Search

International audience; In this paper, we consider a semi-linear fourth order parabolic equation in a bounded smooth domain Ω with homogeneous Dirichlet and Neumann boundary conditions. The main result of this paper is the null controllability and the exact controllability to the trajectories at any time T > 0 for the associated control system with a control function acting at the interior.; Dans ce papier, on considère uneéquation parabolique semi-linéaire de quatrième ordre dans un domaine borné régulier Ω avec des conditions aux limites de type Dirichlet et Neumann homogènes. Le résultat principal de ce papier concerne la contrôlabilitéà zéro et la contrôlabilité exacte pour tout T > 0 du système de contrôle associé avec un contrôle agissantà l'interieur.

The purpose of this paper is to show well-posedness results in L 2 -based Sobolev spaces for transmission, Dirichlet, Neumann, and mixed boundary value problems for the Brinkman system with L ∞ coefficients in Lipschitz domains on a compact Riemannian manifold of dimension m ≥ 2 . The Dirichlet, transmission, and mixed problems for the nonlinear Darcy-Forchheimer-Brinkman system with L ∞ coefficients are also analyzed. First, we focus on the well-posedness of linear transmission, Dirichlet and mixed boundary value problems for the Brinkman system with L ∞ coefficients in Lipschitz domains on compact Riemannian manifolds by using a variational approach that reduces such a boundary value problem to a mixed variational formulation defined in terms of two bilinear continuous forms, one of them satisfying a coercivity condition and another one the inf-sup condition. Further, we show the equivalence between each boundary value problem for the Brinkman system with L ∞ coefficients and its mixed variational counterpart, and then the well posedness in L 2 -based Sobolev spaces by using the Necas-Babuska-Brezzi technique. The second goal of this paper is the construction of the Newtonian and layer potential operators for the Brinkman system with L ∞ coefficients in Lipschitz domains on compact Riemannian manifolds by using the well-posedness results for the analyzed linear transmission problems. Various mapping properties of these operators are also obtained and used to describe the weak solutions of the Poisson problems with Dirichlet and Neumann conditions for the nonsmooth Brinkman system in terms of such potentials. Finally, we combine the well-posedness results of the Poisson problems of Dirichlet, transmission, and mixed type for the nonsmooth Brinkman system with a fixed point theorem in order to show the existence of a weak solution of the Poisson problem of Dirichlet, transmission, or mixed type for the (nonlinear) Darcy-Forchheimer-Brinkman system with L ∞ coefficients in L 2 -based Sobolev spaces in Lipschitz domains on compact Riemannian manifolds of dimension m ∈ { 2 , 3 } .

Though the boundary layer formation in the chemotactic process has been observed in experiment (cf. [63] ), the mathematical study on the boundary layer solutions of chemotaxis models is just in its infant stage. Apart from the sophisticated theoretical tools involved in the analysis, how to impose/derive physical boundary conditions is a state-of-the-art in studying the boundary layer problem of chemotaxis models. This paper will proceed with a previous work [24] in one dimension to establish the convergence of boundary layer solutions of the Keller–Segel model with singular sensitivity in a two-dimensional space (half-plane) with respect to the chemical diffusion rate denoted by e ≥ 0 . Compared to the one-dimensional boundary layer problem, there are many new issues arising from multi-dimensions such as possible Prandtl type degeneracy, curl-free preservation and well-posedness of large-data solutions. In this paper, we shall derive appropriate physical boundary conditions and gradually overcome these barriers and hence establish the convergence of boundary layer solutions of the singular Keller–Segel system in the half-plane as the chemical diffusion rate vanishes. Specially speaking, we justify that the boundary layer converges to the outer layer (solution with e = 0 ) plus the inner layer as e → 0 , where both outer and inner layer profiles are precisely derived and well understood. By doing this, the structure of boundary layer solutions is clearly characterized. We hope that our results and methods can shed lights on the understanding of underlying mechanisms of the boundary layer patterns observed in the experiment for chemotaxis such as the work by Tuval et al. [63] , and open a new window in the future theoretical study of chemotaxis models.

Motivated by the paper Gerard-Varet and Dormy (2010) [6] [JAMS, 2010] about the linear ill-posedness for the Prandtl equations around a shear flow with exponential decay in normal variable, and the recent study of well-posedness on the Prandtl equations in Sobolev spaces, this paper aims to extend the result in [6] to the case when the shear flow has general decay. The key observation is to construct an approximate solution that captures the initial layer to the linearized problem motivated by the precise formulation of solutions to the inviscid Prandtl equations.

Reduction theory has played a major role in the study of Hamiltonian systems. Whilst the Hamilton–Jacobi theory is one of the main tools to integrate the dynamics of certain Hamiltonian problems and a topic of research on its own. Moreover, the construction of several symplectic integrators relies on approximations of a complete solution of the Hamilton–Jacobi equation. The natural question that we address in this paper is how these two topics (reduction and Hamilton–Jacobi theory) fit together. We obtain a reduction and reconstruction procedure for the Hamilton–Jacobi equation with symmetries, even in a generalized sense to be clarified below. Several applications and relations to other reduction of the Hamilton–Jacobi theory are shown in the last section of the paper. It is remarkable that as by-product we obtain a generalization of the Ge–Marsden reduction procedure [18] and the results in [17] . Quite surprisingly, the classical ansatze available in the literature to solve the Hamilton–Jacobi equation (see [2] , [19] ) are also particular instances of our framework.

The main result of this paper provides uniform decay rates obtained for the energy function associated with a three-dimensional structural acoustic model described by coupled system consisting of the wave equation and plate equation with the coupling on the interface between the acoustic chamber and the wall. The uniform stabilization is achieved by introducing a nonlinear dissipation acting via boundary forces applied at the edge of the plate and viscous or boundary damping applied to the wave equation. The results obtained in this paper extend, to the non-analytic, hyperbolic-like setting, the results obtained previously in the literature for acoustic problems modeled by structurally damped plates (governed by analytic semigroups). As a bypass product, we also obtain optimal uniform decay rates for the Euler Bernoulli plate equations with nonlinear boundary dissipation acting via shear forces only and without (i) any geometric conditions imposed on the domain ,(ii) any growth conditions at the origin imposed on the nonlinear function. This is in contrast with the results obtained previously in the literature ([22] and references therein).

This paper studies a dissipative heat equation with Delta sources of non-linear strength located on a periodic lattice. The model arises from intracellular waves in continuum excitable media with discrete release sites. Due to the presence of Delta sources, the solution of the model has discontinuous spatial derivatives. We focus on the bistable regime of the model, determined by the decay strength parameter a and the separation distance L between release sites, in which the model admits exactly three L-periodic steady states. We establish the existence of pulsating waves spatially connecting them. For the case of waves connecting two stable L-periodic steady states, the uniqueness and global exponential stability of pulsating waves are shown. Also a new technique is introduced to find the fine structure of the tails of pulsating waves.

This paper focuses on the Dysthe equation which is a higher order approximation of the water waves system in the modulation (Schrodinger) regime and in the infinite depth case. We first review the derivation of the Dysthe and related equations. Then we study the initial-value problem. We prove a small data global well-posedness and scattering result in the critical space L 2 ( R 2 ) . This result is sharp in view of the fact that the flow map cannot be C 3 continuous below L 2 ( R 2 ) . Our analysis relies on linear and bilinear Strichartz estimates in the context of the Fourier restriction norm method. Moreover, since we are at a critical level, we need to work in the framework of the atomic space U S 2 and its dual V S 2 of square bounded variation functions. We also prove that the initial-value problem is locally well-posed in H s ( R 2 ) , s > 0 . Our results extend to the finite depth version of the Dysthe equation.

We consider the initial boundary value problem for the isentropic compressible Navier-Stokes equations with cylindrical symmetry. The existence of boundary layers is well-known when the shear viscosity vanishes. In this paper, we derive explicit Prandtl type boundary layer equations and prove the global in time stability of the boundary layer profile together with the optimal convergence rate of the vanishing shear viscosity limit without any smallness assumption on the initial and boundary data.

In this paper we study the structure of positive vector solutions for nonlinear Schrodinger systems with 3 components when all interspecies interaction forces are positive and large while all intraspecies interaction forces are positive and fixed. We will show that the structure strongly depends on some relation of large interspecies interaction forces.

We consider the following k-coupled nonlinear Schrodinger systems: { − Δ u j + λ j u j = μ j u j 3 + ∑ i = 1 , i ≠ j k β i , j u i 2 u j in R N , u j > 0 in R N , u j ( x ) → 0 as | x | → + ∞ , j = 1 , 2 , ⋯ , k , where N ≤ 3 , k ≥ 3 , λ j , μ j > 0 are constants and β i , j = β j , i ≠ 0 are parameters. There have been intensive studies for the above systems when k = 2 or the systems are purely attractive ( β i , j > 0 , ∀ i ≠ j ) or purely repulsive ( β i , j 0 , ∀ i ≠ j ); however very few results are available for k ≥ 3 when the systems admit mixed couplings and the components are organized into groups, i.e., there exist ( i 1 , j 1 ) and ( i 2 , j 2 ) such that β i 1 , j 1 > 0 and β i 2 , j 2 0 . In this paper we give the first systematic and an (almost) complete study on the existence of ground states when the systems admit mixed couplings and the components are organized into groups. We first divide these systems into repulsive-mixed and total-mixed cases. In the first case we prove nonexistence of ground states. In the second case we give a necessary condition for the existence of ground states and also provide estimates for Morse index. The key idea is the block decomposition of the systems (optimal block decompositions, eventual block decompositions), and the measure of total interaction forces between different blocks. Finally the assumptions on the existence of ground states are shown to be optimal in some special cases.

In this paper, we study large time behavior of complex-valued solutions to nonlinear Klein-Gordon equation with a gauge invariant quadratic nonlinearity in two spatial dimensions. To find a possible asymptotic behavior, we consider the final value problem. It turns out that one possible behavior is a linear solution with a logarithmic phase correction as in the real-valued case. However, the shape of the logarithmic correction term has one more parameter which is also given by the final data. In the real case the parameter is constant so one cannot see its effect. However, in the complex case it varies in general. The one dimensional case is also discussed., Comment: 25 papges, 2 figures

We make systematic developments on Lawson-Osserman constructions relating to the Dirichlet problem (over unit disks) for minimal surfaces of high codimension in their 1977 Acta paper. In particular, we show the existence of boundary functions for which infinitely many analytic solutions and at least one nonsmooth Lipschitz solution exist simultaneously. This newly-discovered amusing phenomenon enriches the understanding on the Lawson-Osserman philosophy., Supercedes arXiv:1610.08162. To appear in the Journal de Math\'ematiques Pures et Appliqu\'ees

The Boussinesq a b c d system is a 4-parameter set of equations posed in R t × R x , originally derived by Bona, Chen and Saut [11] , [12] as first order 2-wave approximations of the incompressible and irrotational, two dimensional water wave equations in the shallow water wave regime, in the spirit of the original Boussinesq derivation [17] . Among many particular regimes, depending each of them in terms of the value of the parameters ( a , b , c , d ) present in the equations, the generic regime is characterized by the setting b , d > 0 and a , c 0 . If additionally b = d , the a b c d system is Hamiltonian. The equations in this regime are globally well-posed in the energy space H 1 × H 1 , provided one works with small solutions [12] . In this paper, we investigate decay and the scattering problem in this regime, which is characterized as having (quadratic) long-range nonlinearities, very weak linear decay O ( t − 1 / 3 ) because of the one dimensional setting, and existence of non scattering solutions (solitary waves). We prove, among other results, that for a sufficiently dispersive a b c d systems (characterized only in terms of parameters a , b and c), all small solutions must decay to zero, locally strongly in the energy space, in proper subset of the light cone | x | ≤ | t | . We prove this result by constructing three suitable virial functionals in the spirit of works [27] , [28] , and more precisely [42] (valid for the simpler scalar “good Boussinesq” model), leading to global in time decay and control of all local H 1 × H 1 terms. No parity nor extra decay assumptions are needed to prove decay, only small solutions in the energy space.

In this paper we elucidate the connection between various notions of differentiability in the Wasserstein space: some have been introduced intrinsically (in the Wasserstein space, by using typical objects from the theory of Optimal Transport) and used by various authors to study gradient flows, Hamiltonian flows, and Hamilton–Jacobi equations in this context. Another notion is extrinsic and arises from the identification of the Wasserstein space with the Hilbert space of square-integrable random variables on a non-atomic probability space. As a consequence, the classical theory of well-posedness for viscosity solutions for Hamilton–Jacobi equations in infinite-dimensional Hilbert spaces is brought to bear on well-posedness for Hamilton–Jacobi equations in the Wasserstein space.

A general theory of partial balayage on Riemannian manifolds is developed, with emphasis on compact manifolds. Partial balayage is an operation of sweeping measures, or charge distributions, to a prescribed density, and it is closely related to (construction of) quadrature domains for subharmonic functions, growth processes such as Laplacian growth and to weighted equilibrium distributions. Several examples are given in the paper, as well as some specific results. For instance, it is proved that, in two dimensions, harmonic and geodesic balls are the same if and only if the Gaussian curvature of the manifold is constant., Comment: 68 pages

In this paper, we study the expansion problem which arises as two-dimensional (2D) pseudo-steady supersonic flow turns a sharp corner and expands into vacuum. The problem catches interaction of a centered simple wave and a backward planar rarefaction wave, which is deduced a Goursat problem for 2D self-similar Euler equations for compressible flow. By the methods of characteristic decomposition and invariant regions, we get the hyperbolicity in the wave interaction domain and prior C 1 estimates of solutions to the Goursat problem. The global solution up to the interface of gas with vacuum to the expansion problem is obtained constructively.

We consider the zero viscosity limit of the incompressible Navier–Stokes equations with non-slip boundary condition in R + 3 for the initial vorticity located away from the boundary. Unlike 2-D case, this kind of data is not analytic in ( x , y ) for y close to the boundary in 3-D. Maekawa proved the local in time convergence of the Navier–Stokes equations in R + 2 to the Euler equations outside a boundary layer and to the Prandtl equations in the boundary layer. His proof used Cauchy–Kowaleskaya theorem, where the vorticity formulation and Euler–Prandtl decomposition play an important role. In this paper, we generalize Maekawa's result to R + 3 by using a direct energy method, which may be applicable for general physical domain.

In this paper the spectral properties of Dirac operators $A_\eta$ with electrostatic $\delta$-shell interactions of constant strength $\eta$ supported on compact smooth surfaces in $\mathbb{R}^3$ are studied. Making use of boundary triple techniques a Krein type resolvent formula and a Birman-Schwinger principle are obtained. With the help of these tools some spectral, scattering, and asymptotic properties of $A_\eta$ are investigated. In particular, it turns out that the discrete spectrum of $A_\eta$ inside the gap of the essential spectrum is finite, the difference of the third powers of the resolvents of $A_\eta$ and the free Dirac operator $A_0$ is trace class, and in the nonrelativistic limit $A_\eta$ converges in the norm resolvent sense to a Schr\"odinger operator with an electric $\delta$-potential of strength $\eta$., Comment: 32 pages

In [1] , even if f only belongs to L 1 ( Ω ) , we proved that the assumption | f ( x ) | ≤ Q a ( x ) ∈ L 1 ( Ω ) , Q ∈ R + , implies the existence of a bounded solution of boundary value problems, whose simplest example is the linear problem { − div ( M ( x ) ∇ u ) + a ( x ) u = f ( x ) , in Ω ; u = 0 , on ∂ Ω ; where Ω is a bounded open set of R N and M is a bounded elliptic matrix. In this paper, we continue the study of the regularizing effect on the solutions, in some nonlinear Dirichlet problems, when we have L q interaction between the coefficient a ( x ) and the datum f ( x ) (i.e. Q is now a function in L q ( Ω ) ).

This paper is concerned with initial-boundary-value problems (IBVPs) for a class of nonlinear Schrodinger equations posed either on a half line R + or on a bounded interval ( 0 , L ) with nonhomogeneous boundary conditions. For any s with 0 ≤ s 5 / 2 and s ≠ 1 / 2 , it is shown that the relevant IBVPs are locally well-posed if the initial data lie in the L 2 -based Sobolev spaces H s ( R + ) in the case of the half line and in H s ( 0 , L ) on a bounded interval, provided the boundary data are selected from H l o c ( 2 s + 1 ) / 4 ( R + ) and H ( s + 1 ) / 2 l o c ( R + ) , respectively. (For s > 1 2 , compatibility between the initial and boundary conditions is also needed.) Global well-posedness is also discussed when s ≥ 1 . From the point of view of the well-posedness theory, the results obtained reveal a significant difference between the IBVP posed on R + and the IBVP posed on ( 0 , L ) . The former is reminiscent of the theory for the pure initial-value problem (IVP) for these Schrodinger equations posed on the whole line R while the theory on a bounded interval looks more like that of the pure IVP posed on a periodic domain. In particular, the regularity demanded of the boundary data for the IBVP on R + is consistent with the temporal trace results that obtain for solutions of the pure IVP on R , while the slightly higher regularity of boundary data for the IBVP on ( 0 , L ) resembles what is found for temporal traces of spatially periodic solutions.

This paper deals with the finite-time blow-up phenomena of classical solutions for Vlasov/Navier–Stokes equations under suitable assumptions on the initial configurations. We show that a solution to the coupled kinetic-fluid system may be initially smooth, however, it can become singular in a finite period of time. We provide a simple idea of showing the finite time blow up of classical solutions to the coupled system which has not been studied so far. We also obtain analogous results for related systems, such as isentropic compressible Navier–Stokes equations, two-phase fluid equations consisting of pressureless Euler equations and Navier–Stokes equations, and thick sprays model.

In this paper we consider a stationary variational inequality with nonconstant gradient constraint and we prove the existence of solution of a Lagrange multiplier, assuming that the bounded open not necessarily convex set O has a smooth boundary. If the gradient constraint g is sufficiently smooth and satisfies ?g 2 =0 and the source term belongs to L 8 (O), we are able to prove that the Lagrange multiplier belongs to L q (O), for 1 < q < 8, even in a very degenerate case. Fixing q=2, the result is still true if ?g 2 is bounded from above by a positive sufficiently small constant that depends on O, q, minO??g and maxO??g. Without the restriction on the sign of ?g 2 we are still able to find a Lagrange multiplier, now belonging to L 8 (O) ' . We also prove that if we consider the variational inequality with coercivity constant d and constraint g, then the family of solutions (? d ,u d ) d > 0 of our problem has a subsequence that converges weakly to (? 0 ,u 0 ), which solves the transport equation., FCTO -Fuel Cell Technologies Office(UID/MAT/00013/2013), info:eu-repo/semantics/publishedVersion

In this paper we prove global existence and uniqueness of solutions to the stochastic logarithmic Schrodinger equation with linear multiplicative noise. Our approach is mainly based on the rescaling approach and the method of maximal monotone operators. In addition, uniform estimates of solutions in the energy space H-1(R-d) and in an appropriate Orlicz space are also obtained here. (C) 2016 Elsevier Masson SAS. All rights reserved.

The study of Fourier coefficients of meromorphic modular forms dates back to Ramanujan, who, together with Hardy, studied the reciprocal of the weight 6 Eisenstein series. Ramanujan conjectured a number of further identities for other meromorphic modular forms and quasi-modular forms which were subsequently established by Berndt, Bialek, and Yee. In this paper, we place these identities into the context of a larger family by making use of Poincare series introduced by Petersson and a new family of Poincare series which we construct here and which are of independent interest. In addition we establish a number of new explicit identities. In particular, we give the first examples of Fourier expansions for meromorphic modular form with third-order poles and quasi-meromorphic modular forms with second-order poles.

The double ramification hierarchy is a new integrable hierarchy of hamiltonian PDEs introduced recently by the first author. It is associated to an arbitrary given cohomological field theory. In this paper we study the double ramification hierarchy associated to the cohomological field theory formed by Witten's $r$-spin classes. Using the formula for the product of the top Chern class of the Hodge bundle with Witten's class, found by the second author, we present an effective method for a computation of the double ramification hierarchy. We do explicit computations for $r=3,4,5$ and prove that the double ramification hierarchy is Miura equivalent to the corresponding Dubrovin--Zhang hierarchy. As an application, this result together with a recent work of the first author with Paolo Rossi gives a quantization of the $r$-th Gelfand--Dickey hierarchy for $r=3,4,5$., Comment: v3: 26 pages (accepted in Journal de Math\'ematiques Pures et Appliqu\'ees)

We study decreasing rearrangements of functions defined on (possibly non-smooth) metric measure spaces with Ricci curvature bounded below by $K>0$ and dimension bounded above by $N\in (1,\infty)$ in a synthetic sense, the so called $CD(K,N)$ spaces. We first establish a Polya-Szego type inequality stating that the $W^{1,p}$-Sobolev norm decreases under such a rearrangement and apply the result to show sharp spectral gap for the $p$-Laplace operator with Dirichlet boundary conditions (on open subsets), for every $p\in (1,\infty)$. This extends to the non-smooth setting a classical result of B��rard-Meyer and Matei; remarkable examples of spaces fitting out framework and for which the results seem new include: measured-Gromov Hausdorff limits of Riemannian manifolds with Ricci$\geq K>0$, finite dimensional Alexandrov spaces with curvature$\geq K>0$, Finsler manifolds with Ricci$\geq K>0$. In the second part of the paper we prove new rigidity and almost rigidity results attached to the aforementioned inequalities, in the framework of $RCD(K,N)$ spaces, which seem original even for smooth Riemannian manifolds with Ricci$\geq K>0$., 33 pages. Final version published in Journal de Math\'ematiques Pures et Appliqu\'ees

In this paper, various constant curved minimal two-spheres in the complex hyperquadric Q n are obtained, which exhaust all the minimal homogeneous ones in Q n . Their geometric quantities of Gauss curvature, Kahler angle and the length of the second fundamental form are expressed explicitly. As an application, the totally geodesic, constant curved holomorphic and antiholomorphic two-spheres in Q n are completely classified.

The paper is concerned with the asymptotic behavior of positive least energy vector solutions to nonlinear Schrodinger systems with mixed couplings which arise from models in Bose–Einstein condensates and nonlinear optics. We show that due to mixed attractive and repulsive interactions the least energy solutions exhibit new interesting component-wise pattern formations, including co-existence of partial synchronization and segregation. The novelty of our approach is the successful use of multiple scaling to carry out a refined asymptotic analysis of convergence to a multiply scaled limiting system.

This work is the continuation of the recent paper (Danchin, 2010) [9] devoted to the density-dependent incompressible Euler equations. Here we concentrate on the well-posedness issue in Besov spaces of type B ∞ , r s embedded in the set of Lipschitz continuous functions, a functional framework which contains the particular case of Holder spaces C 1 , α and of the endpoint Besov space B ∞ , 1 1 . For such data and under the non-vacuum assumption, we establish the local well-posedness and a continuation criterion in the spirit of that of Beale, Kato and Majda (1984) [2] . In the last part of the paper, we give lower bounds for the lifespan of a solution. In dimension two, we point out that the lifespan tends to infinity when the initial density tends to be a constant. This is, to our knowledge, the first result of this kind for the density-dependent incompressible Euler equations.

In this paper we study the asymptotic phase space energy distribution of solution of the Schr\"{o}dinger equation with a time-dependent random potential. The random potential is assumed to be with slowly decaying correlations. We show that the Wigner transform of a solution of the random Schr\"{o}dinger equation converges in probability to the solution of a radiative transfer equation. Moreover, we show that this radiative transfer equation with long-range coupling has a regularizing effect on its solutions. Finally, we give an approximation of this equation in term of a fractional Laplacian. The derivations of these results are based on an asymptotic analysis using perturbed-test-functions, martingale techniques, and probabilistic representations., Comment: 33 pages

In this paper, we study the problem of controllability of Schrodinger equation. We prove that the system is exactly controllable in infinite time to any position. The proof is based on an inverse mapping theorem for multivalued functions. We show also that the system is not exactly controllable in finite time in lower Sobolev spaces.

Let X be a symmetric right process, and let Z = { Z t , t ⩾ 0 } be a multiplicative functional of X that is the product of a Girsanov transform, a Girsanov transform under time-reversal and a continuous Feynman–Kac transform. In this paper we derive necessary and sufficient conditions for the strong L 2 -continuity of the semigroup { T t , t ⩾ 0 } given by T t f ( x ) = E x [ Z t f ( X t ) ] , expressed in terms of the quadratic form obtained by perturbing the Dirichlet form of X in the appropriate way. The transformations induced by such Z include all those treated previously in the literature, such as Girsanov transforms, continuous and discontinuous Feynman–Kac transforms, and generalized Feynman–Kac transforms.

The present paper proves that all limit points of sequences of renormalized solutions of the Boltzmann equation in the limit of small, asymptotically equivalent Mach and Knudsen numbers are governed by Leray solutions of the Navier-Stokes equations. This convergence result holds for hard cutoff potentials in the sense of H. Grad, and therefore completes earlier results by the same authors [Invent. Math. 155, 81-161(2004)] for Maxwell molecules., 56 pages, LaTeX, a few typos have been corrected, a few remarks added, one uncited reference removed

In 1992, C. Vallee showed that the metric tensor field C = ∇ Θ T ∇ Θ associated with a smooth enough immersion Θ : Ω → R 3 defined over an open set Ω ⊂ R 3 necessarily satisfies the compatibility relation CURL Λ + COF Λ = 0 in Ω , where the matrix field Λ is defined in terms of the field U = C 1 / 2 by Λ = 1 det U { U ( CURL U ) T U − 1 2 ( tr [ U ( CURL U ) T ] ) U } . The main objective of this paper is to establish the following converse: If a smooth enough field C of symmetric and positive-definite matrices of order three satisfies the above compatibility relation over a simply-connected open set Ω ⊂ R 3 , then there exists, typically in spaces such as W loc 2 , ∞ ( Ω ; R 3 ) or C 2 ( Ω ; R 3 ) , an immersion Θ : Ω → R 3 such that C = ∇ Θ T ∇ Θ in Ω. This global existence theorem thus provides an alternative to the fundamental theorem of Riemannian geometry for an open set in R 3 , where the compatibility relation classically expresses that the Riemann curvature tensor associated with the field C vanishes in Ω. The proof consists in first determining an orthogonal matrix field R defined over Ω, then in determining an immersion Θ such that ∇ Θ = R C 1 / 2 in Ω, by successively solving two Pfaff systems. In addition to its novelty, this approach thus also possesses a more “geometrical” flavor than the classical one, as it directly seeks the polar factorization ∇ Θ = RU of the immersion gradient in terms of a rotation R and a pure stretch U = C 1 / 2 . This approach also constitutes a first step towards the analysis of models in nonlinear three-dimensional elasticity where the rotation field is considered as one of the primary unknowns.

In this paper we discuss the problem of computing and analyzing the static equilibrium of a nonrigid water tank. Specifically, we fix the amount of water contained in the tank, modelled as a membrane. In addition, there are rigid obstacles that constrain the deformation. This amounts to a nonconvex variational problem. We derive the optimality system and its interpretation in terms of equilibrium of forces. A second-order sensitivity analysis, allowing to compute derivatives of solutions and a second-order Taylor expansion of the cost function, is performed, in spite of the fact that the cost function is not twice differentiable. We also study the finite elements discretization, introduce a decomposition algorithm for the numerical computation of the solution, and display numerical results.

RésuméDans cet article, nous énonçons et résolvons des problèmes d'approximation de Padé nouveaux et très généraux dont les solutions s'expriment à l'aide de séries hypergéométriques : par spécialisation, ces séries permettent de retrouver l'irrationalité de ζ(3), d'une infinité de ζ(2n+1), n entier ⩾1, et essentiellement tous les résultats de cette nature déjà présents dans la littérature. Nous présentons également deux nouvelles applications diophantiennes de notre méthode.AbstractIn this paper, we present and solve some very general new Padé approximant problems, whose solutions can be expressed with hypergeometric series. These series appear in the proofs of the irrationality of ζ(3), of infinitely many ζ(2n+1), and in essentially all results of this kind in the literature. We also prove two new Diophantine results with this method.

In this paper we introduce a new transportation distance between non-negative measures inside a domain Ω. This distance enjoys many nice properties, for instance it makes the space of non-negative measures inside Ω a geodesic space without any convexity assumption on the domain. Moreover we will show that the gradient flow of the entropy functional ∫ Ω [ ρ log ( ρ ) − ρ ] d x with respect to this distance coincides with the heat equation, subject to the Dirichlet boundary condition equal to 1.

In computational electromagnetic and acoustic scattering, the unbounded Euclidean space R 3 is often modeled by a bounded domain with an absorbing boundary condition. One possible approach to create such absorbing boundary condition is to surround the computational domain by a non-reflecting artificial sponge layer that absorbs quickly the scattered waves. This approach is called the method of a Perfectly Matched Layer (PML). In this paper we prove that such absorbing boundary layers can be obtained by using complex Riemannian metric gij. We show that the boundary layer is non-reflecting when g is flat, that is, the curvature tensor of the complex metric gij is zero. This fact gives an invariant formulation for the absorbing boundary layers as well as give us new kind of absorbing boundary layers for Maxwell and Helmholtz equations. Moreover, we show that all Perfectly Matched Layers, that is, absorbing boundary layers obtained through a complexification of coordinates corresponds to flat complex manifolds. Finally, we discuss the relation of the absorbing boundary layers and the complex scaling technique, developed by Sjostrand and Zworski for the study of scattering poles.

International audience; In this paper we study generalized solutions (in the Brenier's sense) for the Euler equations. We prove that uniqueness holds in dimension one whenever the pressure field is smooth, while we show that in dimension two uniqueness is far from being true. In the case of the two-dimensional disc we study solutions to Euler equations where particles located at a point $x$ go to $-x$ in a time $\pi$, and we give a quite general description of the (large) set of such solutions. As a byproduct, we can construct a new class of classical solutions to Euler equations in the disc.