



Approximation of a quantum algorithm for order finding
Abstract
A quantum algorithm for the computation the order of an integer, which uses the quantum Fourier transform, is discussed. The cases of the exact and approximate Fourier transform are considered, and estimates of the probability of the successful solution of the problem that significantly improve the available results are obtained. The quantum algorithm for order finding is simulated using the QuantumCircuit package written in the computer algebra system Wolfram Mathematica, and its efficiency in combination with the approximate Fourier transform is demonstrated.



Algorithm for construction of volume forms on toric varieties starting from a convex integer polytope
Abstract
This paper presents a method and a corresponding algorithm for constructing volume forms (and related forms that act as kernels of integral representations) on toric varieties from a convex integer polytope. The algorithm is implemented in the Maple computer algebra system. The constructed volume forms are similar to the volume forms of the Fubini–Study metric on a complex projective space and can be used for constructing integral representations of holomorphic functions in polycircular regions of a multidimensional complex space.



On checking existence of infinitely differentiable solutions of partial differential equations with boundary conditions
Abstract
The problem of checking existence of infinitely differentiable solutions of linear partial differential equations with zero boundary conditions is considered. The coefficients of the equations are assumed to be polynomials over Z in independent variables. It is proved that this problem is algorithmically undecidable. This result extends results of our earlier studies of analytic solutions. The proof relies on the result obtained by Denef and Lipshitz concerning the relationship between a certain subset of nonhomogeneous differential equations of the considered form (but without boundary conditions) and Diophantine equations.



An algorithm for constructing Darboux transformations of type I for third-order hyperbolic operators of two variables
Abstract
Darboux transformations of type I are invertible Darboux transformations with explicit short formulas for inverse transformations. These transformations are invariant with respect to gauge transformations, and, for gauge transformations acting on third-order hyperbolic operators of two variables, a general-form system of generating differential invariants is known. In the paper, first-order Darboux transformations of type I for this class of operators are considered. The corresponding operator orbits are directed graphs with at most three edges originating from each vertex. In the paper, an algorithm for constructing such orbits is suggested. We have derived criteria for existence of first-order Darboux transformations of type I in terms of the generating invariants, formulas for transforming invariants, and the so-called “triangle rule” property of orbits. The corresponding implementation in the LPDO package is described. The orbits are constructed in two different forms, one of which outputs the graph in the format of the well-known built-in Maple package Graph Theory.


