A generalized normal form and formal equivalence of two-dimensional systems with quadratic zero approximation: IV

We continue the study of invertible formal transformations of two-dimensional autonomous systems of differential equations with zero approximation represented by homogeneous polynomials of degree 2 and with perturbations in the form of power series without terms of order < 3. In the regular case, we consider systems that have the canonical form (αx ₁ ² ? sgnα x ₂ ² , x ₁ x ₂) with α ≠ 0 as the zero approximation. For such systems, we obtain resonance equations in closed form and use them to prove the theorem on the formal equivalence of systems and establish a generalized normal form to which any original system can be reduced by an invertible change of variables.     

Invariant <Emphasis Type="Italic">f</Emphasis>-structures on naturally reductive homogeneous spaces

We study invariant metric f-structures on naturally reductive homogeneous spaces and establish their relation to generalized Hermitian geometry. We prove a series of criteria characterizing geometric and algebraic properties of important classes of metric f-structures: nearly Kähler, Hermitian, Kähler, and Killing structures. It is shown that canonical f-structures on homogeneous Φ-spaces of order k (homogeneous k-symmetric spaces) play remarkable part in this line of investigation. In particular, we present the final results concerning canonical f-structures on naturally reductive homogeneous Φ-spaces of order 4 and 5.     

Boundary value problems for a higher-order parabolic equation with growing coefficients

We establish the unique solvability of boundary value problems in Hölder function classes for a linear parabolic equation of order 2m in noncylindrical domains of the class C ^{2m ? 1,α} , possibly unbounded (with respect to x as well as t), with nonsmooth (with respect to t) lateral boundary under the condition that the lower-order coefficients and the right-hand side of the equation can grow in a certain way when approaching the parabolic boundary of the domain and the leading coefficients may fail to satisfy the Dini condition near this boundary.     

On the <Emphasis Type="Italic">L</Emphasis><Superscript><Emphasis Type="Italic">p</Emphasis></Superscript>-Poisson Semigroup Associated with Elliptic Systems

We study the infinitesimal generator of the Poisson semigroup in L ^p associated with homogeneous, second-order, strongly elliptic systems with constant complex coefficients in the upper-half space, which is proved to be the Dirichlet-to-Normal mapping in this setting. Also, its domain is identified as the linear subspace of the L ^p -based Sobolev space of order one on the boundary of the upper-half space consisting of functions for which the Regularity problem is solvable. Moreover, for a class of systems containing the Lamé system, as well as all second-order, scalar elliptic operators, with constant complex coefficients, the action of the infinitesimal generator is explicitly described in terms of singular integral operators whose kernels involve first-order derivatives of the canonical fundamental solution of the given system. Furthermore, arbitrary powers of the infinitesimal generator of the said Poisson semigroup are also described in terms of higher order Sobolev spaces and a higher order Regularity problem for the system in question. Finally, we indicate how our techniques may be adapted to treat the case of higher order systems in graph Lipschitz domains.     

Symmetries of Fundamental Solutions and Their Application in Continuum Mechanics

This paper is a brief survey of the recent results in problems of approximating functions by solutions of homogeneous elliptic systems of PDEs on compact sets in the plane in the norms of C^m spaces, m ≥ 0. We focus on general second-order systems. For such systems the paper complements the recent survey by M. Mazalov, P. Paramonov, and K. Fedorovskiy (2012), where the problems of C^m approximation of functions by holomorphic, harmonic, and polyanalytic functions as well as by solutions of homogeneous elliptic PDEs with constant complex coefficients were considered.     

The Nodal Cubic is a Quantum Homogeneous Space

The cusp was recently shown to admit the structure of a quantum homogeneous space, that is, its coordinate ring B can be embedded as a right coideal subalgebra into a Hopf algebra A such that A is faithfully flat as a B-module. In the present article such a Hopf algebra A is constructed for the coordinate ring B of the nodal cubic, thus further motivating the question which affine varieties are quantum homogeneous spaces.     

Two-dimensional homogeneous cubic systems: Classification and normal forms. I

This work is the first in a series of papers devoted to classifying of two-dimensional homogeneous cubic systems based on partitioning into classes of linear equivalence. Principles have been developed that are capable of constructively distinguishing the structure of a simplest system in each class and a canonical set that defines the admissible values that can be assumed by its coefficients. The polynomial vector in the right-hand part of this system identified with a 2 × 4 matrix is called the canonical form (CF) and the system itself is called the normal cubic form. One of the main objectives of this series of papers is to maximally simplify the reduction of a system with a homogeneous cubic polynomial in the unperturbed part to the various structures of a generalized normal form (GNF). Generalized normal form refers to a system in which the perturbed part has the simplest form in some sense. The constructive implementation of the normalization process depends on the ability to explicitly specify the conditions of compatibility and possible solutions of the so-called bonding system, which is understood to be a countable set of linear algebraic equations that specify the normalizing transformations of the perturbed system. The above principles are based on the idea of the maximum possible simplification of the bonding system. This will allow one to first reduce the initial perturbed system by an invertible linear substitution of variables to a system with some CF in the unperturbed part, then reduce the resulting system, which is optimal for normalization, by almost identical substitutions to various structures of the GNF. In this paper, the following tasks are carried out: (1) the general problem is set, close problems are formulated, and the available results are described; (2) a bonding system is derived that is capable of determining the equivalence of any two perturbed systems with the same homogeneous cubic part, the possibilities of its simplification are discussed, the GNF is defined, and the method of resonant equations is given allowing one to constructively obtain all its structures; (3) special forms of recording homogeneous cubic systems in the presence of a common homogeneous factor in their right-hand parts with a degree of 1–3 are introduced, and the linear equivalence of these systems, as well as of systems without a common factor is studied, and key linear invariants are offered.     

Determinantal representations of smooth cubic surfaces

For every smooth (irreducible) cubic surface S we give an explicit construction of a representative for each of the 72 equivalence classes of determinantal representations. Equivalence classes (under GL₃ × GL₃ action by left and right multiplication) of determinantal representations are in one to one correspondence with the sets of six mutually skew lines on S and with the 72 (two-dimensional) linear systems of twisted cubic curves on S. Moreover, if a determinantal representation M corresponds to lines (a ₁,...,a ₆) then its transpose M ^t corresponds to lines (b ₁,...,b ₆) which together form a Schläfli’s double-six \(a_1\ldots a_6 \choose b_1\ldots b_6\) . We also discuss the existence of self-adjoint and definite determinantal representation for smooth real cubic surfaces. The number of these representations depends on the Segre type F _i . We show that a surface of type F _i , i = 1,2,3,4 has exactly 2(i?1) nonequivalent self-adjoint determinantal representations none of which is definite, while a surface of type F ₅ has 24 nonequivalent self-adjoint determinantal representations, 16 of which are definite.     

On the Arkhipov–Karatsuba multivariate system of congruences

The Arkhipov–Karatsuba multivariate system of congruences modulo any prime greater than the degrees of forms in this system is solvable for any right-hand sides and any number of variables larger than 8(n + 1)mlog₂(rn) + 12(n + 1)m + 4(n + 1), where n is the degree of the forms in the system and \(m = \left( {\begin{array}{*{20}{c}} {n + r - 1} \\ {r - 1} \end{array}} \right)\) is the number of congruences.     

Local analysis of a polynomial system and its perturbations

On the real (x, y)-plane, we consider an autonomous system of differential equations whose right-hand sides are polynomials of special form in x and y and a perturbed system obtained from the former by varying the coefficients in the class of functions of (x, y) satisfying the Lipschitz condition. We study the behavior of trajectories of the system in a neighborhood of the isolated equilibrium point O = (0, 0). For the main (polynomial) system, we find all possible types of arrangement of the trajectories in a neighborhood of O. For the case in which the system has TO-curves, we give coefficient criteria for each of the possible types of the point O and study conditions under which the type is preserved in the perturbed system.     

On the boundedness of generalized solutions of higher-order nonlinear elliptic equations with data from an Orlicz–Zygmund class

In the present paper, a 2mth-order quasilinear divergence equation is considered under the condition that its coefficients satisfy the Carathéodory condition and the standard conditions of growth and coercivity in the Sobolev space W^m,p(Ω), Ω ? Rⁿ, p > 1. It is proved that an arbitrary generalized (in the sense of distributions) solution u ∈ W₀^m,p (Ω) of this equation is bounded if m ≥ 2, n = mp, and the right-hand side of this equation belongs to the Orlicz–Zygmund space L(log L)^n?1(Ω).     

Descent properties of hermitian Witt groups in inseparable extensions

Let k be a field of characteristic ≠ 2, A be a central simple algebra with involution σ over k and W(A, σ) be the associated Witt group of hermitian forms. We prove that for all purely inseparable extensions L of k, the canonical map \({r_{L/k}: W(A, \sigma) \longrightarrow W(A_L, \sigma_L)}\) is an isomorphism.     

Curvature identities for principle <Emphasis Type="Italic">T</Emphasis><Superscript>1</Superscript>-bundles over almost Hermitian manifolds

We show that the identities R ₁, R ₂ and R ₃ for an almost Hermitian structure S on the base of the canonical principal T ¹-bundle are equivalent to their contact analogs for the induced almost contact metric structure S ^# on the total space of this bundle. We prove that the canonical connection of the canonical principal T ¹-bundle over a Hermitian or a quasi-Kähler manifold of class R ₃ is normal. We also prove that that the canonical connection of the canonical principal T ¹-bundle over a Vaisman-Gray manifold M of class R ₃ is normal if and only if the Lee vector of M belongs to the center of the adjoint K-algebra.     

Nearly Kähler and Hermitian <Emphasis Type="Italic">f</Emphasis>-structures on homogeneous Φ-spaces of order <Emphasis Type="Italic">k</Emphasis> with special metrics

We consider arbitrary homogeneous Φ-spaces of order k ≥ 3 of semisimple compact Lie groups G in the case of a series of special metrics. We give formulas for the Nomizu function of the Levi-Civita connection of these metrics. Using these formulas and other relations for Φ-spaces of order k, we prove necessary and sufficient conditions for the canonical f-structures on these spaces to lie in some generalized Hermitian geometry classes of f-structures: nearly Kähler (NKf-structures) and Hermitian (Hf-structures).     

Compactification of the space of branched coverings of the two-dimensional sphere

For a closed oriented surface Σ we define its degenerations into singular surfaces that are locally homeomorphic to wedges of disks. Let X_Σ,n be the set of isomorphism classes of orientation-preserving n-fold branched coverings Σ → S ² of the two-dimensional sphere. We complete X_Σ,n with the isomorphism classes of mappings that cover the sphere by the degenerations of Σ. In the case Σ = S ², the topology that we define on the obtained completion \({\overline X _{\Sigma ,n}}\) coincides on \({X_{{s^2},n}}\) with the topology induced by the space of coefficients of rational functions P/Q, where P and Q are homogeneous polynomials of degree n on ?P¹ ≌ S ². We prove that \({\overline X _{\Sigma ,n}}\) coincides with the Diaz–Edidin–Natanzon–Turaev compactification of the Hurwitz space H(Σ, n) ? X _Σ,n consisting of isomorphism classes of branched coverings with all critical values being simple.     

Gradient continuity estimates

We consider degenerate elliptic equations of p-Laplacean type

$$-{\rm{div}}\, (\gamma(x)|Du|^{p-2}Du)=\mu\,,$$

and give a sufficient condition for the continuity of Du in terms of a natural non-linear Wolff potential of the right-hand side measure μ. As a corollary we identify borderline condition for the continuity of Du in terms of the data: namely μ belongs to the Lorentz space L(n, 1/(p ? 1)), and γ(x) is a Dini continuous elliptic coefficient. This last result, together with pointwise gradient bounds via non-linear potentials, extends to the non homogeneous p-Laplacean system, thereby giving a positive answer in the vectorial case to a conjecture of Verbitsky. Continuity conditions related to the density of μ, or to the decay rate of its L ⁿ -norm on small balls, are identified as well as corollaries of the main non-linear potential criterium.

     

On using cubic spline for the solution of problems in calculus of variations

Two different approaches based on cubic B-spline are developed to approximate the solution of problems in calculus of variations. Both direct and indirect methods will be described. It is known that, when using cubic spline for interpolating a function g∈C⁴[a,b] on a uniform partition with the step size h, the optimal order of convergence derived is O(h⁴). In Zarebnia and Birjandi (J. Appl. Math. 1–10, 2012) a non-optimal O(h²) method based on cubic B-spline has been used to solve the problems in calculus of variations. In this paper at first we will obtain an optimal O(h⁴) indirect method using cubic B-spline to approximate the solution. The convergence analysis will be discussed in details. Also a locally superconvergent O(h⁶) indirect approximation will be describe. Finally the direct method based on cubic spline will be developed. Some test problems are given to demonstrate the efficiency and applicability of the numerical methods.     

Riesz Transforms Associated with Higher-Order Schrödinger Type Operators

Let L = L ₀ + V be a Schrödinger type operator, where L ₀ is a higher order elliptic operator with bounded complex coefficients in divergence form and V is a signed measurable function. Under the strongly subcritical assumption on V, we study the L ^q boundedness of Riesz transform ? ^m L ^?1/2 for q ≤ 2 based on the off-diagonal estimates of semigroup e ^{?t L} . Furthermore, the authors impose extra regularity assumptions on V to obtain the L ^q boundedness of Riesz transform ? ^m L ^?1/2 for some q > 2. In particular, these results are applied to the more interesting Schrödinger operators L = P(D) + V, where P(D) is any homogeneous positive elliptic operator with constant coefficients.     

Gradient estimates for parabolic systems from composite material

In this paper, we derive W~(1,∞) and piecewise C~(1,α) estimates for solutions, and their t-derivatives, of divergence form parabolic systems with coefficients piecewise H¨older continuous in space variables x and smooth in t. This is an extension to parabolic systems of results of Li and Nirenberg [Comm Pure Appl Math, 2003, 56:892–925] on elliptic systems. These estimates depend on the shape and the size of the surfaces of discontinuity of the coefficients, but are independent of the distance between these surfaces.