Nonlinear second order evolution inclusions with noncoercive viscosity term

In this paper we deal with a second order nonlinear evolution inclusion, with a nonmonotone, noncoercive viscosity term. Using a parabolic regularization (approximation) of the problem and a priori bounds that permit passing to the limit, we prove that the problem has a solution.     

Nonlocal higher order evolution equations

In this article, we study the asymptotic behaviour of solutions to the nonlocal operator u _t (x, t) = (?1) ^n?1 (J * Id ? 1) ⁿ (u(x, t)), x ∈ ? ^N , which is the nonlocal analogous to the higher order local evolution equation v _t = (?1) ^n?1(Δ) ⁿ v. We prove that the solutions of the nonlocal problem converge to the solution of the higher order problem with the right-hand side given by powers of the Laplacian when the kernel J is rescaled in an appropriate way. Moreover, we prove that solutions to both equations have the same asymptotic decay rate as t goes to infinity.     

Solutions of nonlinear evolution inclusions

In this paper, the solution of the nonlinear evolution inclusion problem of the form u^′(t)+B(t,u(t))∋f(t) is studied. In this problem, the operators are of type (M) or type (S₊), which are different from those of pseudo-monotone operators that had been studied by many authors. At the same time, we study the perturbation problem. In fact, many kinds of evolution equations can be generalized by this problem. The former results are improved and generalized by our conclusions, and we will give more applications.     

Optimal switching for systems governed by nonlinear evolution equations

Optimal switching for a nonlinear evolution equation is considered. A type of approximating problems with value functions {u_u0<u<u₀} is studied. We prove that each u_u is the unique viscositY solution of corresponding Hami1ton-Jacoby system, u_u convergesto the value function u of the original problem as µ goes to 0, and an optimal control is determined.     

Globally smooth solutions of quasilinear hyperbolic systems in diagonal form

We prove the existence of a large class of globally smooth solutions of the Cauchy problem for the system of n equations u_t + Λ(x, t, u)u_x = 0, where Λ is a diagonal matrix. We show that, under certain monotonicity conditions on both Λ and the initial data u₀, the solution u will be locally Lipschitz continuous at positive times. Since u₀ is a function of locally bounded variation, our result thus provides both for the smoothing of discontinuities in u₀ as well as for the global preservation of smoothness. The global existence results from an a priori estimate of $\text{?u}\text{?x}$, which we obtain by a device which enables us to effectively uncouple the system of equations for $\text{?u}\text{?x}$. Finally, we prove a partial converse which demonstrates that our hypotheses are not overly restrictive.     

On the structure of solutions to a class of quasilinear elliptic Neumann problems

We study the structure of positive solutions to the equation ?^mΔ_mu-u^m-1+f(u)=0 with homogeneous Neumann boundary condition. First, we show the existence of a mountain-pass solution and find that as ?→0⁺ the mountain-pass solution develops into a spike-layer solution. Second, we prove that there is an uniform upper bound independent of ? for any positive solution to our problem. We also present a Harnack-type inequality for the positive solutions. Finally, we show that if 1<m?2 holds and ? is sufficiently large, any positive solution must be a constant.     

The exterior Dirichlet problem for a class of fourth order elliptic equations

In some exterior domain G of the Euclidian p-space $\text{R}$^p the Dirichlet boundary value problem is considered for the equation (L + κ²)²u = f, where L is a uniformly elliptic operator and κ is a real number different from 0. It can be shown that each solution u of this equation splits into u = x_l?_lu₁ + u₂, where u₁ and u₂ satisfy Heimholte equations. Asymptotic conditions for u are formulated by imposing Sommerfeld radiation conditions on u₁ and u₂. If u₁ and u₂ are assumed to satisfy the same radiation condition, we prove a “Fredholm alternative theorem.” If u₁ and u₂ satisfy different radiation conditions, existence and uniqueness of the solution can be shown, provided the space dimension p is greater than 2.     

On the local existence of solutions to a class of nonconvex evolution inclusions

We prove the local existence of solutions to the Cauchy problemx'∈-?_F V(x)+F(x+f(t,x),x(0)=x ₀, where? _FV is the Fréchet subdifferential of a functionV with aψ-monotone subdifferential of order 2,F is an upper semicontinuous set-valued map contained in the Fréchet subdifferential of aφ- convex function of order two andf is a Carathéodory mapping.     

Entrainment of frequency in evolution equations

The existence of periodic solutions near resonance is discussed using elementary methods for the evolution equation ·u = Au + ?f(t, u) when the linear problem is totally degenerate (e^2πA = I) and the period of f is entrained with ? (T = 2π(1 + ?μ)). The approach is to solve the periodicity equation u(T,p,?) = p for an element p(?) in D, the domain of A, as a perturbation from an approximate solution p₀. p₀ is a solution of the nonlinear boundary value problem 2πμAp + ∝₀^2πe^?Asf(s, e^Asp) ds = 0 obtained from the periodicity equation by dividing by ?, applying the entrainment assumption, and letting ? → 0. Once p₀ is known, the conventional inverse function theorem is applied in a slightly unconventional manner. Two particular cases where results are obtained are u_t = u_x + ?{g(u) ? h(t, x)} with g strongly monotone and

$\text{d}\text{dt}\text{v}\text{w}\text{=}\text{0}\text{d}\text{dx}\text{d}\text{dx}\text{0}\text{v}\text{w}\text{+ ?}\text{v}{}^3\text{h(t,x)}$

, where in both cases D is a certain class of 2π-periodic functions of x.     

Higher order of fully discreate solution for parabolic problem inL ∞

In this work, we approximate the solution of initial boundary value problem using a Galerkin-finite element method for the spatial discretization, and Implicit Runge-Kutta methods for the time stepping. To deal with the nonlinear termf(x, t, u), we introduce the well-known extrapolation sheme which was used widely to prove the convergence inL ₂-norm. We present computational results showing that the optimal order of convergence arising underL ₂-norm will be preserved inL _∞-norm.     

Existence of β-martingale solutions of stochastic evolution functional equations of parabolic type with measurable locally bounded coefficients

We prove a theorem on the existence of ??-martingale solutions of stochastic evolution functional equations of parabolic type with Borel measurable locally bounded coefficients. A ??-martingale solution of a stochastic evolution functional equation is understood as a martingale solution of a stochastic evolution functional inclusion constructed on the basis of the equation. We find sufficient conditions for the existence of ??-martingale solutions that do not blow up in finite time.     

Accuracy of discrete schemes for a class of abstract evolution inequalities

For a triple of Hilbert spaces {V, H, V*}, we study a discrete and a semidiscrete scheme for an evolution inclusion of the form u′(t) + A(t)u(t) + ??(t, u(t)) ? f(t), u(0) = u ₀, t ∈ (0, T], where the pair {A(t), ?(t, ·)} consists of a family of nonlinear operators from V into V* and a family of proper convex lower semicontinuous functionals with common effective domain D(?) ? V. The discrete scheme is a combination of the Galerkin method with perturbations and the implicit Euler method. Under conditions on the data providing the existence and uniqueness of the solution of the problem in the space H ¹(0, T; V) ∩ W _∞ ¹ (0, T;H), we obtain an abstract estimate for the method error in the energy norm of first-order accuracy with respect to the time increment. By way of application, we consider a problem with an obstacle inside the domain, for which we obtain an optimal estimate of the accuracy of two implicit schemes (standard and new) on the basis of the finite element method.     

Periodic solutions for doubly nonlinear evolution equations

We discuss the existence of periodic solution for the doubly nonlinear evolution equation A(u^′(t))+∂?(u(t))∋f(t) governed by a maximal monotone operator A and a subdifferential operator ∂? in a Hilbert space H. As the corresponding Cauchy problem cannot be expected to be uniquely solvable, the standard approach based on the Poincaré map may genuinely fail. In order to overcome this difficulty, we firstly address some approximate problems relying on a specific approximate periodicity condition. Then, periodic solutions for the original problem are obtained by establishing energy estimates and by performing a limiting procedure. As a by-product, a structural stability analysis is presented for the periodic problem and an application to nonlinear PDEs is provided.     

Global well-posedness and scattering for the focusing energy-critical nonlinear Schrödinger equations of fourth order in the radial case

We consider the focusing energy-critical nonlinear Schrödinger equation of fourth order , d?5. We prove that if a maximal-lifespan radial solution obeys supt∈I‖Δu(t)_‖2<‖ΔW_‖2, then it is global and scatters both forward and backward in time. Here W denotes the ground state, which is a stationary solution of the equation. In particular, if a solution has both energy and kinetic energy less than those of the ground state W at some point in time, then the solution is global and scatters.     

Schauder estimates for a class of second order elliptic operators on a cube

We consider a class of second order elliptic operators on a d-dimensional cube S_d. We prove that if the coefficients are of class C^k+δ(S_d), with k=0,1 and δ∈(0,1), then the corresponding elliptic problem admits a unique solution u belonging to C^k+2+δ(S_d) and satisfying non-standard boundary conditions involving only second order derivatives.     

Nonoscillatory solutions of forced differential equations of the second order

We prove that under certain assumptions if lu + b²u = 0, b > 0, where l is an operator Schrödinger equation defined in section 2 and u exp(?br) → 0 as r → ∞ then u?L². This result shows in particular that the resonant states behave exactly as O(r^?1, exp(ikr)) as r → ∞. It also provides dichotomy theorems in a general situation.     

Singular sets for curvature equation of order k

We consider the removability of singular sets for the curvature equations of the form Hk[u]=ψ, which is determined by the kth elementary symmetric function, in an n-dimensional domain Ω. We prove that, for 1?k?n−1 and a compact set K whose (n−k)-dimensional Hausdorff measure is zero, any generalized solution to the curvature equation on Ω?K is always extendable to a generalized solution on the whole domain Ω.     

The heterogeneous Allen-Cahn equation in a ball: Solutions with layers and spikes

Let u? be a single layered radially symmetric unstable solution of the Allen-Cahn equation −?²Δu=u(u−a(|x|))(1−u) over the unit ball with Neumann boundary conditions. Based on our estimate of the small eigenvalues of the linearized eigenvalue problem at u? when ? is small, we construct solutions of the form u?+v?, with v? non-radially symmetric and close to zero in the unit ball except near one point x₀ such that |x₀| is close to a nondegenerate critical point of a(r). Such a solution has a sharp layer as well as a spike.     

Existence and non-existence of solutions for a class of Monge-Ampère equations

We study the boundary value problems for Monge-Ampère equations: detD²u=e−u in Ω⊂Rn, n?1, u_|∂Ω=0. First we prove that any solution on the ball is radially symmetric by the argument of moving plane. Then we show there exists a critical radius such that if the radius of a ball is smaller than this critical value there exists a solution, and vice versa. Using the comparison between domains we can prove that this phenomenon occurs for every domain. Finally we consider an equivalent problem with a parameter detD²u=e−tu in Ω, u_|∂Ω=0, t?0. By using Lyapunov-Schmidt reduction method we get the local structure of the solutions near a degenerate point; by Leray-Schauder degree theory, a priori estimate and bifurcation theory we get the global structure.     

The radiation condition at infinity for the high-frequency Helmholtz equation with source term: a wave-packet approach

We consider the high-frequency Helmholtz equation with a given source term, and a small absorption parameter α>0. The high-frequency (or: semi-classical) parameter is ?>0. We let ? and α go to zero simultaneously. We assume that the zero energy is non-trapping for the underlying classical flow. We also assume that the classical trajectories starting from the origin satisfy a transversality condition, a generic assumption.Under these assumptions, we prove that the solution u^? radiates in the outgoing direction, uniformly in ?. In particular, the function u^?, when conveniently rescaled at the scale ? close to the origin, is shown to converge towards the outgoing solution of the Helmholtz equation, with coefficients frozen at the origin. This provides a uniform version (in ?) of the limiting absorption principle.Writing the resolvent of the Helmholtz equation as the integral in time of the associated semi-classical Schrödinger propagator, our analysis relies on the following tools: (i) for very large times, we prove and use a uniform version of the Egorov Theorem to estimate the time integral; (ii) for moderate times, we prove a uniform dispersive estimate that relies on a wave-packet approach, together with the above-mentioned transversality condition; (iii) for small times, we prove that the semi-classical Schrödinger operator with variable coefficients has the same dispersive properties as in the constant coefficients case, uniformly in ?.