Optimal Design of Brittle Composite Materials: a Nonsmooth Approach

Our goal is to design brittle composite materials yielding maximal energy dissipation for a given static load case. We focus on the effect of variation of fiber shapes on resulting crack paths and thus on the fracture energy. To this end, we formulate a shape optimization problem, in which the cost function is the fracture energy and the state problem consists in the determination of the potentially discontinuous displacement field in the two-dimensional domain. Thereby, the behavior at the crack surfaces is modeled by cohesive laws. We impose a nonpenetration condition to avoid interpenetration of opposite crack sides. Accordingly, the state problem is formulated as variational inequality. This leads to potential nondifferentiability of the shape-state mapping. For the numerical solution, we derive first-order information in the form of subgradients. We conclude the article by numerical results.     

Higher order curvature flows on surfaces

We consider a sixth and an eighth order conformal flow on Riemannian surfaces, which arise as gradient flows for the Calabi energy with respect to a higher order metric. Motivated by a work of Struwe which unified the approach to the Hamilton-Ricci and Calabi flow, we extend the method to these higher order cases. Our results contain global existence and exponentially fast convergence to a constant scalar curvature metric.Uniform bounds on the conformal factor are obtained via the concentration-compactness result for conformal metrics. In the case of the sphere we use the idea of DeTurck's gauge flow to derive bounds up to conformal transformation.We prove exponential convergence by showing that the Calabi energy decreases exponentially fast. The problem of the non-trivial kernel in the evolution of the Calabi energy on the sphere is resolved by using Kazdan-Warner's identity. Mathematics subject classifications (2000) {Primary 53C44 Secondary 35K25}     

The Cauchy Problem on a Characteristic Cone for the Einstein Equations in Arbitrary Dimensions

We derive explicit formulae for a set of constraints for the Einstein equations on a null hypersurface, in arbitrary space–time dimensions n + 1 ≥ 3. We solve these constraints and show that they provide necessary and sufficient conditions so that a spacetime solution of the Cauchy problem on a characteristic cone for the hyperbolic system of the reduced Einstein equations in wave-map gauge also satisfies the full Einstein equations. We prove a geometric uniqueness theorem for this Cauchy problem in the vacuum case.     

Universally coupled massive gravity

We derive Einstein’s equations from a linear theory in flat space-time using free-field gauge invariance and universal coupling. The gravitational potential can be either covariant or contravariant and of almost any density weight. We adapt these results to yield universally coupled massive variants of Einstein’s equations, yielding two one-parameter families of distinct theories with spin 2 and spin 0. The Freund-Maheshwari-Schonberg theory is therefore not the unique universally coupled massive generalization of Einstein’s theory, although it is privileged in some respects. The theories we derive are a subset of those found by Ogievetsky and Polubarinov by other means. The question of positive energy, which continues to be discussed, might be addressed numerically in spherical symmetry. We briefly comment on the issue of causality with two observable metrics and the need for gauge freedom and address some criticisms by Padmanabhan of field derivations of Einstein-like equations along the way. __________ Translated from Teoreticheskaya i Matematicheskaya Fizika, Vol. 151, No. 2, pp. 311–336, May, 2007.     

Time decay of maxwell—klein—gordon equations in 4—dimensional minkowski space

Abstract. In this paper I derive a gauge invariant decay estimate of the solutions of massive Maxwell—Klein—Gordon fields equations in the 4—dimensional Minkowski space, provided that the initial energy of the system is bounded. This estimate implies that the Klein—Gordon field decays to zero in the local L² norm. I also show that the local energy decays. The proof is based on gauge invariant energy identities.     

Finite element approximation of a free boundary plasma problem

In this article, we study a finite element approximation for a model free boundary plasma problem. Using a mixed approach (which resembles an optimal control problem with control constraints), we formulate a weak formulation and study the existence and uniqueness of a solution to the continuous model problem. Using the same setting, we formulate and analyze the discrete problem. We derive optimal order energy norm a priori error estimates proving the convergence of the method. Further, we derive a reliable and efficient a posteriori error estimator for the adaptive mesh refinement algorithm. Finally, we illustrate the theoretical results by some numerical examples.     

Uniform stability of monopoles

Two proofs of uniform stability in energy norm of BPS monopoles in the Yang-Mills-Higgs equations on Minkowski space-time are presented. BPS monopoles are minimisers of the static Yang-Mills-Higgs functional. The space of monopoles is an infinite dimensional submanifold whose quotient by the group of gauge transformations is finite dimensional. The problem of proving the existence and regularity of monopoles which are closest to the solution at each time is also discussed. The first proof establishes the existence of monopoles which are closest in ; this is achieved using Schoen-Uhlenbeck regularity theory for harmonic maps, and works for solutions of fairly high regularity ( with ). A uniform bound is then obtained for the distance function which gives the distance from the solution to the projected point in a certain norm, first defined by Taubes, which is related to the energy. The second proof involves the direct minimisation of the distance function with respect to which stability is proved. Although this function is minimised only subject to a closeness condition, this approach has the advantage of yielding a proof of uniform stability valid for solutions with only the regularity required for finite energy (). There are two principal sources of difficulty in the present problem: the fact that the continuous spectrum of the Hessian extends all the way to zero and the presence of the infinite dimensional group of gauge symmetries. Received February 13, 1997 / Accepted January 30, 1998     

Peculiarities of the electron energy spectrum in the Coulomb field of a superheavy nucleus

We consider the peculiarities of the electron energy spectrum in the Coulomb field of a superheavy nucleus and discuss the long history of an incorrect interpretation of this problem in the case of a pointlike nucleus and its current correct solution. We consider the spectral problem in the case of a regularized Coulomb potential. For some special regularizations, we derive an exact equation for the point spectrum in the energy interval (-m,m) and find some of its solutions numerically. We also derive an exact equation for charges yielding bound states with the energy E = -m; some call them supercritical charges. We show the existence of an infinite number of such charges. Their existence does not mean that the oneparticle relativistic quantum mechanics based on the Dirac Hamiltonian with the Coulomb field of such charges is mathematically inconsistent, although it is physically unacceptable because the spectrum of the Hamiltonian is unbounded from below. The question of constructing a consistent nonperturbative second-quantized theory remains open, and the consequences of the existence of supercritical charges from the standpoint of the possibility of constructing such a theory also remain unclear.     

Risk measurement and risk-averse control of partially observable discrete-time Markov systems

We consider risk measurement in controlled partially observable Markov processes in discrete time. We introduce a new concept of conditional stochastic time consistency and we derive the structure of risk measures enjoying this property. We prove that they can be represented by a collection of static law invariant risk measures on the space of function of the observable part of the state. We also derive the corresponding dynamic programming equations. Finally we illustrate the results on a machine deterioration problem.     

Asymptotic behaviour in a conducting electromagnetic medium

We investigate the initial-boundary value problem for Maxwell'sequations in linear conducting materials together with dissipativeboundary conditions. We show that it is possible to introducethe free energy and derive from it a domain of dependence. Weprove the existence, uniqueness and asymptotic stability ofthe solution when one of Maxwell's equations is considered asa constraint for the electromagnetic fields.     

A remark on the junction in a thin multi-domain: the non convex case

Our aim consists of studying, in the spirit of Gamma convergence, a dimension reduction problem for a multi-domain filled of either an hyperelastic material or a non simple grade-two one. We derive asymptotically the limit energy density starting from a sample described trough non convex bulk energy densities, depending either on the first or second order derivative of the displacement.      

Fermi-Walker transport and the Weyl connection

We derive equations relating the Fermi-Walker and the congruent Weyl transports. Using these equations, we show that a non-Abelian gauge field can result in the Thomas precession of a gyroscope. We find solutions to the equations for such a non-Abelian gauge field. Translated from Teoreticheskaya i Matematicheskaya Fizika, Vol. 119, No. 1, pp. 136–141, April, 1999.     

Quasi-averages in the solution of the classification problem for equilibriums of condensed media with a spontaneously broken symmetry

We develop a statistical approach for solving the classification problem for equilibriums of degenerate condensed media. We introduce generators of unbroken and spatial symmetries of an equilibrium and use them to derive the classification equations for the order parameter. We elucidate the mechanism of the appearance of additional thermodynamic parameters characterizing both homogeneous and inhomogeneous equilibriums. We solve the classification problem for equilibriums of various liquid crystals analytically.     

String-like excitations in quantum electrodynamics

A study is made of the evolution of string-like excitations of the gauge field (exponentials ordered along the contour of integration) in free quantum electrodynamics and in electrodynamics with static sources. It is shown that such excitations are unstable and are transformed after radiation of excess energy into the field of Coulomb sources. These results are compared with calculations in lattice quantum electrodynamics. It is also shown that in this case the strong-coupling method is not suitable for solving the confinement problem.Joint Institute for Nuclear Research, Dubna. Translated from Teoreticheskaya i Matematicheskaya Fizika, Vol. 97, No. 3, pp. 373–385, December, 1993.     

Maxwell–Dirac Equations in Four-Dimensional Minkowski Space

Abstract

I derive the global existence and asymptotic behavior of small amplitude solutions to the system of massive coupled classical Maxwell–Dirac equations in the four-dimensional Minkowski space. Because the physically defined energy of the system is not positive definite, I transform it into an equivalent system of Maxwell–Klein–Gordon equations, which I study with a method based on gauge invariant energy estimates and geometric properties of the equations.     

Homogenization of a Ginzburg–Landau model for a nematic liquid crystal with inclusions

We consider a nonlinear homogenization problem for a Ginzburg–Landau functional with a (positive or negative) surface energy term describing a nematic liquid crystal with inclusions. Assuming that inclusions are separated by distances of the same order ɛ as their size, we find a limiting functional as ɛ approaches zero. We generalize the variational method of mesocharacteristics to show that a corresponding homogenized problem for arbitrary, periodic or non-periodic geometries is described by an anisotropic Ginzburg–Landau functional. We obtain computational formulas for material characteristics of an effective medium. As a byproduct of our analysis, we show that the limiting functional is a Γ-limit of a sequence of Ginzburg–Landau functionals. Furthermore, we prove that a cross-term corresponding to interactions between the bulk and the surface energy terms does not appear at the leading order in the homogenized limit.     

Energy estimates and numerical verification of the stabilized Domain Decomposition Finite Element/Finite Difference approach for time-dependent Maxwell’s system

We rigorously derive energy estimates for the second order vector wave equation with gauge condition for the electric field with non-constant electric permittivity function. This equation is used in the stabilized Domain Decomposition Finite Element/Finite Difference approach for time-dependent Maxwell’s system. Our numerical experiments illustrate efficiency of the modified hybrid scheme in two and three space dimensions when the method is applied for generation of backscattering data in the reconstruction of the electric permittivity function.     

A Posteriori Error Estimators of Gradient Recovery Type for Fem of a Model Optimal Control Problem

In this paper, we derive an a posteriori error estimator of gradient recovery type for a model optimal control problem. We show that the a posteriori error estimator is equivalent to the discretization error in a norm of energy type on general meshes. Furthermore, when the solution of the control problem is smooth and the meshes are uniform, it is shown to be asymptotically exact.     

Optimal control of time discrete two-phase flow governed by a diffuse interface model

We present results on optimal control of two-phase flows. The fluid is modeled by a thermodynamically consistent diffuse interface model and allows to treat fluids of different densities and viscosities. In earlier work we proposed an energy stable time discretization for this model that we now employ to derive existence of optimal controls for a time discrete optimal control problem. The control aim is to obtain a desired distribution of the two phases in the system. For this we investigate three control actions. We use tangential Dirichlet boundary control and distributed control. We further consider the inverse problem of finding an initial distribution such that the evolution over a given time horizon starting from this value is close to a desired distribution. (© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

The equations of elastostatics in a Riemannian manifold

To begin with, we identify the equations of elastostatics in a Riemannian manifold, which generalize those of classical elasticity in the three-dimensional Euclidean space. Our approach relies on the principle of least energy, which asserts that the deformation of the elastic body arising in response to given loads minimizes over a specific set of admissible deformations the total energy of the elastic body, defined as the difference between the strain energy and the potential of the loads. Assuming that the strain energy is a function of the metric tensor field induced by the deformation, we first derive the principle of virtual work and the associated nonlinear boundary value problem of nonlinear elasticity from the expression of the total energy of the elastic body. We then show that this boundary value problem possesses a solution if the loads are sufficiently small (in a sense we specify).