Vol 99, No 2 (2019)
- Year: 2019
- Articles: 31
- URL: https://journals.rcsi.science/1064-5624/issue/view/13882
Mathematics
Peano-Type Curves, Liouville Numbers, and Microscopic Sets
Abstract
Peano-type curves in multidimensional Euclidean space are considered in terms of number theory. In contrast to curves constructed by D. Hilbert, H. Lebesgue, V. Sierpinski, and others, this paper presents results showing that each such curve is a continuous image of universal (shared by all curves) nowhere dense perfect subsets of the interval [0, 1] with a zero s-dimensional Hausdorff measure that consist of only Liouville numbers. An example of a problem in which a pair of continuous functions controlling the behavior of an oscillating system generates a Peano-type curve in the plane is given.
Logical Language of Description of Polynomial Computing
Abstract
The concept of a term and, accordingly, the concept of a formula are extended using new operators. These extensions of the language preserve the expressiveness of \(\Sigma \)-formulas and, at the level of \({{\Delta }_{0}}\)-formulas and terms, they ensure the polynomiality of algorithms for calculating the value of a term and deciding the truth of a \({{\Sigma }_{0}}\)-formula.
Accelerated Primal-Dual Gradient Descent with Linesearch for Convex, Nonconvex, and Nonsmooth Optimization Problems
Abstract
A new version of accelerated gradient descent is proposed. The method does not require any a priori information on the objective function, uses a linesearch procedure for convergence acceleration in practice, converge according to well-known lower bounds for both convex and nonconvex objective functions, and has primal-dual properties. A universal version of this method is also described.
2-Factor Newton Method for Solving Constrained Optimization Problems with a Degenerate Kuhn–Tucker System
Abstract
A new method is proposed for solving constrained optimization problems with inequality constraints in the case when the system of Kuhn–Tucker necessary optimality conditions is degenerate. This situation arises, for example, when the strict complementarity conditions do not hold for the solution. A reduction of the inequality-constraint problem to an equality-constraint one and the use of a new 2 factor Newton method for efficiently solving the resulting degenerate system of optimality conditions are justified.
On the Complexity of Reductive Group Actions over Algebraically Nonclosed Field and Strong Stability of the Actions on Flag Varieties
Abstract
Classification of One-Dimensional Attractors of Diffeomorphisms of Surfaces by Means of Pseudo-Anosov Homeomorphisms
Abstract
Axiom A diffeomorphisms of closed 2-manifold of genus \(p \geqslant 2\) whose nonwandering set contains a perfect spaciously situated one-dimensional attractor are considered. It is shown that such diffeomorphisms are topologically semiconjugate to a pseudo-Anosov homeomorphism with the same induced automorphism of fundamental group. The main result of this paper is as follows. Two diffeomorphisms from the given class are topologically conjugate on perfect spaciously situated attractors if and only if the corresponding homotopic pseudo-Anosov homeomorphisms are topologically conjugate by means of a homeomorphism that maps a certain subset of one pseudo-Anosov homeomorphism onto a subset of the other.
Optimal Feedback Control Problem for Bingham Media Motion with Periodic Boundary Conditions
Abstract
We study the optimal feedback control problem for the motion of Bingham media with periodic boundary conditions in two- and three-dimensional cases. First, the considered problem is interpreted as an operator inclusion with a multivalued right-hand side. Then, the approximation-topological approach to hydrodynamic problems and the degree theory for a class of multivalued maps are used to prove the existence of solutions of this inclusion. Finally, we prove that, among the solutions of the considered problem, there exists one minimizing the given cost functional.
The Least Distance between Extrema and the Minimum Period of Solutions of Autonomous Vector Differential Equations
Abstract
Solutions x(t) of the equation \(\dot {x} = f(x)\), where \(x \in {{{\text{R}}}^{n}}\) and the function f(x) satisfies the Lipschitz condition with an arbitrary vector norm, are considered. It is proved that the lower bound for the distances between successive extrema xk(t), k = 1, 2, …, n, is \(\frac{\pi }{L}\), where L is the Lipschitz constant. For nonconstant periodic solutions, the lower bound for the periods is \(\frac{{2\pi }}{L}\). These estimates are sharp for norms that are invariant with respect to permutations of indices.
Interpolation by Sums of Series of Exponentials and Global Cauchy Problem for Convolution Operators
Abstract
The study is made of the problem of multiple interpolation on an infinite nodes set by the sums of absolutely convergent series of exponentials whose exponents are from a given set. For entire function conditions on nodes and exponents are obtained that give solubility of the problem. A new approach is demonstrated that enable us, for the case of holomorphic function in a domain, to obtain criteria of solubility of the problem for some class of exponents set and for a special class of nodes set. Moreover the necessity of the conditions is proved in great generality namely for arbitrary nodes sets and in the setting of interpolation by functions that are represented as the Laplace transforms of the Radon measures over the exponents set. Solubility is obtained of the global Cauchy problem for convolution operator with data on the nodes set in domain, in the form of the series of exponentials whose exponents belong to a sparse subset of zero set of characteristic function of the operator. The results substantially strengthen the known results on the theme.
Controllability and Optimal Controllability for Operator Equations of the First Kind in (B)-Spaces: Examples for ODE in \({{\mathbb{R}}^{n}}\)
Abstract
For control and observation problems considered for operator equations of the first kind in Banach spaces, an controllability criterion is stated. In the case of reflexive strictly convex (B)-spaces, the BUME method and the method of monotone mappings are used to find optimal controls and an abstract maximum principle is formulated. The indicated problems for ODE systems in \({{\mathbb{R}}^{n}}\) are investigated as an example.
Homogenization of the Boundary Value Problem for the Poisson Equation with Rapidly Oscillating Nonlinear Boundary Conditions: Space Dimension n ≥ 3, Critical Case
Abstract
The homogenization of the Poisson equation in a bounded domain with rapidly oscillating boundary conditions specified on a part of the domain boundary is studied. A Neumann boundary condition alternates with an ε-periodically distributed nonlinear Robin condition involving the coefficient \({{\varepsilon }^{{ - \beta }}}\), where \(\beta \in \mathbb{R}\). The diameter of the boundary portions with a nonlinear Robin condition is of order \(O({{\varepsilon }^{\alpha }}),\)\(\alpha > 1\). A critical relation between the parameters \(\alpha \) and \(\beta \) is considered.
Degenerate Boundary Conditions on a Geometric Graph
Abstract
The boundary conditions of the Sturm–Liouville problem defined on a star-shaped geometric graph of three edges are studied. It is shown that, if the lengths of the edges are different, then the Sturm–Liouville problem does not have degenerate boundary conditions. If the lengths of the edges and the potentials are identical, then the characteristic determinant of the Sturm–Liouville problem cannot be equal to a constant different from zero. However, the set of Sturm–Liouville problems for which the characteristic determinant is identically zero is infinite (continuum). In this way, in contrast to the Sturm–Liouville problem defined on an interval, the set of boundary value problems on a star-shaped graph whose spectrum completely fills the entire plane is much richer. In the particular case when the minor \({{A}_{{124}}}\) of the coefficient matrix is nonzero, this set consists of not two problems, as in the case of the Sturm–Liouville problem given on an interval, but rather of 18 classes, each containing two to four arbitrary constants.
Sub-Finsler Structures on the Engel Group
Abstract
A one-parameter family of left-invariant sub-Finsler problems on a four-dimensional nilpotent Lie group of depth 3 with two generators is considered. The indicatrix of sub-Finsler structures is a square rotated by an arbitrary angle in the distribution. Methods of optimal control theory are applied. Abnormal and singular normal trajectories are described, and their optimality is proved. Singular trajectories arriving at the boundary of the reachable set in fixed time are characterized. A bang-bang phase flow is constructed, and estimates for the number of switchings on bang-bang trajectories are obtained. The structure of all normal extremals is described. Mixed trajectories are studied.
Analogues of Korn’s Inequality on Heisenberg Groups
Abstract
Two analogues of Korn’s inequality on Heisenberg groups are constructed. First, the norm of the horizontal differential is estimated in terms of its symmetric part. Second, Korn’s inequality is treated as a coercive estimate for a differential operator whose kernel coincides with the Lie algebra of the isometry group. For this purpose, we construct a differential operator whose kernel coincides with the Lie algebra of the isometry group on Heisenberg groups and prove a coercive estimate for this operator. Additionally, a coercive estimate is proved for a differential operator whose kernel coincides with the Lie algebra of the group of conformal mappings on Heisenberg groups.
On Minimal Surfaces on Two-Step Carnot Groups
Abstract
For graph mappings constructed from contact mappings of arbitrary two-step Carnot groups, conditions for the correct formulation of the minimal surfaces problem are found. A suitable notion of the increment of the (sub-Riemannian) area functional is introduced, the differentiability of this functional is proved, and necessary minimality conditions for graph surfaces are deduced. These conditions are also expressed in terms of the sub-Riemannian mean curvature.
Outer Billiard outside a Regular Dodecagon
Abstract
On the Uniqueness of a Solution to an Inverse Problem of Scattering by an Inhomogeneous Solid with a Piecewise Hölder Refractive Index in a Special Function Class
Abstract
The problem of reconstructing a piecewise Hölder continuous function describing the refractive index of an inhomogeneous obstacle scattering a monochromatic wave is considered. The boundary value scattering problem is reduced to a system of integral equations. The equivalence of the integral and differential formulations of the problem is proved. A two-step method for solving the inverse problem is proposed. A linear integral equation of the first kind is solved at the first step. Sufficient conditions for the uniqueness of its solution in the class of piecewise constant functions are obtained. At the second step, the unknown refractive index is explicitly expressed in terms of the solution obtained at the first step.
On Lindeberg–Feller Limit Theorem
Abstract
In the Lindeberg–Feller theorem, the Lindeberg condition is present. The fulfillment of this condition must be checked for any ε > 0. We formulae a new condition in terms of some generalization of moments of order 2 + \(\alpha \), which does not depend on ε, and show that this condition is equivalent to the Lindeberg condition, and if this condition is valid for some \(\alpha > 0\) then it is valid for any \(\alpha \) > 0. In the nonclassical setting (in the absence of conditions of a uniform infinitely smallness) V.I. Rotar formulated an analogue of the Lindeberg condition in terms of the second pseudo-momens. The paper presents the same modification of Rotar’s condition, which does not depend on ε. In addition, we discuss variants of the simple proofs of theorems of Lindeberg–Feller and Rotar and some related inequalities.
New Approach to Farkas’ Theorem of the Alternative
Abstract
The classical Farkas theorem of the alternative is considered, which is widely used in various areas of mathematics and has numerous proofs and formulations. An entirely new elementary proof of this theorem is proposed. It is based on the consideration of a functional that, under Farkas’ condition, is bounded below on the whole space and attains a minimum. The assertion of Farkas’ theorem that a vector belongs to a cone is equivalent to the fact that the gradient of this functional is zero at the minimizer.
On the Bisection Method for Normal Matrices
On Infinite-Dimensional Integer Hankel Matrices
Abstract
Mathematical Physics
Vlasov–Poisson–Poisson Equations, Critical Mass, and Kordylewski Clouds
Abstract
The Vlasov–Poisson–Poisson equation is derived to study stationary solutions for a system of gravitating charged particles in a neighborhood of triangular libration points (Kordylewski clouds). Stationary solutions are sought in the form of functions of integrals, which leads to elliptic nonlinear equations for the gravitational and electrostatic field potentials. This yields a critical mass: gravitation dominates for bodies with large masses, while electrostatics dominates for bodies with smaller masses.
Bifurcations of Liouville Tori in a System of Two Vortices of Positive Intensity in a Bose–Einstein Condensate
Abstract
A completely Liouville integrable Hamiltonian system with two degrees of freedom that describes the dynamics of two vortex filaments in a Bose–Einstein condensate enclosed in a harmonic trap is considered. For a pair of vortices of positive intensity, a bifurcation of three Liouville tori into a single one is detected. Such a bifurcation was previously found in the Goryachev–Chaplygin–Sretensky integrable case in rigid body dynamics. For an integrable perturbation of the intensity ratio parameter, the identified bifurcation proves to be unstable, which leads to bifurcations of the type of two tori into one and vice versa.
Computer Science
Explicit Methods for Integrating Stiff Cauchy Problems
Abstract
An explicit method for solving stiff Cauchy problems is proposed. The method relies on explicit schemes and a step size selection algorithm based on the curvature of an integral curve. Closed-form formulas are derived for finding the curvature. For Runge–Kutta schemes with up to four stages, the corresponding sets of coefficients are given. The method is validated on a test problem with a given exact solution. It is shown that the method is as accurate and robust as implicit methods, but is substantially superior to them in efficiency. A numerical example involving chemical kinetics computations with 9 components and 50 reactions is given.
Classification of Lung Nodules Using CT Images Based on Texture Features and Fractal Dimension Transformation
Abstract
A new computer-aided detection (CAD) system for lung nodule detection and selection in computed tomography scans is substantiated and implemented. The method consists of the following stages: preprocessing based on threshold and morphological filtration, the formation of suspicious regions of interest using a priori information, the detection of lung nodules by applying the fractal dimension transformation, the computation of informative texture features for identified lung nodules, and their classification by applying the SVM and AdaBoost algorithms. A physical interpretation of the proposed CAD system is given, and its block diagram is constructed. The simulation results based on the proposed CAD method demonstrate advantages of the new approach in terms of standard criteria, such as sensitivity and the false-positive rate.