Abstract

For a rapidly spatially oscillating nonlinearity 𝑔 we compare solutions 𝑢𝜖 of non-Newtonian filtration equation 𝜕𝑡𝛽(𝑢𝜖)𝐷(|𝐷𝑢𝜖|𝑝2𝐷𝑢𝜖+𝜑(𝑢𝜖)𝐷𝑢𝜖)+𝑔(𝑥,𝑥/𝜖,𝑢𝜖)=𝑓(𝑥,𝑥/𝜖) with solutions 𝑢0 of the homogenized, spatially averaged equation 𝜕𝑡𝛽(𝑢0)𝐷(|𝐷𝑢0|𝑝2𝐷𝑢0+𝜑(𝑢0)𝐷𝑢0)+𝑔0(𝑥,𝑢0)=𝑓0(𝑥). Based on an 𝜀-independent a priori estimate, we prove that ||𝛽(𝑢𝜖)𝛽(𝑢0)||𝐿1(Ω)𝐶𝜖𝑒𝜌𝑡 uniformly for all 𝑡0. Besides, we give explicit estimate for the distance between the nonhomogenized 𝐴𝜖 and the homogenized 𝐴0 attractors in terms of the parameter 𝜖; precisely, we show fractional-order semicontinuity of the global attractors for 𝜖0dist𝐿1(Ω)(𝐴𝜖,𝐴0)𝐶𝜖𝛾.

1. Introduction

This paper is devoted to the study of nonlinear parabolic equations related to the 𝑝-Laplacian operator. The problems related to such type of equation arise in many applications in the fields of mechanics, physics, and biology (non-Newtonian fluids, gas flow in porous media, spread of biological populations, etc.).

For example, in the study of the water in filtration through porous media, Darcy’s linear relation 𝑉=𝐾(𝜏)𝜔𝑥() satisfactorily describes flow conditions provided that the velocities are small. Here 𝑉𝛼=𝐾(𝜏)𝜔𝑥(1.1) represents the velocity of the water, 𝑢𝑡div|𝑢|𝑚||||𝑢𝑝2𝑢=𝑓(𝑥,𝑡,𝑢).() is the volumetric water (moisture) content, is the hydraulic conductivity, and 𝑝[3/2,2) is the total potential, which can be expressed as the sum of a hydrostatic potential 𝑚>𝑝1 and a gravitational potential |𝑢|𝜎𝑢=𝑣: 𝜕𝑡𝛽(𝑣)+div𝐵(𝑥,𝑡,𝑣,𝑣)+𝐵0(𝑥,𝑡,𝑣,𝑣)=0(1.2)𝐵. However, from the physical point of view, (𝐵0) fails to describe the flow for large velocities. To get a more accurate description of the flow in this case, several nonlinear versions of (𝑡) have been proposed. One of these versions is 𝛽, which leads us to the equation𝐵, For the first time, nonlinear relationship instead of Darcy`s relation are suggested by Dupuit and Forchheimer. This modification, known as Darcy-Forchheimer's relation, has led to much research from experimental, theoretical, and numerical points of view. For example, Darcy-Forchheimer equations are widely used in reservoir engineering and other subsurface applications.

One-dimensional (by spatial variable) variant of (𝐵0) with 𝑓 and 𝑔 arises in the study of a turbulent flow of a gas in a one-dimensional porous medium. This phenomenon was first described by Leibenson in [1]. One of the first papers devoted to the existence problem for such type of equation was an initiating paper by Raviart [2]. In [3], the author investigated the regularity problem for more general case (which includes Leibenson's equation after changing 𝜕𝑡𝛽𝑢0||𝐷𝐷𝑢0||𝑝2𝐷𝑢0𝑢+𝜑0𝐷𝑢0+𝑔0𝑥,𝑢0=𝑓0(𝑥)(1.3)) of non-Newtonian equation as𝐴𝜖 with some restriction on the functions 𝜕𝑡𝛽(𝑢𝜖||)𝐷𝐷𝑢𝜖||𝑝2𝐷𝑢𝜖+𝜑(𝑢𝜖)𝐷𝑢𝜖𝑥+𝑔𝑥,𝜖,𝑢𝜖𝑥=𝑓𝑥,𝜖,(1.4) and 𝑔(𝑥,𝑥/𝜖,𝑢𝜖). Earlier, in [4] authors have investigated existence and the stabilization of solutions (as 𝑓(𝑥,𝑥/𝜖)) of nonlinear parabolic equation of mentioned type describing certain models related to turbulent flows. Also note that there is an extensive literature devoted to the existence, regularity, and the large-time behavior of solutions of (1.2) under various conditions on functions ||𝛽(𝑢𝜖)𝛽(𝑢0)||𝐿1(Ω)𝐶𝜖𝑒𝜌𝑡𝑡0 and 𝑔 (see [210] and references therein).

In this paper we show that, under natural assumptions on the terms 𝑓 and 𝑧=𝑥/𝜖., the longtime behavior of solutions of the equation𝐴𝜖 can be described in terms of the global attractor 𝐴0 of the associated dynamical system related to the equation𝜖0 where functions 𝑢𝜖𝑡Δ𝑢𝜖+𝑓(𝑢𝜖)=𝜖(𝑥)(1.5) and , are constructed according to some ideas previously presented in [11], where authors carry out a quantitative comparison with the averaged or homogenized equations, in particular for quasiperiodic inhomogeneities with Diophantine frequencies (see [11, pages 176–180]). Note that a quantitative homogenization aims at determining a specific rate of convergence. Method of the construction of the homogenized equation allows us to assert that 𝜕𝑡𝛽(𝑢𝜖𝑎𝑥)𝐷𝜖||𝐷𝑢𝜖||𝑝2+𝜑(𝑢𝜖)𝐷𝑢𝜖𝑥+𝑔𝑥,𝜖,𝑢𝜖𝑥=𝑓𝑥,𝜖,(1.6) (Theorem 3.1 below) uniformly for all 𝜑(𝑠). As we mentioned above, this result requires [𝑎,𝑏] and 𝑢(𝑥,𝑡)𝑣(𝑥,𝑡) to depend quasiperiodically on the rapid space variable 𝑢(𝑥,𝑡) At the same time, being interested in quantitative strong convergence not only of individual trajectories but also of global attractors, we also show that global attractors 𝑣(𝑥,𝑡) tend to attractor 𝑡. in a suitable sense providing fractional-order semicontinuity of the global attractors for 𝛽. On a related note, the author would like to mention several results on homogenization of the attractor for nonlinear parabolic equations.

In [12] the Cauchy problem for parabolic equations on Riemannian manifolds with complicated microstructure has been considered and a connection between global attractors of the initial problem of the homogenized one has been established. The asymptotical behavior of the global attractor of the boundary value problem for semilinear equation𝜑. was investigated in [13] and it was shown that this tended in a suitable sense to the finite-dimensional weak global attractor of some system of a parabolic p.d.e. coupled with an o.d.e.

Also it is necessary to note the papers in [1416], where the media properties are assumed to be oscillatory, focusing on the homogenization of the attractor for semilinear parabolic equations (see [14]) and systems (see [15])𝜕𝑡||||𝛽(𝑢)𝐷𝐷𝑢𝑝2𝑢𝐷𝑢+𝜑(𝑢)𝐷𝑢+𝑔(𝑥,𝑢)=𝑓(𝑥),(1.7)|Γ=0,Γ=𝜕Ω×(0,𝑇),(1.8)𝑢(𝑥,0)=𝑢0(𝑥),(1.9) and of the quasilinear parabolic equations (see [16]).

Note that previous approaches on homogenization of attractors are limited to certain types of nonlinear equations. In particular, they assume the main part of the equation to be linear or monotone. The consideration of equation (1.4) is a first step in the investigation of the more general equation𝐷=𝜕𝑥, where the media properties are assumed to be oscillatory. This often arises in porous media flows [17]. We also note the paper in [18] where authors deal with the homogenization problem for a one-dimensional parabolic PDE with random stationary mixing coefficients in the presence of a large zero-order term and show that the family of solutions of the studied problem converges in law.

Our work is inspired, on one hand, by results of the studies in [7, 8] which are devoted to the existence and regularity of the attractor and, on the other hand, by the paper in [11] that is related to quantitative homogenization of the global attractor for reaction-diffusion systems. For proof of the existence we use sketch of the proof of corresponding result; however, compared to the studies in [7, 8] we consider the equation with an additional nonlinear term Ω=(𝑎,𝑏). Note that this term prevents us from applying the result from the paper in [4] which is often used to prove the uniqueness of the solution. Here we use a technique (see [19]) which is based on the fact that the interval (𝐻1) may be divided into subintervals where the sign of the “difference” 𝑢0(𝑥) (𝛽(𝑢0) and 𝐿2(Ω), are solutions with the same initial condition) does not change for fixed (𝐻2) Also note that the results of the paper in [11] cannot be directly applied to prove homogenization of the attractor for (1.4) because of nonlinear terms 𝛽(𝜁) and 𝑅

Our paper is organized as follows. In Section 2 we consider the Dirichlet problem𝑅, where 𝛽(0)=0.(𝐻3) under the following hypotheses𝑔(,𝜁)𝜁𝑐1|𝜁|𝑘+𝑐2, and 𝜕𝜁𝑔(,𝜁)>𝜆 are in 𝑔(,0)=0)𝑐3𝑐4 is an increasing locally Lipschitzian function from sign(𝜁)𝑔(𝑥,𝜁) to 𝑐3|𝛽(𝜁)|𝑞1𝑐4 with (𝑥,𝑡)Ω×𝑅𝑞>2.(𝐻4)𝜑(𝜁)𝐶1, (without loss of generality we suppose that 𝜑(𝜁)0,𝜑(0)=0, and there exist positive constants 𝑀 and |𝜁|𝑀 such that |𝜑(𝜁)|(|𝜁|𝑙+𝑐)(𝛽(𝜁))𝛼𝑙,𝑐0 for a. e. 𝛼1. and (𝐻5)𝑓(𝑥) is a function from the space 𝐿(Ω), such that 𝑝4. and, for each 𝐿2- and (𝐻6), 𝑀 almost everywhere, for some |𝜁|𝑀, and |𝑔(𝑥,𝜁)𝑓(𝑥)|(|𝜁|𝑙1+𝑐)(𝛽(𝜁))𝛼1𝑙1,𝑐0𝛼11.(𝐻1)-(𝐻5),𝛽(𝑢𝜖𝑢)𝛽0𝐿1(Ω)𝐶𝜖𝑒𝜌𝑡.(1.10)𝑔

Under the mentioned condition we show the existence and uniqueness of the solution and the existence of the 𝜖0global attractor.

Further, in order to show more regularity of the solution we suppose the following condition.dist𝐿1(Ω)(𝐴𝜖,𝐴0)𝐶𝜖𝛾,For each 𝐴𝜖 and 𝐴0𝜖 almost everywhere, for some 𝑔(𝑥,𝑢) and 𝑓(𝑥)

Finally in Section 3, we derive, under conditions 𝑔(𝑥,𝑥/𝜖,𝑢𝜖) the following estimate:𝑓(𝑥,𝑥/𝜖)

Also, under additional condition on 𝜕𝑡𝛽𝜂||||(𝑢)𝐷𝐷𝑢𝑝2𝑢+𝜑(𝑢)+𝜂𝐷𝑢+𝑔(𝑥,𝑢)=𝑓(𝑥),(2.1)|Γ=0,Γ=𝜕Ω×(0,𝑇),(1.8)𝑢(𝑥,0)=𝑢0(𝑥),(1.9) we prove fractional-order semicontinuity of the global attractors for 𝛽𝜂𝐶1(𝑅), namely, the validity of the estimation 𝛽𝜂(0)=0, where 𝛽𝜂𝛽 and 𝐶loc(𝑅) are global attractors of the semigroups generated by the problems (1.4), (1.8), (1.9) and (1.3), (1.8), (1.9), respectively.

2. Existence and Uniqueness

First, we suppose that 𝑀𝜂>𝛽𝜂>𝜂, in equation (1.4) is a constant. This leads us to the problem (1.7), (1.8), (1.9) where |𝛽𝜂||𝛽|. and (𝑢0𝜂)𝜂>0 are denoted as 𝐶0(Ω) and 𝑢0𝜂𝑢0 correspondingly.

We use the standard regularization of the problem (1.7), (1.8), (1.9):Ω where the sequence ||𝑢0𝜂||𝐿2(Ω),||𝛽𝜂(𝑢0𝜂)||𝐿2(Ω)𝑐 is such that 𝑐>04.1(5.2), in 𝜑, 𝛽𝜂, and |𝛽𝜂(𝑢𝜂)|𝑘𝛽𝜂(𝑢𝜂),

Let (𝐻1)-(𝐻5) be a sequence in 𝜂(0,1) such that ||𝑢𝜂||𝐿(𝜏,𝑇;𝐿(Ω))𝑐(𝜏,𝑇), almost everywhere in ||||𝛽𝜂𝑢𝜂||||𝐿(0,𝑇;𝐿2(Ω))𝐿𝑞(𝑄𝑇)||||𝑢𝑐(𝑇),𝜂||||𝐿𝑝(0,𝑇;𝑊01,𝑝(Ω))𝑐(𝑇).(2.2) and |𝛽𝜂(𝑢𝜂)|𝑘𝛽𝜂(𝑢𝜂), with constant 1𝑑𝑘+2𝑑𝑡Ω||𝛽𝜂𝑢𝜂||𝑘+2𝑑𝑥+(𝑘+1)Ω||𝐷𝑢𝜂||𝑝2𝑢+𝜑𝜂||𝛽+𝜂𝜂𝑢𝜂||𝑘𝛽𝜂𝑢𝜂×||𝐷𝑢𝜂||2𝑑𝑥+Ω𝑔𝑥,𝑢𝜂||𝛽𝜂𝑢𝜂||𝑘𝛽𝜂𝑢𝜂𝑑𝑥=Ω||𝛽𝑓(𝑥)𝜂𝑢𝜂||𝑘𝛽𝜂𝑢𝜂𝑑𝑥.(2.3). To show the solvability of the problem (1.8), (1.9), (2.1) we use the result for the classical solvability in the large given by Ladyžhenskaya et al. (see [20, Theorem 𝛽𝜂(𝑢𝜂) ch. VI]).

Our proof is based on a priori estimates and is similar to that in [7]. First, using the properties of 𝜑(𝑢𝜂) and 1𝑑𝑘+2𝑑𝑡Ω||𝛽𝜂𝑢𝜂||𝑘+2𝑑𝑥+Ω𝑔𝑥,𝑢𝜂||𝛽𝜂𝑢𝜂||𝑘𝛽𝜂𝑢𝜂𝑑𝑥Ω||𝛽𝑓(𝑥)𝜂𝑢𝜂||𝑘𝛽𝜂𝑢𝜂𝑑𝑥.(2.4) and then multiplying the equation by (𝐻2) we deduce, analogously to [7, 8], the following lemma.

Lemma 2.1. Under the hypotheses (𝐻3) for any 1𝑑𝑘+2𝑑𝑡Ω||𝛽𝜂𝑢𝜂||𝑘+2𝑑𝑥+𝑐3Ω||𝛽𝜂𝑢𝜂||𝑘+𝑞𝑑𝑥Ω||𝛽𝑓(𝑥)𝜂𝑢𝜂||𝑘𝛽𝜂𝑢𝜂𝑑𝑥+𝑐4Ω||𝛽𝜂𝑢𝜂||𝑘𝛽𝜂𝑢𝜂𝑑𝑥.(2.5) the following estimates hold: (𝐻5)1𝑑𝑘+2𝑑𝑡Ω||𝛽𝜂𝑢𝜂||𝑘+2𝑑𝑥+𝑐3Ω||𝛽𝜂𝑢𝜂||𝑘+𝑞𝑑𝑥𝑐(𝑓)+𝑐4Ω||𝛽𝜂𝑢𝜂||𝑘+1𝑑𝑥.(2.6)

Proof. Multiplying equation (2.1) by 𝛼0>0 and integrating by parts, we get 𝜆0>0
By virtue of the positivity of the 𝑑||||𝛽𝑑𝑡𝜂𝑢𝜂||||𝐿𝑘+2(Ω)+𝜆0||||𝛽𝜂𝑢𝜂||||𝐿𝑞1𝑘+2(Ω)𝛼0,(2.7) and 𝛽𝑛𝑢𝜂𝐿𝑘+2(Ω)𝛼0𝜆01/(𝑞1)+1𝜆0(𝑞2)𝑡1/(𝑞2)=𝑐(𝑡).(2.8) we have 𝑘
By conditions 𝑡𝜏>0, and ||||𝛽𝜂𝑢𝜂||||(𝑡)𝐿(Ω)𝑐(𝜏),(2.9) we derive ||||𝑢𝜂||||(𝑡)𝐿(Ω)𝛽max𝜂1||𝛽(𝑐(𝜏)),𝜂1||(𝑐(𝜏))=𝛿𝜂.(2.10)
Consequently, from 𝛽𝜂 we obtain 𝛽
Further, using Holder's inequality on both sides of the latter inequality, we deduce that there exist two constants 𝐶loc(), and 𝛿𝜂 such that which implies from Ghidaglia's lemma [9] that 𝜂+
As 𝛿𝜂 in (2.8), and for max(𝛽𝜂1(𝑐(𝜏)),|𝛽𝜂1(𝑐(𝜏))|), we have ||𝑢𝜂||𝐿(𝜏,𝑇;𝐿(Ω))𝑐(𝜏). which implies that 𝑘=0
Since [0,𝑇], converges to ||||𝛽𝜂𝑢𝜂||||𝐿(0,𝑇;𝐿2(Ω))𝐿𝑞(𝑄𝑇)𝑐(𝑇).(2.11) in 𝑢𝜂 then the sequence 𝑄𝑡 is bounded in 1𝑘+2𝑡0Ω𝜕𝑡𝛽𝜂𝑢𝜂𝑢𝜂𝑑𝑥𝑑𝜃+(𝑘+1)𝑡0Ω||𝐷𝑢𝜂||𝑝2𝑢+𝜑𝜂||+𝜂𝐷𝑢𝜂||2+𝑑𝑥𝑑𝜃𝑡0Ω𝑔𝑥,𝑢𝜂𝑢𝜂𝑑𝑥𝑑𝜃=𝑡0Ω𝑓(𝑥)𝑢𝜂𝑑𝑥𝑑𝜃.(2.12) as 1𝑘+2𝑡0Ω𝜕𝑡𝛽𝜂𝑢𝜂𝑢𝜂𝑑𝑥𝑑𝜃+(𝑘+1)𝑡0Ω||𝐷𝑢𝜂||𝑝2𝑢+𝜑𝜂||+𝜂𝐷𝑢𝜂||2+𝑑𝑥𝑑𝜃𝑡0Ω𝑔𝑥,𝑢𝜂𝑢𝜂𝑑𝑥𝑑𝜃=𝑡0Ω𝑓(𝑥)𝑢𝜂1𝑑𝑥𝑑𝜃+𝑘+2𝑡0Ω𝛽𝜂𝑢𝜂𝜕𝑡𝑢𝜂𝑑𝑥𝑑𝜃.(2.13). Thus 1𝑘+2Ω𝛽𝜂𝑢𝜂(𝑢𝑡)𝜂(1𝑡)𝑑𝑥𝑘+2Ω𝛽𝜂𝑢0𝜂𝑢0𝜂𝑑𝑥+(𝑘+1)𝑡0Ω||𝐷𝑢𝜂||𝑝2𝑢+𝜑𝜂||+𝜂𝐷𝑢𝜂||2𝑑𝑥𝑑𝜃+𝑡0Ω𝑔𝑥,𝑢𝜂𝑢𝜂=𝑑𝑥𝑑𝜃𝑡0Ω𝑓(𝑥)𝑢𝜂1𝑑𝑥𝑑𝜃+𝑘+2𝑡0Ω𝑑𝐵𝜂𝑢𝜂𝑑𝜏𝑑𝑥𝑑𝜃,(2.14) is bounded by 𝐵𝜂(𝑠)=𝑠0𝛽𝜂(𝜃)𝑑𝜃. which is finite. Whence 𝛽𝜂(0)=0
On the other hand, taking 𝛽𝜂>𝜂>0, in (2.3), using Holder’s inequality, and integrating over 𝐵𝜂(𝑠)𝛽𝜂(𝑠)𝑠 we obtain the second estimate of the statement of Lemma 2.1 as 1𝑘+2Ω𝛽𝜂𝑢𝜂(𝑡)𝑢𝜂(1𝑡)𝑑𝑥𝑘+2Ω𝛽𝜂𝑢0𝜂𝑢0𝜂𝑑𝑥+(𝑘+1)𝑡0Ω||𝐷𝑢𝜂||𝑝2𝑢+𝜑𝜂||+𝜂𝐷𝑢𝜂||2𝑑𝑥𝑑𝜃+𝑡0Ω𝑔𝑥,𝑢𝜂𝑢𝜂𝑑𝑥𝑑𝜃𝑡0Ω𝑓(𝑥)𝑢𝜂1𝑑𝑥𝑑𝜃+𝑘+2Ω𝐵𝜂𝑢𝜂(1𝑡)𝑑𝑥𝑘+2Ω𝐵𝜂𝑢0𝜂𝑑𝑥𝑡0Ω𝑓(𝑥)𝑢𝜂+1𝑑𝑥𝑑𝜃𝑘+2Ω𝛽𝜂𝑢𝜂(𝑡)𝑢𝜂(𝑡)𝑑𝑥,(2.15)
Further, in order to prove the latter estimate of Lemma 2.1, we multiply (2.1) by (𝑘+1)𝑡0Ω||𝐷𝑢𝜂||𝑝2𝑢+𝜑𝜂||+𝜂𝐷𝑢𝜂||2𝑑𝑥𝑑𝜃+𝑡0Ω𝑔𝑥,𝑢𝜂𝑢𝜂𝑑𝑥𝑑𝜃𝑡0Ω𝑓(𝑥)𝑢𝜂1𝑑𝑥𝑑𝜃+𝑘+2Ω𝛽𝜂𝑢0𝜂𝑢0𝜂𝑑𝑥.(2.16) and integrate over 𝜕𝜁𝑔>𝜆: 𝑔(,𝜁)>𝜆𝜁
Therefore, (𝑘+1)𝑡0Ω||𝐷𝑢𝜂||𝑝2𝑢+𝜑𝜂||+𝜂𝐷𝑢𝜂||2𝑑𝑥𝑑𝜃𝑡0Ω||𝑢𝜂||2(𝜆+1)𝑑𝑥𝑑𝜃+𝑡0Ω||𝑓||21𝑑𝑥𝑑𝜃+𝑘+2Ω||𝛽𝜂𝑢0𝜂||21𝑑𝑥+𝑘+2Ω||𝑢0𝜂||2𝑑𝑥.(2.17)
Hence, 𝑇0Ω||𝐷𝑢𝜂||𝑝𝑑𝑥𝑑𝜃𝑐(𝑇).(2.18) where 𝑐
Now, taking into account that 1𝜏𝑡𝑡+𝜏Ω||𝐷𝑢𝜂||𝑝𝑑𝑥𝑑𝜏𝑐forarbitrary𝜏.(2.19) and (𝑡,𝑡+𝜏) we conclude that (0,𝑡). Thus, (𝐻1)-(𝐻5) or 𝑐𝑖
From condition 𝜂(0,1) we obtain that ||𝑢𝜂||𝐿(𝜏,𝑇;𝑊01,𝑝(Ω))𝑐(𝜏,𝑇), and, consequently, 𝑇𝜏Ω𝛽𝜂𝑢𝜂2||𝜕𝑡𝑢𝜂||2𝑑𝑥𝑑𝜃𝑐(𝜏,𝑇),𝑡𝑡+𝜏Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||2𝑑𝑥𝑑𝜃𝑐(𝜏)(𝑡𝜏).(2.20)
Therefore, 𝜕𝑡𝑢𝜂
Thereby assertion follows.

Corollary 2.2. Under condition of Lemma 2.1 there exists [𝑠,𝜏+𝑡], such that 𝜏𝑡𝑠,

(It is sufficient to carry out the proof of Lemma 2.1 using integration on the interval 𝑠𝑡+𝜏Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||2𝑑𝑥𝑑𝜃𝑠𝑡+𝜏Ω𝐷||𝐷𝑢𝜂||𝑝2𝑢+𝜑𝜂+𝜂𝐷𝑢𝜂𝜕𝑡𝑢𝜂+𝑑𝑥𝑑𝜃𝑠𝑡+𝜏Ω𝑔𝑥,𝑢𝜂𝜕𝑡𝑢𝜂𝑑𝑥𝑑𝜃=𝑠𝑡+𝜏Ω𝑓(𝑥)𝜕𝑡𝑢𝜂𝑑𝑥𝑑𝜃.(2.21) instead of 𝑠𝑡+𝜏Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||2𝑑𝑥𝑑𝜃+𝑠𝑡+𝜏Ω||𝐷𝑢𝜂||𝑝2𝑢+𝜑𝜂+𝜂𝐷𝑢𝜂𝐷𝜕𝑡𝑢𝜂+𝑑𝑥𝑑𝜃𝑠𝑡+𝜏Ω𝜕𝑡𝐺(𝑥,𝑢)𝑑𝑥𝑑𝜃=𝑠𝑡+𝜏Ω𝜕𝑡𝑓(𝑥)𝑢𝜂𝑑𝑥𝑑𝜃,(2.22) and taking into account (2.8).)

Lemma 2.3. Under the hypotheses 𝐺(,𝑠)=𝑠0𝑔(,𝜁)𝑑𝜁. there exist constants (𝐻4), such that for any 𝑠𝑡+𝜏Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||2+𝑑𝑥𝑑𝜃𝑠𝑡+𝜏𝑑1𝑑𝜃𝑝+2Ω||𝐷𝑢𝜂(||𝜃)𝑝1𝑑𝑥+2Ω𝜑𝑢𝜂(||𝜃)𝐷𝑢𝜂(||𝜃)21𝑑𝑥+2Ω𝜂||𝐷𝑢𝜂(||𝜃)2+𝑑𝑥Ω𝐺𝑥,𝑢𝜂(𝜃)𝑑𝑥Ω𝑓(𝑥)𝑢𝜂=1(𝜃)𝑑𝜃2𝑠𝑡+𝑟Ω𝜑𝑢𝜂×||(𝜃)𝐷𝑢𝜂||(𝜃)2𝜕𝑡𝑢𝜂1𝑑𝑥𝑑𝜃2𝑠𝑡+𝑟Ω||𝑢𝜂||𝑙+𝑐𝛽(𝑠)𝛼||𝐷𝑢𝜂||2||𝜕𝑡𝑢𝜂||𝑑𝑥𝑑𝜃.(2.23) the following estimates hold: 𝛽𝜂(𝑢𝜂)𝜂

Proof. Multiplying (2.1) by 𝑢𝜂[𝛿,𝛿] and integrating over 𝛿 where 𝛽𝜂 we get 𝛽𝜂𝐿, Then, using integration by parts, we derive 𝐿 where 𝛽(𝜁)
Hence, using condition [𝛿,𝛿]. we get 12𝑡+𝜏𝑠Ω||𝑢𝜂||𝑙+𝑐𝛽(𝑠)𝛼||𝐷𝑢𝜂||2||𝜕𝑡𝑢𝜂||𝜖𝑑𝑥𝑑𝜃2𝑠𝑡+𝜏Ω𝛽𝜂𝑢𝜂𝛼||𝜕𝑡𝑢𝜂||21𝑑𝑥𝑑𝜃+2𝜖𝑠𝑡+𝜏Ω||𝑢𝜂||𝑙+𝑐2||𝐷𝑢𝜂||4𝜖𝑑𝑥𝑑𝜃2𝐿𝛼1𝑠𝑡+𝜏Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||21𝑑𝑥𝑑𝜃+2𝜖𝑠𝑡+𝜏Ω||𝑢𝜂||𝑙+𝑐2||𝐷𝑢𝜂||4𝑑𝑥𝑑𝜃.(2.24)
Further, taking into account that 𝑠𝑡+𝜏Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||2+𝑑𝑥𝑑𝜃𝑠𝑡+𝜏𝑑1𝑑𝜃𝑝+2Ω||𝐷𝑢𝜂(||𝜃)𝑝1𝑑𝑥+2Ω𝜑𝑢𝜂(||𝜃)𝐷𝑢𝜂(||𝜃)21𝑑𝑥+2Ω𝜂||𝐷𝑢𝜂(||𝜃)2+𝑑𝑥Ω𝐺𝑥,𝑢𝜂(𝜃)𝑑𝑥Ω𝑓(𝑥)𝑢𝜂𝜖(𝜃)𝑑𝑥𝑑𝜃2𝐿𝛼1𝑠𝑡+𝜏Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||21𝑑𝑥𝑑𝜃+2𝜖𝑠𝑡+𝜏Ω||𝑢𝜂||𝑙+𝑐2||𝐷𝑢𝜂||𝑝+1𝑑𝑥𝑑𝜃.(2.25) is uniformly bounded by 12𝑠𝑡+𝜏Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||21𝑑𝑥𝑑𝜃+𝑝+2Ω||𝐷𝑢𝜂||𝑝(𝑡+𝜏)𝑑𝑥+Ω𝜑𝑢𝜂||𝐷𝑢𝜂||2(+1𝑡+𝜏)𝑑𝑥2Ω𝜂||𝐷𝑢𝜂||2(𝑡+𝜏)𝑑𝑥+Ω𝐺𝑥,𝑢𝜂(𝑡+𝜏)𝑑𝑥Ω𝑓(𝑥)𝑢𝜂(1𝑡+𝜏)𝑑𝑥𝑝+2Ω||𝐷𝑢𝜂||𝑝(𝑠)𝑑𝑥+Ω𝜑𝑢𝜂||𝐷𝑢𝜂||21(𝑠)𝑑𝑥+2Ω𝜂||𝐷𝑢𝜂||2+(𝑠)𝑑𝑥Ω𝐺𝑥,𝑢𝜂(𝑠)𝑑𝑥Ω𝑓(𝑥)𝑢𝜂1(𝑠)𝑑𝑥+2𝜖𝑐𝜏.(2.26) ([𝑡,𝑡+𝜏],, where 𝜏1𝑝+2Ω||𝐷𝑢𝜂||𝑝(𝑡+𝜏)𝑑𝑥+𝜏Ω𝜑𝑢𝜂||𝐷𝑢𝜂||2(1𝑡+𝜏)𝑑𝑥+2𝜏Ω𝜂||𝐷𝑢𝜂||2(1𝑡+𝜏)𝑑𝑥𝑝+2𝑡𝑡+𝜏Ω||𝐷𝑢𝜂||𝑝(𝑠)𝑑𝑥𝑑𝑠+𝑡𝑡+𝜏Ω𝜑𝑢𝜂||𝐷𝑢𝜂||2+1(𝑠)𝑑𝑥𝑑𝑠2𝑡𝑡+𝜏Ω𝜂||𝐷𝑢𝜂||2(𝑠)𝑑𝑥𝑑𝑠+𝑡𝑡+𝜏Ω𝐺𝑥,𝑢𝜂(𝑠)𝑑𝑥𝑑𝑠+𝑡𝑡+𝜏Ω𝑓(𝑥)𝑢𝜂(𝑠)𝑑𝑥𝑑𝑠𝜏Ω𝑓(𝑥)𝑢𝜂(𝑡+𝜏)𝑑𝑥𝜏Ω𝐺𝑥,𝑢𝜂(𝑡+𝜏)𝑑𝑥+𝑐(𝜏)𝜏.(2.27) is the bound in the proof of Lemma 2.1), it is possible to choose 𝑔, so that (𝜆/2)|𝜁|2<𝐺(,𝜁)𝑐|𝜁|𝑘+(𝜆/2)|𝜁|2+𝑐, where 1𝑝+2Ω||𝐷𝑢𝜂||𝑝(𝑡+𝜏)𝑑𝑥+Ω𝜑𝑢𝜂||𝐷𝑢𝜂||2(1𝑡+𝜏)𝑑𝑥𝜏1𝑝+2𝑡𝑡+𝜏Ω||𝐷𝑢𝜂||𝑝(𝑠)𝑑𝑥𝑑𝑠+𝑡𝑡+𝜏Ω𝜑𝑢𝜂||𝐷𝑢𝜂||2+1(𝑠)𝑑𝑥𝑑𝑠2𝑡𝑡+𝜏Ω𝜂||𝐷𝑢𝜂||2(𝑠)𝑑𝑥𝑑𝑠+𝑡𝑡+𝜏Ω𝑐||𝑢𝜂||𝑘+𝜆2||𝑢𝜂||2++𝑐𝑑𝑥𝑑𝑠𝑡𝑡+𝜏Ω𝑓(𝑥)𝑢𝜂+(𝑠)𝑑𝑥𝑑𝑠Ω𝑐||𝑢𝜂||𝑘+𝜆2||𝑢𝜂||2++𝑐(𝑡)𝑑𝑥Ω𝑓(𝑥)𝑢𝜂1(𝑡)𝑑𝑥+𝑐(𝜏)𝜏1𝑝+2𝑡𝑡+𝜏Ω||𝐷𝑢𝜂||𝑝1(𝑠)𝑑𝑥𝑑𝑠+2𝜀𝑡𝑡+𝜏Ω||𝐷𝑢𝜂||4+1(𝑠)𝑑𝑥𝑑𝑠2𝜀1𝑡𝑡+𝜏Ω𝜑2𝑢𝜂1(𝑠)𝑑𝑥𝑑𝑠+2𝑡𝑡+𝜏Ω𝜂||𝐷𝑢𝜂||2+(𝑠)𝑑𝑥𝑑𝑠𝑡𝑡+𝜏Ω𝑐||𝑢𝜂||(𝑠)𝑘+𝜆2||𝑢𝜂||(𝑠)2+𝑐𝑑𝑥𝑑𝑠+𝑡𝑡+𝜏Ω||𝑓(𝑥)𝑢𝜂||(𝑠)𝑑𝑥𝑑𝑠+𝑐(𝜏).(2.28) is the Lipschitz constant of 𝛽𝜂𝐿 on [𝛿,𝛿] Therefore, 𝑡𝑡+𝜏Ω𝜑2𝑢𝜂𝑑𝑥𝑑𝑠𝑡𝑡+𝜏Ω𝑢𝜂0𝜑(𝑠)𝑑𝑠2𝑑𝑥𝑑𝑠𝑡𝑡+𝜏Ω||||𝑢𝜂||||𝐿(𝜏,𝑇;𝐿(Ω))||||𝑢𝜂||||𝑙𝐿(𝜏,𝑇;𝐿(Ω))𝐿+𝑐𝛼𝑑𝑥𝑑𝑠.(2.29)
Consequently, 1𝑝Ω||𝐷𝑢𝜂||𝑝(𝑡+𝜏)𝑑𝑥+Ω𝜑𝑢𝜂||𝐷𝑢𝜂||2(𝑡+𝜏)𝑑𝑥𝑐(𝜏)(2.30)
Now, using Lemma 2.1 and Corollary 2.2, we easily deduce 𝑡𝜏.
After integrating over 𝑡𝑡+𝜏Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||2𝑑𝑥𝑑𝜃𝑐(𝜏).(2.31) we derive 𝛽𝜂
Since, by virtue of conditions imposed on the function 𝛽𝜂𝐿,𝑇𝜏Ω𝛽𝜂𝑢𝜂2||𝜕𝑡𝑢𝜂||2𝑑𝑥𝑑𝜃𝑐(𝜏,𝑇).(2.32) then 𝑢𝜂
As we noted above, 𝑢𝜂𝑢 on 𝐿𝑝(0,𝑇;𝑊01,𝑝(Ω)). Hence, 𝑎
Thus, we obtain that 𝑒 for 𝛽𝜂(𝑢𝜂)𝛽(𝑢)
Now returning to (2.23), we have 𝐶([0,𝑇];𝐿2(Ω)),
Also, choosing 𝜕𝑡𝛽𝜂(𝑢𝜂)𝜕𝑡𝛽(𝑢) so that 𝐿2(𝑄), we derive 𝐷((|𝐷𝑢𝜂|𝑝2𝐷𝑢𝜂)𝜒 Thus, the lemma is proved.

Passage to the Limit in (2.1).
Analogously to [7, 8], by estimates from Lemmas 2.1 and 2.3, we deduce that there exist that a subsequence 𝐿𝑝(0,𝑇;𝑊1,𝑝(Ω)). such that𝑔𝑥,𝑢𝜂𝑔(𝑥,𝑢)weaklyin𝐿2(𝑄).(2.33)𝑢𝜂𝑢 weakly in (𝑄)𝐶(𝑄) and (𝑄).𝑔(𝑄)𝐿2(𝑄).,𝑔(𝑥,𝑢)𝑔 strongly in 𝑔(𝑥,𝑢𝜂)𝑔(𝑥,𝑢)𝑎.𝑒1.3 weakly in 𝑔𝑥,𝑢𝜂𝑔(𝑥,𝑢)weaklyin𝐿2(𝑄).(2.34)1.3𝜑(𝜁) weakly in 𝜑𝑢𝜂𝐷𝑢𝜂𝜑(𝑢)𝐷𝑢weaklyin𝐿2(𝜑𝑢𝑄),(2.35)𝜂𝐷𝑢𝜂𝜑(𝑢)𝐷𝑢weaklyin𝐿2𝐷𝜑𝑢(𝑄),𝜂𝐷𝑢𝜂𝐷(𝜑(𝑢)𝐷𝑢)weaklyin𝐿20,𝑇;𝑊1,2.(Ω)(2.36)
Further, it is easy to see that 𝑄𝜑𝑢𝜂||𝐷𝑢𝜂||2||||||𝑑𝑥𝑑𝑡=||||||𝜑(𝑢)𝐷𝑢𝐿2(𝑄)lim𝑛||||||inf𝜑𝑢𝜂𝐷𝑢𝜂||||||𝐿2(𝑄)=𝑄𝜑𝑢𝜂||𝐷𝑢𝜂||2𝑑𝑥𝑑𝑡.(2.37)

Indeed, we know that 𝜒 a.e. Besides, by virtue of the embedding =𝐷(|𝐷𝑢|𝑝2𝐷𝑢)., where (𝐻1)-(𝐻5) is a domain of the solvability, the operator 𝑢(𝑥,𝑡) generated by the expression 𝑢 is bounded. Hence, by the continuity of we obtain that 𝐿𝑝(0,𝑇;𝑊01,𝑝(Ω))𝐿(0,𝑇;𝐿2(Ω))𝐿(𝜏,𝑇;𝑊01,𝑝(Ω)),. By applying a known lemma from [10, (1.1, Lemma 𝜏>0)] we can conclude that𝛽(𝑢)

Arguing similarly, by help of the lemma from [10, (1.1, Lemma 𝐶([0,𝑇];𝐿2(Ω))𝐿𝑞(𝑄))] and conditions imposed on 𝑐, we obtain𝜁𝑔(𝑥,𝜁)+𝑐𝛽(𝜁)

Observe now that from (2.35), in view of the known theorem,𝑎.𝑒.

Consequently, using standard monotonicity argument [10], we derive that𝑥Ω𝑢(𝑥,𝑡)

Therefore, the following theorems hold.

Theorem 2.4. Under hypotheses 𝑣(𝑥,𝑡) the problem (1.7), (1.8), (1.9) has a weak solution 𝑢(𝑥,0)=𝑣(𝑥,0)=𝑢0(𝑥). such that 𝑢(𝑥,𝑡0)𝑣(𝑥,𝑡0),𝑡0(0,𝑇).𝑡>0 for all 𝑇𝜏Ω(𝛽𝜂(𝑢𝜂))2|𝜕𝑡𝑢𝜂|2𝑑𝑥𝑑𝑡𝑐2(𝜏,𝑇), and |𝐷𝑢|𝑝2𝐷𝑢𝐿2(𝜏,𝑇;𝑊01,2(Ω))𝐷𝑢(𝑥,𝑡)𝐶(Ω).

Now, we prove that the solution of the problem is unique.

Lemma 2.5. Suppose that the assumptions of Theorem 2.4 are fulfilled and there exists a constant 𝑡. such that the function 𝑡0 is increasing for 𝑢(𝑥,𝑡0)𝐶1(Ω)[𝑎,𝑏]. Then the solution of the problem (1.7), (1.8), (1.9) is unique.

Proof. Let 𝑢(𝑥,𝑡0)𝑣(𝑥,𝑡0) and (𝑥1,𝑥2) be two solutions of the problem (1.7), (1.8), (1.9) with the same initial condition: 𝑢(𝑥,𝑡0)𝑣(𝑥,𝑡0)>0 Consider the “difference” 𝑢(𝑥𝑖,𝑡)=𝑣(𝑥𝑖,𝑡) where (𝑖=1,2). is an arbitrary point from The above “difference” is a continuously differentiable function for almost all 𝑥2𝑥1𝜕𝑡(𝛽(𝑢)𝛽(𝑣))𝑑𝑥𝑥2𝑥1𝐷||||𝐷𝑢𝑝2||||𝐷𝑢𝐷𝑣𝑝2𝐷𝑣𝑑𝑥𝑥2𝑥1+𝐷(𝜑(𝑢)𝐷𝑢𝜑(𝑣)𝐷𝑣)𝑑𝑥𝑥2𝑥1(𝑔(𝑥,𝑢)𝑔(𝑥,𝑣))𝑑𝑥=0.(2.38), because by virtue of estimation 𝑑𝑑𝑡𝑥2𝑥1||||(𝛽(𝑢)𝛽(𝑣))𝑑𝑥𝐷𝑢𝑝2𝑥𝐷𝑢2+||||,𝑡𝐷𝑣𝑝2𝑥𝐷𝑣2+||||,𝑡𝐷𝑢𝑝2𝑥𝐷𝑢1||||,𝑡𝐷𝑣𝑝2𝑥𝐷𝑣1𝑥,𝑡𝜑(𝑢)𝐷𝑢2𝑥,𝑡+𝜑(𝑣)𝐷𝑣2𝑥,𝑡+𝜑(𝑢)𝐷𝑢1𝑥,𝑡𝜑(𝑣)𝐷𝑣1+,𝑡𝑥2𝑥1𝑔(𝑥,𝑢)+𝑐𝛽(𝑢)𝑔(𝑥,𝑣)𝑐𝛽(𝑣)𝑑𝑥𝑥2𝑥1𝑐𝛽(𝑢)𝑐𝛽(𝑣)𝑑𝑥𝑡=𝑡0=0.(2.39) using the equation, we can conclude that 𝐷𝑢(𝑥1,𝑡0)𝐷𝑣(𝑥1,𝑡0), and consequently 𝐷𝑢(𝑥2,𝑡0)𝐷𝑣(𝑥2,𝑡0), for almost all 𝑢(𝑥𝑖,𝑡)=𝑣(𝑥𝑖,𝑡) We choose (𝑖=1,2), such that 𝑑𝑑𝑡𝑥2𝑥1||||(𝛽(𝑢)𝛽(𝑣))𝑑𝑥𝑡=𝑡0<𝑐𝑥2𝑥1||||(𝛽(𝑢)𝛽(𝑣))𝑑𝑥𝑡=𝑡0.(2.40). Hence, the interval 𝜓(𝑡)=𝑥2𝑥1((𝛽(𝑢)𝛽(𝑣))𝑑𝑥 may be divided into the subintervals where sign of “difference” 𝜓(𝑡)=𝑥2𝑥1|𝛽(𝑢)𝛽(𝑣)|𝑑𝑥 does not change. Let 𝜕𝑡𝛽(𝑢)𝐿2(𝜏,𝑇,𝐿2(Ω)) be an interval such that 𝑡 and 𝜓(𝑡)𝜓(𝑡), Then from (1.7) we obtain that 𝛽(𝑢(𝑥,𝑡))𝛽(𝑣(𝑥,𝑡))0𝑥(𝑥1,𝑥2), By applying Newton-Leibniz formula we have
Since𝜓(𝑡)=𝜓(𝑡)𝑑𝑑𝑡𝑥2𝑥1|||||𝛽(𝑢)𝛽(𝑣)|𝑑𝑥𝑡=𝑡𝑑𝑑𝑡𝑥2𝑥1||||(𝛽(𝑢)𝛽(𝑣))𝑑𝑥𝑡=𝑡(2.41)𝑡.𝑑𝑑𝑡𝑥2𝑥1|||||𝛽(𝑢)𝛽(𝑣)|𝑑𝑥𝑡=𝑡0𝑑𝑑𝑡𝑥2𝑥1||||(𝛽(𝑢)𝛽(𝑣))𝑑𝑥𝑡=𝑡0.(2.42) it follows that 𝑑𝑑𝑡𝑥2𝑥1||𝛽||||||(𝑢)𝛽(𝑣)𝑑𝑥𝑡=𝑡0<𝑐𝑥2𝑥1||||||||𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝑡=𝑡0.(2.43) Note that the functions 𝑢(𝑥,𝑡0)𝑣(𝑥,𝑡0) and 𝑑𝑑𝑡𝑏𝑎|||||𝛽(𝑢)𝛽(𝑣)|𝑑𝑥𝑡=𝑡0<𝑐𝑏𝑎||||||||𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝑡=𝑡0(2.44) are absolutely continuous by virtue of the inclusion 𝛽(𝑢)𝑊1,2(𝜏,𝑇;𝐿2(Ω))). Hence, these functions possess a derivate for almost all 𝑡𝑡𝑑𝑑𝑡𝑏𝑎||||𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝑑𝑡<𝑐𝑡𝑡𝑏𝑎||||𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝑑𝑡,(2.45). Besides, it is obvious that 𝑡>𝑡>𝜏,𝑏𝑎|||||𝛽(𝑢)𝛽(𝑣)|𝑑𝑥𝑡=𝑡<𝑏𝑎|||||𝛽(𝑢)𝛽(𝑣)|𝑑𝑥𝑡=𝑡+𝑐𝑡𝑡𝑏𝑎||||𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝑑𝑡.(2.46) and if 𝑏𝑎||𝛽||||||(𝑢)𝛽(𝑣)𝑑𝑥𝑡=𝑡𝑒𝑐(𝑡𝑡)𝑏𝑎||||||||𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝑡=𝑡.(2.47) for 𝛽(𝑢)𝐶(0,𝑇;𝐿2(Ω))𝑢(𝑥,0)=𝑣(𝑥,0)=𝑢0(𝑥),then, 𝑢(𝑥,𝑡)=𝑣(𝑥,𝑡). Consequently, 𝜁𝑔(𝑥,𝜁)+𝑐𝛽(𝜁) for almost all (𝐻2) From here, without loss of generality, we can assume that 𝛽1(𝑠)
It follows that 𝑅 The same estimation holds for an arbitrary interval on which 𝑅 does not change its sign. Summing up similar inequalities over subintervals, we get (𝐻4) almost everywhere. In view of integrability of both sides of the latter inequality ((𝐻4) we have (𝐻6) where 𝑆𝑡𝐿2(Ω)𝐿2(Ω) or 𝑆𝑡𝑢0=𝑢(𝑡,)
Thus, by Gronwall's lemma, 𝑆𝑡
Now, taking into account that 𝐴 and 𝑊01,𝑝(Ω)𝐿(Ω), we obtain 𝐿2(Ω)., which finishes the proof of uniqueness.

Remark 2.6. Note that the condition on the function 𝑢(𝑥,𝑡). from the lemma can be changed to the following.
𝑢0(𝑥)𝑊1,(Ω) The function 𝛽(𝑢0)𝐿(Ω), is a locally Lipschitzian function from (𝐻2)-(𝐻6) to 𝐷𝑢𝐿(𝑄). (it follows directly from the proof).
In this case, we can exclude the condition |𝐷𝑢𝜂|𝜎𝜕𝑡𝑢𝜂 of Theorem 2.4 and conditions 𝜎 and 𝑇0Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||2||𝐷𝑢𝜂||𝜎𝑑𝑥𝑑𝑡𝑇0Ω𝐷||𝐷𝑢𝜂||𝑝2𝐷𝑢𝜂𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎𝑑𝑥𝑑𝑡𝜂𝑇0Ω𝐷2𝑢𝜂𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎𝑑𝑥𝑑𝑡𝑇0Ω𝐷𝜑𝑢𝜂𝐷𝑢𝜂𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎+𝑑𝑥𝑑𝑡𝑇0Ω𝑔𝑥,𝑢𝜂𝜕𝑓(𝑥)𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎𝑑𝑥𝑑𝑡=0.(2.48) of Theorem 2.8 below.

So, analogously to the corresponding result from [8], we obtain that problem (1.7), (1.8), (1.9) generates a continuous semigroup 𝑇0Ω𝐷||𝐷𝑢𝜂||𝑝2𝐷𝑢𝜂𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎𝑑𝑥𝑑𝑡=(𝑝1)𝑇0Ω||𝐷𝑢𝜂||𝑝2𝐷2𝑢𝜂𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎𝑑𝑥𝑑𝑡=(𝑝1)𝑇0Ω||𝐷𝑢𝜂||𝑝+𝜎2𝐷2𝑢𝜂𝜕𝑡𝑢𝜂𝑑𝑥𝑑𝑡=𝑝1𝑝+𝜎1𝑇0Ω𝐷||𝐷𝑢𝜂||𝑝+𝜎2𝐷𝑢𝜂𝜕𝑡𝑢𝜂=𝑑𝑥𝑑𝑡𝑝1𝑝+𝜎1𝑇0Ω||𝐷𝑢𝜂||𝑝+𝜎2𝐷𝑢𝜂𝐷𝜕𝑡𝑢𝜂=𝑑𝑥𝑑𝑡𝑝1(𝑝+𝜎1)(𝑝+𝜎)𝑇0𝑑𝑑𝑡Ω||𝐷𝑢𝜂||𝑝+𝜎𝑑𝑥𝑑𝑡.(2.49)𝜂𝑇0Ω𝐷2𝑢𝜂𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎𝜂𝑑𝑥𝑑𝑡=𝜎+1𝑇0Ω𝐷||𝐷𝑢𝜂||𝜎𝐷𝑢𝜂𝜕𝑡𝑢𝜂=𝜂𝑑𝑥𝑑𝑡𝜎+1𝑇0Ω||𝐷𝑢𝜂||𝜎𝐷𝑢𝜂𝐷𝜕𝑡𝑢𝜂=𝜂𝑑𝑥𝑑𝑡(𝜎+1)(𝜎+2)𝑇0𝑑𝑑𝑡Ω||𝐷𝑢𝜂||𝜎+2𝑑𝑥𝑑𝑡.(2.50) and the following theorem holds.

Theorem 2.7. Assume that assumptions of Lemma 2.5 are satisfied. Then the semigroup 𝑇0Ω𝐷𝜑𝑢𝜂𝐷𝑢𝜂𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎=1𝑑𝑥𝑑𝑡𝜎+2𝑇0Ω𝜑𝑢𝜂𝜕𝑡||𝐷𝑢𝜂||𝜎+2𝜎𝑑𝑥𝑑𝑡+𝜎+1𝑇0Ω𝜑𝑢𝜂𝜕𝑡𝑢𝜂𝐷||𝐷𝑢𝜂||𝜎𝐷𝑢𝜂=1𝑑𝑥𝑑𝑡𝜎+2𝑇0Ω𝜑𝑢𝜂𝜕𝑡||𝐷𝑢𝜂||𝜎+2𝜎𝑑𝑥𝑑𝑡𝜎+1𝑇0Ω𝜑𝑢𝜂𝐷𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎𝐷𝑢𝜂𝜎𝑑𝑥𝑑𝑡𝜎+1𝑇0Ω𝜑𝑢𝜂𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎+21𝑑𝑥𝑑𝑡=(𝜎+1)(𝜎+2)𝑇0Ω𝜑𝑢𝜂𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎+2+1𝑑𝑥𝑑𝑡(𝜎+1)(𝜎+2)𝑇0Ω𝜕𝑡𝜑𝑢𝜂||𝐷𝑢𝜂||𝜎+2𝜎𝑑𝑥𝑑𝑡𝜎+1𝑇0Ω𝜑𝑢𝜂𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎+2𝑑𝑥𝑑𝑡=𝜎+1𝜎+2𝑇0Ω𝜑𝑢𝜂𝜕𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎+21𝑑𝑥𝑑𝑡+(𝜎+1)(𝜎+2)𝑇0Ω𝜕𝑡𝜑𝑢𝜂||𝐷𝑢𝜂||𝜎+2𝑑𝑥𝑑𝑡𝜎+1𝜎+2𝑇0Ω||𝑢𝜂||𝑙1+𝑐𝛽𝜂𝑢𝜂𝛼1||𝜕𝑡𝑢𝜂||||𝐷𝑢𝜂||𝜎+21𝑑𝑥𝑑𝑡+(×𝜎+1)(𝜎+2)𝑇0Ω𝜕𝑡𝜑𝑢𝜂||𝐷𝑢𝜂||𝜎+2(𝑑𝑥𝑑𝑡𝜎+1)𝜖(𝜎+2)2𝑇0Ω𝛽𝜂𝑢𝜂𝛼1||𝜕𝑡𝑢𝜂||2||𝐷𝑢𝜂||𝜎𝑑𝑥𝑑𝑡(𝜎+1)2𝜖(𝜎+2)𝑇0Ω||𝑢𝜂||𝑙1+𝑐2||𝐷𝑢𝜂||𝜎+41𝑑𝑥𝑑𝑡+×(𝜎+1)(𝜎+2)𝑇0Ω𝜕𝑡𝜑𝑢𝜂||𝐷𝑢𝜂||𝜎+2𝑑𝑥𝑑𝑡(𝜎+1)𝜖𝐿(𝜎+2)2𝛼11𝑇0Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||2||𝐷𝑢𝜂||𝜎𝑑𝑥𝑑𝑡(𝜎+1)2𝜖(𝜎+2)𝑇0Ω||𝑢𝜂||𝑙1+𝑐2||𝐷𝑢𝜂||𝜎+𝑝+2𝑑𝑥𝑑𝑡(𝜎+1)2𝜖(𝜎+2)𝑇0Ω||𝑢𝜂||𝑙1+𝑐2𝑑𝑥𝑑𝑡Ω𝜑𝑢𝜂||𝐷𝑢𝜂||𝜎+2𝑑𝑥Ω𝜑𝑢0||𝐷𝑢0||𝜎+2𝑑𝑥.(2.51) associated with the boundary value problem (1.7), (1.8), (1.9) possesses a maximal attractor 𝛽𝜂(𝑢𝜂) which is bounded in 𝜂 compact, and connected in 𝑢𝜂[𝛿,𝛿]

(For the concepts of absorbing sets and global attractors used here, we refer the reader to [9]).

Under an additional condition we can obtain more regularity for 𝛿

Theorem 2.8. Assume that 𝛽𝜂, 𝛽𝜂𝐿, and the conditions 𝐿 are fulfilled. Then 𝛽(𝜁)

Proof. We prove this fact by multiplying (2.1) on expression [𝛿,𝛿]., taking into account of the arbitrarity of ||||𝑇0Ω𝑔𝑥,𝑢𝜂𝜕𝑓(x)𝑡𝑢𝜂||𝐷𝑢𝜂||𝜎||||||||𝑑𝑥𝑑𝑡𝑇0Ω||𝑢𝜂||𝑙1+𝑐𝛽𝜂𝑢𝜂𝛼1||𝜕𝑡𝑢𝜂||||𝐷𝑢𝜂||𝜎||||𝜖𝑑𝑥𝑑𝑡2𝑇0Ω𝛽𝜂𝑢𝜂𝛼1||𝜕𝑡𝑢𝜂||2||𝐷𝑢𝜂||𝜎1𝑑𝑥𝑑𝑡+2𝜖𝑇0Ω||𝑢𝜂||𝑙1+𝑐2||𝐷𝑢𝜂||𝜎𝜖𝑑𝑥𝑑𝑡2𝐿𝛼11𝑇0Ω𝛽𝜂𝑢𝜂||𝜕𝑡𝑢𝜂||2||𝐷𝑢𝜂||𝜎1𝑑𝑥𝑑𝑡+2𝜖𝑇0Ω||𝑢𝜂||𝑙1+𝑐2||𝐷𝑢𝜂||𝜎𝑑𝑥𝑑𝑡.(2.52)𝛽(𝑢0)𝐿(Ω)
Let us estimate the terms of (2.48) separately.
Integrating by parts the second integral of (2.48), we get ||𝑢𝜂||𝐿(𝑄𝑇)𝑐.
Arguing similarly, we obtain 𝜖
A series simple calculations give us the estimate of the third term of (2.48): Ω||𝐷𝑢𝜂||𝑝+𝜎𝑑𝑥𝑐1𝑇0Ω||𝐷𝑢𝜂||𝑝+𝜎𝑑𝑥𝑑𝑡+𝑐2𝑇+𝑐3Ω||𝐷𝑢0||𝑝+𝜎𝑑𝑥.(2.53)
Further, taking into account that Ω||𝐷𝑢𝜂||𝑝+𝜎𝑐𝑑𝑥2𝑇+𝑐3Ω||𝐷𝑢0||𝑝+𝜎𝑒𝑑𝑥𝑐1𝑇,(2.54) is uniformly bounded by 𝑐𝑖 (𝜎, where 1||Ω||Ω||𝐷𝑢𝜂||𝑝+𝜎𝑑𝑥1/(𝑝+𝜎)𝑐3||Ω||Ω||𝐷𝑢0||𝑝+𝜎𝑐𝑑𝑥+2||Ω||𝑇1/(𝑝+𝜎)𝑒𝑐1𝑇/(𝑝+𝜎)𝑒𝑐1𝑇/(𝑝+𝜎)𝑐3||Ω||Ω||𝐷𝑢0||𝑝+𝜎𝑑𝑥1/(𝑝+𝜎)+𝑒𝑐1𝑇/(𝑝+𝜎)𝑐2||Ω||𝑇1/(𝑝+𝜎).(2.55) is the bound in the proof of Lemma 2.1), it is possible to choose 𝜎 so that +, where 𝑢𝜖(𝑥,𝑡) is the Lipschitz constant of 𝜖0 on 𝑢=𝑢0(𝑥,𝑡) Therefore, 𝜕𝑡||||𝛽(𝑢)𝐷𝐷𝑢𝑝2𝐷𝑢+𝜑(𝑢)𝐷𝑢+𝑔0(𝑥,𝑢)=𝑓0(𝑥),(3.1)
Also note that if 𝑢𝜖(𝑥,0)=𝑢0(𝑥,0)=𝑢0(𝑥), then using the same arguments as in proof of (2.8) we conclude that 𝑔0 Thus, combining (2.49)–(2.52) and choosing 𝑓0 sufficiently small, we rewrite (2.48) in the form 𝑔(𝑥,𝑧,𝜁)
Applying Gronwall 's lemma, we get 𝑔(𝑥,𝑧,𝜁)=𝑀𝑗=1𝑏𝑗(𝑥,𝑧)𝑔𝑗(𝜁),(3.2) where the constant 𝑔𝑗𝐶1 does not depend on (𝐻3).. Consequently, 𝑗=1,
By letting 𝑏𝑗(𝑥,𝑧) tend to ||𝑏𝑗(||𝑥,𝑧)𝐶,(3.3) we get necessary estimation. Theorem is proved.

3. Quantitative Homogenization

As we mentioned earlier, our goal is to compare the global behavior of solutions 𝑏0𝑗(𝑥) of (1.4) for 𝑏𝑗(𝑥,𝑥/𝜖) with solutions 𝐿𝑤(Ω) of the homogenized equation𝜖0 where (3.1) and (1.4) are supplied with the same initial data 𝑏𝑗𝑥𝑥,𝜖,𝑣(𝑥)𝜖0𝑏0𝑗,(𝑥),𝑣(𝑥)(3.4) and homogenized nonlinearity 𝑣(𝑥)𝐿1(Ω), and inhomogeneity , are defined according to assumption from [11, pages 172-174].

We suppose that the function 𝑓𝑥𝑥,𝜖,𝑣(𝑥)𝜖0𝑓0,(𝑥),𝑣(𝑥)(3.5) has the following structure:𝑣(𝑥)𝐿2(Ω) where 𝜕𝑡𝛽(𝑢0)𝐷((|𝐷𝑢0|𝑝2𝐷𝑢0+𝜑(𝑢0)𝐷𝑢0)+𝑔0(𝑥,𝑢0)=𝑓0(𝑥) is supposed to satisfy a condition 𝑔0𝑥,𝑢0=𝑀𝑗=1𝑏0𝑗(𝑥)𝑔𝑗𝑢0.(3.6)

For all ̃𝑏𝑗(𝑥,𝑧)=𝑏𝑗(𝑥,𝑧)𝑏0𝑗(𝑥) we suppose that 𝑥Ω,𝑧𝑅 are bounded:𝐵𝑗(𝑥,𝑧) and the average 𝑥Ω, of 𝑧𝑅 exist in ||𝐵𝑗||(𝑥,𝑧)𝐶(3.7), for ̃𝑏𝑗̃𝑏𝑗(𝑥,𝑧)=𝜕𝑧𝐵𝑗(𝑥,𝑧). for 𝑥- where 𝜖- indicates duality.

We also suppose that𝐿1- for any ||||||𝜕𝑥𝐵𝑗,𝜖||||||𝐿1(Ω)𝐶,(3.8).

The equation 𝑗=1,,𝑀 is called the homogenization of equation if𝜕𝑥

Denoting 𝑥, for 𝐵𝑗(𝑥,𝑧)., we assume that there exist functions 𝑓(𝑥,𝑧)=𝑓(𝑥,𝑧)𝑓0(𝑥) which are uniformly bounded for all 𝐽(𝑥,𝑧)𝑓(𝑥,𝑧) given as𝑓(𝑥,𝑧)=𝜕𝑧𝐽(𝑥,𝑧). and which represent ||||||||||𝜕𝐽(𝑥,𝑧)𝐶,𝑥𝐽,𝜖||||||𝐿1(Ω)𝐶.(3.9) such that ̃𝑏𝑗(𝑥,𝑧) With respect to the 𝑓(𝑥,𝑧)derivatives we assume 𝐵𝑗(𝑥,𝑧)independent 𝐽(𝑥,𝑧)bound3.2, uniformly for all 𝑔(𝑥,𝑧,𝑤). Here 𝑓(𝑥,𝑧) is a partial’s derivatives with respect to the first argument 𝜌 of the function 𝑢𝜖(𝑥,𝑡)

Analogously, we denote 𝑢0(𝑢,𝑥) and require the existence of a function 𝑢0(𝑥)𝐿2(Ω) such that |||𝛽(𝑢𝜖)𝛽(𝑢0||)|𝐿1(Ω)𝐶𝜖𝑒𝜌𝑡uniformlyfor0𝑡<.(3.10) admits a divergence representation 𝑢𝜖(𝑥,𝑡0)𝑢0(𝑥,𝑡0),

Besides, we assume bounds𝑡0

Note that sufficient conditions which guarantee the existence of divergence representations for (0,𝑇). and [𝑎,𝑏] by help of 𝑢𝜖(𝑥,𝑡0)𝑢0(𝑥,𝑡0) and (𝑥1,𝑥2), respectively, are established in [11, Theorem 𝑢𝜖(𝑥,𝑡0)𝑢0(𝑥,𝑡0)>0 pages 176-180].

The following theorem holds.

Theorem 3.1. Let 𝑢𝜖(𝑥𝑖,𝑡0)=𝑢0(𝑥𝑖,𝑡0) and (𝑖=1,2). satisfy conditions (3.2)–(3.9) and let assumptions of Lemma 2.5 be fulfilled. Then there exists a positive constant such that the solutions 𝑑𝑑𝑡𝑥2𝑥1𝛽(𝑢𝜖𝑢)𝛽0𝑑𝑥𝑥2𝑥1𝐷||𝐷𝑢𝜖||𝑝2𝐷𝑢𝜖||𝐷𝑢0||𝑝2𝐷𝑢0𝑑𝑥𝑥2𝑥1𝐷𝜑(𝑢𝜖)𝐷𝑢𝜖𝑢𝜑0𝐷𝑢0𝑑𝑥+𝑥2𝑥1𝑔0(𝑥,𝑢𝜖)𝑔0𝑥,𝑢0𝑑𝑥=𝑥2𝑥1𝑔𝑥𝑥,𝜖,𝑢𝜖𝑔0(𝑥,𝑢𝜖)𝑑𝑥+𝑥2𝑥1𝑓𝑥𝑥,𝜖𝑓0(𝑥)𝑑𝑥.(3.11) and 𝑑𝑑𝑡𝑥2𝑥1𝛽(𝑢𝜖𝑢)𝛽0||𝑑𝑥𝐷𝑢𝜖||𝑝2𝐷𝑢𝜖𝑥2+||,𝑡𝐷𝑢0||𝑝2𝐷𝑢0𝑥2+||,𝑡𝐷𝑢𝜖||𝑝2𝐷𝑢𝜖𝑥1||,𝑡𝐷𝑢0||𝑝2𝐷𝑢0𝑥1,𝑡+𝜑(𝑢𝜖)𝐷𝑢𝜖𝑥2𝑢,𝑡𝜑0𝐷𝑢0𝑥2,𝑡𝜑(𝑢𝜖)𝐷𝑢𝜖𝑥1𝑢,𝑡+𝜑0𝐷𝑢0𝑥1+,𝑡𝑥2𝑥1𝑔0(𝑥,𝑢𝜖)𝑔0𝑥,𝑢0+𝑑𝑥𝑥2𝑥1𝑔𝑥𝑥,𝜖,𝑢𝜖𝑔0(𝑥,𝑢𝜖)𝑑𝑥𝑥2𝑥1𝑓𝑥𝑥,𝜖𝑓0(𝑥)𝑑𝑥𝑡=𝑡0=0.(3.12) of the respective problems (1.4), (1.8), (1.9) and (1.3), (1.8), (1.9) with equal initial data 𝐷𝑢𝜖(𝑥1,𝑡0)𝐷𝑢0(𝑥1,𝑡0), satisfy the quantitative homogenization estimation 𝐷𝑢𝜖(𝑥2,𝑡0)𝐷𝑢0(𝑥2,𝑡0),

Proof. Consider the “difference” 𝑢𝜖(𝑥𝑖,𝑡)=𝑢0(𝑥𝑖,𝑡)(𝑖=1,2), where 𝑑𝑑𝑡𝑥2𝑥1𝛽(𝑢𝜖𝑢)𝛽0||||𝑑𝑥𝑡=𝑡0<𝑥2𝑥1𝑔0(𝑥,𝑢𝜖)𝑔0𝑥,𝑢0𝑑𝑥𝑥2𝑥1𝑔𝑥𝑥,𝜖,𝑢𝜖𝑔0(𝑥,𝑢𝜖)𝑑𝑥+𝑥2𝑥1𝑓𝑥𝑥,𝜖𝑓0||||(𝑥)𝑑𝑥𝑡=𝑡0.(3.13) is an arbitrary point from 𝑥2𝑥1𝑔𝑥𝑥,𝜖,𝑢𝜖𝑔0(𝑥,𝑢𝜖)||||𝑑𝑥𝑡=𝑡0=𝑀𝑗=1𝑥2𝑥1𝑏𝑗𝑥𝑥,𝜖𝑔𝑗(𝑢𝜖)𝑏0𝑗(𝑥)𝑔𝑗(𝑢𝜖)|||||𝑑𝑥𝑡=𝑡0=𝑀𝑗=1𝑥2𝑥1̃𝑏𝑗𝑥𝑥,𝜖𝑔𝑗(𝑢𝜖|||||)𝑑𝑥𝑡=𝑡0.(3.14) The above “difference” is a continuously differentiable function, in virtue of Lemma 2.3. Hence, the interval 𝜕𝑧𝐵𝑗𝑥𝑥,𝜖𝑑=𝜖𝐵𝑑𝑥𝑗𝑥𝑥,𝜖𝜖𝜕𝑥𝐵𝑗𝑥𝑥,𝜖,(3.15) may be divided into the subintervals where sign of the “difference” 𝜕𝑥 does not change. Let 𝑥 be an interval such that 𝐵𝑗(𝑥,𝑧). and 𝑥2𝑥1𝑔𝑥𝑥,𝜖,𝑢𝜖𝑔0(𝑥,𝑢𝜖)||||𝑑𝑥𝑡=𝑡0=𝑀𝑗=1𝑥2𝑥1𝜕𝑧𝐵𝑗𝑥𝑥,𝜖𝑔𝑗(𝑢𝜖)|||||𝑑𝑥𝑡=𝑡0=𝜖𝑀𝑗=1𝑥2𝑥1𝑑𝐵𝑑𝑥𝑗𝑥𝑥,𝜖𝑔𝑗(𝑢𝜖)𝑑𝑥+𝜖𝑀𝑗=1𝑥2𝑥1𝜕𝑥𝐵𝑗𝑥𝑥,𝜖𝑔𝑗(𝑢𝜖|||||)𝑑𝑥𝑡=𝑡0.(3.16)(𝐻3) Then from equations (1.3) and (1.4) we obtain that 𝑥2𝑥1𝑔𝑥𝑥,𝜖,𝑢𝜖𝑔0(𝑥,𝑢𝜖)||||𝑑𝑥𝑡=𝑡0𝜖𝑀𝑗=1𝑥2𝑥1|||𝐵𝑗𝑥𝑥,𝜖|||||𝜕𝑥𝑔𝑗(𝑢𝜖)||𝑑𝑥+𝜖𝑀𝑗=1||||||𝜕𝑥𝐵𝑗..,𝜖||||||𝐿1(𝑥1,𝑥2)||||𝑔𝑗(𝑢𝜖)||||𝐿(𝑥1,𝑥2)|||||𝑡=𝑡0𝐶𝜖𝑀𝑗=1𝑥2𝑥1||𝜕𝑥𝑔𝑗(𝑢𝜖)||𝑑𝑥+𝜖𝑀𝑗=1𝜕𝑥𝐵𝑗,𝜖𝐿1(𝑥1,𝑥2)𝑔𝑗(𝑢𝜖)𝐿(𝑥1,𝑥2)|||𝑡=𝑡0.(3.17)𝑢𝜖𝜂
By applying Newton-Leibniz formula we have
Since𝐿(𝜏,𝑇;𝑊01,𝑝(Ω)) and 𝜖, it follows that 𝑐𝑖
Note that 𝜆
Obviously, (𝐻3) where, as we mentioned above, 𝑔𝑗(𝜔) indicate partial derivatives with respect to the first argument (𝐻5) of the function |𝑓(𝑥,𝑥/𝜖)|𝐶
Consequently, ||𝑢𝜖||𝐿(𝜏,𝑇;𝑊01,𝑝(Ω))𝐶.
Therefore, in view of condition 𝑀𝑗=1𝑥2𝑥1|𝜕𝑥𝑔𝑗(𝑢𝜖)|𝑑𝑥𝐶 and (3.7) 𝑥2𝑥1𝑔𝑥𝑥,𝜖,𝑢𝜖𝑔0(𝑥,𝑢𝜖)||||𝑑𝑥𝑡=𝑡0𝐶𝜖+𝐶𝜖𝑀𝑗=1𝜕𝑥𝐵𝑗,𝜖𝐿1(𝑥1,𝑥2)|||||𝑡=𝑡0.(3.18)
Observe now that the approximate solution 𝑥2𝑥1𝑓𝑥𝑥,𝜖𝑓0(𝑥)𝑑𝑥𝐶𝜖.(3.19) is bounded 𝜁𝑔0(𝑥,𝜁)+𝑐𝛽(𝜁)in the space 𝑑𝑑𝑡𝑥2𝑥1𝛽(𝑢𝜖𝑢)𝛽0||||𝑑𝑥𝑡=𝑡0<𝐶𝑥2𝑥1||𝛽(𝑢𝜖𝑢)𝛽0||||||𝑑𝑥𝑡=𝑡0+𝐶𝜖+𝐶𝜖𝑀𝑗=1||||||𝜕𝑥𝐵𝑗,𝜖||||||𝐿1(𝑥1,𝑥2)|||||𝑡=𝑡0.(3.20)𝜓(𝑡)=𝑥2𝑥1((𝛽(𝑢𝜖)𝛽(𝑢0))𝑑𝑥 uniformly with respect to 𝜓(𝑡)=𝑥2𝑥1|𝛽(𝑢𝜖)𝛽(𝑢0)|𝑑𝑥 because in the proof of this Lemma 2.3 we use the same constant 𝛽(𝑢𝜖)𝑡𝐿2(𝜏,𝑇,𝐿2(Ω) and 𝜓(𝑡)𝜓(𝑡) (condition 𝜓(𝑡0)=𝜓(𝑡0)) to estimate every 𝑑𝑑𝑡𝑥2𝑥1||𝛽(𝑢𝜖𝑢)𝛽0||||||𝑑𝑥𝑡=𝑡0𝑑𝑑𝑡𝑥2𝑥1𝛽(𝑢𝜖𝑢)𝛽0||||𝑑𝑥𝑡=𝑡0𝐶𝑥2𝑥1||𝛽(𝑢𝜖𝑢)𝛽0||||||𝑑𝑥𝑡=𝑡0+𝐶𝜖+𝐶𝜖𝑀𝑗=1𝜕𝑥𝐵𝑗,𝜖𝐿1(𝑥1,𝑥2)|||||𝑡=𝑡0.(3.21) from (3.2), and from condition (3.5) and 𝑑𝑑𝑡𝑥2𝑥1||𝛽(𝑢𝜖𝑢)𝛽0||||||𝑑𝑥𝑡=𝑡0𝐶𝑥2𝑥1||𝛽(𝑢𝜖𝑢)𝛽0||||||𝑑𝑥𝑡=𝑡0+𝐶𝜖.(3.22) (using known theorem on boundedness of the weakly convergence sequence in normed space) we can conclude that 𝑢𝜖(𝑥,𝑡)𝑢0(𝑢,𝑥). Thus, 𝑑𝑑𝑡𝑏𝑎||𝛽(𝑢𝜖𝑢)𝛽0||||||𝑑𝑥𝑡=𝑡0<𝐶𝑏𝑎||𝛽(𝑢𝜖𝑢)𝛽0||||||𝑑𝑥𝑡=𝑡0+𝐶𝜖.(3.23) Consequently, 𝑏𝑎||𝛽(𝑢𝜖𝑢)𝛽0||||||𝑑𝑥𝑡=𝑡𝐶𝜖𝑒𝜌(𝑡𝑡)𝑡>𝑡.>𝜏(3.24) and (3.17) becomes 𝛽(𝑢𝜖)𝐶(0,𝑇;𝐿2(Ω)),
Analogously, by (3.9), using the same arguments as in proof of (3.18), we deduce that 𝑏𝑎|𝛽(𝑢𝜖)𝛽(𝑢0||||)|𝑑𝑥𝑡𝐶𝜖𝑒𝜌𝑡.(3.25)
Now, bearing in mind that the function 𝜁𝑔(𝑥,𝜁)+𝑐𝛽(𝜁) is increasing and combining (3.18) and (3.19), we rewrite (3.13) in the form (𝐻2)
Further, note that the functions (𝐻4) and 𝑀𝑗=1𝑏0𝑗(𝑥)<̃𝑐<0 are absolutely continuous ((𝐻2)). Besides, it is obvious that 𝜈>0 and dist𝐿1(Ω)(𝑆0𝑡𝑢0;𝐴0)𝑐𝑒𝜈𝑡.. Consequently, 𝐴0
Hence, by condition (3.8) 𝑣
The same estimation holds for an arbitrary interval on which 𝑢=𝑆0𝑡𝑢0, does not change its sign. Summing up similar inequalities over subintervals, we get 𝑑𝑑𝑡𝑥2𝑥1𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝑥2𝑥1𝐷||||𝐷𝑢𝑝2||||𝐷𝑢𝐷𝑣𝑝2𝐷𝑣𝑑𝑥𝑥2𝑥1+𝐷(𝜑(𝑢)𝐷𝑢𝜑(𝑣)𝐷𝑣)𝑑𝑥𝑥2𝑥1𝑔0(𝑥,𝑢)𝑔0(𝑥,𝑣)𝑑𝑥=0,(3.26)
Consequently, 𝑥1
Thus, taking into account that 𝑥2𝛽(𝑢)𝛽(𝑣)0
Thereby, assertion follows.

Remark 3.2. It is easy to see that the condition on the function (𝑥1,𝑥2). in Lemma 2.5, which we use in Theorem 3.1, can be changed to the 𝑑𝑑𝑡𝑥2𝑥1𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝑥2𝑥1𝐷||||𝐷𝑢𝑝2||||𝐷𝑢𝐷𝑣𝑝2𝐷𝑣𝑑𝑥𝑥2𝑥1𝐷(𝜑(𝑢)𝐷𝑢𝜑(𝑣)𝐷𝑣)𝑑𝑥+𝑀𝑗=1𝑏0𝑗(𝑥)𝑥2𝑥1𝑔𝑗(𝑢)𝑔𝑗(𝑣)𝑑𝑥=0.(3.27). In this case we can exclude condition 𝑔𝑗(𝑤).

Now, note that if (𝐻3) and condition 𝜕𝜁𝑔𝑗>𝜆 is fulfilled then we derive, using simple reasoning, the following exponential attraction with exponential rate 𝑑𝑑𝑡𝑥2𝑥1𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝜆̃𝑐𝑥2𝑥1(𝑢𝑣)𝑑𝑥.(3.28): (𝐻2),

Indeed, we know that solution of the corresponding stationary problem belongs to the attractor, that is, it belongs to 𝑑𝑑𝑡𝑥2𝑥1𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝜆̃𝑐𝑥2𝑥1(𝛽(𝑢)𝛽(𝑣))𝑑𝑥.(3.29). Denoting this solution by 𝑑𝑑𝑡𝑥2𝑥1||||𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝜆̃𝑐𝑥2𝑥1||||𝛽(𝑢)𝛽(𝑣)𝑑𝑥.(3.30), we will use the same arguments as in proof of Lemma 2.5.

Defining 𝑑𝑑𝑡𝑏𝑎||||𝛽(𝑢)𝛽(𝑣)𝑑𝑥<𝜆̃𝑐𝑏𝑎||||𝛽(𝑢)𝛽(𝑣)𝑑𝑥,(3.31) we have𝑏𝑎||||𝑢𝛽(𝑢)𝛽(𝑣)𝑑𝑥<𝑐0𝑒,𝑣𝜆̃𝑐𝑡.(3.32) where (𝐻2) and 𝑥2𝑥1|𝑢𝑣|𝑑𝑥𝐿1𝑥2𝑥1||||𝛽(𝑢)𝛽(𝑣)𝑑𝑥𝐿1𝑐𝑢0𝑒,𝑣𝜆̃𝑐𝑡=𝑐𝑒𝜈𝑡.(3.33) are such that dist𝐿1(Ω)(𝑆0𝑡𝑢0;𝐴0)𝑐𝑒𝜈𝑡 on the interval 𝜈=𝜆̃𝑐, Hence,𝐿1 Bearing into mind that 4.1 is supposed to satisfy condition 𝐴𝜖 (𝐴0), we have𝜖dist𝐿1(Ω)(𝐴𝜖,𝐴0) Using 𝐶𝜖𝛾 we derive𝑔(𝑥,𝑧,𝑤) Hence, from(2.42)𝑓(𝑥,𝑧) Further, arguing similarly to the proof of Lemma 2.5, we arrive to(𝐻1)-(𝐻3),(𝐻2), or(𝐻5)

Consequently, by condition 𝑀𝑗=1𝑏0𝑗(𝑥)<̃𝑐<0𝐴𝜖

Thus, we obtain that dist𝐿1(Ω)(𝐴𝜖,𝐴0), where 𝐶𝜖𝛾. is the Lipschitz constant.

Hence, using Remarks 2.6 and 3.2, with the help of Lemma of the study in [11], we obtain the estimate for the distance between the nonhomogenized and the homogenized attractors in terms of the parameter . So, the following theorem holds.

Theorem 3.3. Let and satisfy conditions (3.2)–(3.9), and assumptions and are fulfilled. Also suppose that . Then the global attractors of the problem (1.4), (1.8), (1.9) satisfy an upper semicontinuity distance estimate of the form