We study a class of fractional stochastic integrodifferential equations
considered in a real Hilbert space. The existence and uniqueness of the Mild solutions
of the considered problem is also studied. We also give an application for stochastic
integropartial differential equations of fractional order.
1. Introduction
Let H and K denote real Hilbert spaces equipped with norms ∥·∥H and ∥·∥K, respectively, and let the space of bounded linear operators from K to H be denoted by BL(K;H). For Banach space X and Y, the space of continuous functions from X into Y (equipped with the usual sup-norm) will be denoted by C(X;Y), while Lp(0,T;X) will represent the space of X-valued functions that are p-integrable on [0,T]. Let (Ω,Z,P) be a complete probability space equipped with a normal filtration {Zt:0≤t≤T}. An H-valued random variable is an Z-measurable function X:Ω→H, and a collection of random variables ψ={X(t;ω):Ω→H:0≤t≤T} is called a stochastic process. The collection of all strongly measurable square integrable H-valued random variables, denoted by L2(Ω;H), is a Banach space equipped with norm ∥X(·)∥L2(Ω;H)=(E∥X(·;ω)∥H2)1/2.
An important subspace is given by L02(Ω;H)={f∈L2(Ω;H):fisZ0measurable}. Next we define the space γ((0,T);H) to be the set {v∈C([0,T];L2(Ω;H):visZt-adapted} with norm
∥v∥γ=sup0≤t≤T(E∥v(t)∥H2)1/2
(see in [1–5]). In this paper we study the existence and uniqueness of the mild solution of the fractional stochastic integrodifferential equation of the form
dαx(t)dtα=Ax(t)+F(x)(t)+∫0tG(x)(s)dW(s),0≤t≤T,x(0)=h(x)+x0,
in a real separable Hilbert space H. Here, 1/2≤α≤1,A:D(A)⊂H→H is a linear closed operator generating semigroup, F:γ([0,T];H)→Lp([0,T];L2(Ω;H))(1≤p<∞),G:γ([0,T];H)→C([0,T];L2(Ω;BL(K;H))) (where K is a real separable Hilbert space), W is a K-valued Wiener process with incremental covariance described by the nuclear operator Q, x0 is an Z0-measurable H-valued random variable independent of W and h:γ([0,T];H)→L02(Ω;H).
Definition 1.1.
An Zt-adapted stochastic process x:[0,T]→H is called a mild solution of (1.2) if x(t) is measurable, for all t∈[0,T],∫0T∥x(s)∥H2ds<∞,x(t)=∫0∞ξα(θ)S(tαθ)(h(x)+x0)dθ+α∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)F(x)(η)dθdη+α∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)[∫0ηG(x)(τ)dW(τ)]dθdη,0≤t≤T,
where ξα(θ) is a probability density function defined on (0,∞),
∫0∞ξα(θ)dθ=1
(see [6–12]). In the next section, we will prove the existence and uniqueness of the mild solutions to (1.2).
2. Existence and Uniqueness
Consider the initial value problem (1.2) in a real separable Hilbert space H under the following assumptions:
the linear operator A:D(A)⊂H→H generates a C0-semigroup{S(t):t≥0} on H;
F:γ([0,T];H)→Lp(0,T;L2(Ω;H)) is such that there exists MF>0 for which
∥F(x)-F(y)∥Lp≤MF∥x-y∥γ,∀x,y∈γ([0,T];H);
G:γ([0,T];H)→C([0,T];L2(Ω;BL(K;H)))(=γBL) is such that there exists MG>0 for which
∥G(x)-G(y)∥γBL≤MG∥x-y∥γ∀x,y∈γ([0,T];H);
h:γ([0,T];H)→L02(Ω;H) is such that there exists Mh>0 for which
∥h(x)-h(y)∥L02≤Mh∥x-y∥γ∀x,y∈γ([0,T];H);
x0∈L02(Ω;H).
We can therefore state the following theorem.
Theorem 2.1.
Assume that (I)–(V) hold. Then (1.2) has a unique solution on [0,T], provided that
MS[Mh+CFTα+MGCGTα+1/2]<1,
where Mh>0, MS>0, and CG>0.
Proof.
Define the solution map J:γ([0,T];H)→γ([0,T];H) by
(Jx)(t)=∫0∞ξα(θ)S(tαθ)(h(x)+x0)dθ+α∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)F(x)(η)dθdη+α∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)×[∫0ηG(x)(τ)dW(τ)]dθdτ,0≤t≤T.
From Holder's inequality, we get
[E∥∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)F(x)(η)dθdη∥H2]1/2≤MS[∫0T∥(T-η)α-1F(x)(η)∥L2(Ω;H)2dη]1/2≤MS[∫0T(T-η)2(α-1)dη]1/2[∫0T∥F(x)(η)∥L2(Ω;H)2dη]1/2≤MSTα-1/2(2α-1)1/2[∫0T∥F(x)(η)∥L2(Ω;H)2dη]1/2≤CFMSTα-1/2∥F(x)∥Lp,
where CF is a constant depending on α.
Subsequently, an application of (II), together with Minkowski's inequality enables us to continue the string of inequalities in (2.6) to conclude that
[E∥∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)F(x)(η)dθdη∥H2]1/2≤MSCFTα-1/2[MF∥x∥γ+∥F(0)∥Lp].
Taking the supermum over [0,T] in (2.7) then implies that
∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)F(x)(η)dθdη∈γ([0,T];H),
for any x∈γ([0,T];H). Furthermore for such x,G(x)(η)∈BL(K;H), and h(x)+x0∈L02(Ω;H) (by (IV) and (V). Consequently, one can argue as in [13–15] to conclude that J is well defined.
Next we show that J is a strict contraction.
Observe that for x,y∈γ([0,T];H), we infer from (2.5) that
(Jx)(t)-(Jy)(t)=∫0∞ξα(θ)S(tαθ)(h(x)-h(y))dθ+α∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)(F(x)(η)-F(y)(η))dθdη+α∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)[∫0η(G(x)(τ)-G(y(τ)))dW(τ)]dθdη,0≤t≤T.
Squaring both sides and taking the expectation in (2.9) yields, with the help of Young's inequality,
E∥(Jx)(t)-(Jy)(t)∥H2≤4E∥∫0∞ξα(θ)S(tαθ)(h(x)+x0)dθ∥H2+4α2[E∥∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)(F(x)(η)-F(y)(η))dθdη∥H2+E∥∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)×[∫0η(G(x)(τ)-G(y)(τ))dW(τ)]dθdη∥H2],
and subsequently,
∥(Jx)(t)-(Jy)(t)∥γ≤∥∫0∞ξα(θ)S(tαθ)(h(x)-h(y))dθ∥γ+4α2[∥∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)(F(x)(η)-F(y)(η))dθdη∥γ+∥∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)×[∫0η(G(x)(τ)-G(y)(τ))dW(τ)]dθdη∥γ].
Using reasoning similar to that which led to (2.6), one can show that
∥∫0∞ξα(θ)S(tαθ)(h(x)-h(y))dθ∥γ=E∥∫0∞ξα(θ)S(tαθ)(h(x)+x0)dθ∥H2≤MsMh∥x-y∥γ,∥∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)(F(x)(η)-F(y)(η))dθdη∥γ=∥∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)(F(x)(η)-F(y)(η))dθdη∥γ≤CFMSTα∥x-y∥γ,
where CF depending on α and MF. We also infer that
∥∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)[∫0η(G(x)(τ)-G(y)(τ))dW(τ)]dθdη∥γ=[E∥∫0t∫0∞θ(t-η)α-1ξα(θ)S((t-η)αθ)[∫0η(G(x)(τ)-G(y)(τ))dW(τ)]dθdη∥H2]1/2≤Tr(Q)Tα-1/2(2α-1)1/2MS[∫0T∫0T∥G(x)(τ)-G(y)(τ)∥L2(Ω;H)2dτdη]1/2≤CGTα+1/2MsMG∥x-y∥γ,
where CG is a constant depending on (α and Tr(Q). Using (2.12) and (2.13) in (2.11) enables us to conclude that J is a strict contraction, provided that (2.4) is satisfied, and has a unique fixed point which coincides with a mild solution of (1.2). This completes the proof.
3. Application
Let D be a bounded domain in RN with smooth boundary ∂D, and consider the initial boundary value problem:
∂α(t,z)∂tα=Δzx(t,z)+∫0Ta(t,s)f1(s,x(s,z),∫0sk(s,τ,x(τ,z))dτ)ds+∫0Tb(t,s)f2(s,x(s,z))dW(s),on(0,T)×D,x(0,z)=∑i=1ngi(z)x(ti,z)+∫0Tc(s)f3(s,x(s,z))ds,onD,x(t,z)=0,on(0,T)×∂D,
where 0≤t1<t2⋯<tn≤T are given and W is an L2(D)-valued Wiener process. We consider the equation (3.1) under the following conditions.
f1:[0,T]×R×R→R satisfies the Caratheodory conditions as well as
f1(·,0,0)∈L2(0,T),
|f1(t,x1,y1)-f1(t,x2,y2)|≤Mf1[|x1-x2|+|y1-y2|], for all x1,x2,y1,y2∈R and almost t∈(0,T) for some Mf1>0,
f2:[0,T]×R→BL(L2(D)) where BL(L2(D)) is the space of bounded linear operator from L2(D) to L2(D) satisfies the Caratheodory conditions as well as
f2(·,0)∈L2(0,T),
|f2(t,x)-f2(t,y)|BL(H)≤Mf2|x-y|, for all x,y∈R and almost all t∈(0,T), for some Mf2>0.
f3:[0,T]×R→R satisfies the Caratheodory conditions as well as
f3(·,0)∈L2(0,T),
|f3(t,x)-f3(t,y)|≤Mf3|x-y|, for all x,y∈R and almost t∈(0,T) for some Mf3>0,
a∈L2((0,T)2),
b∈L∞((0,T)2),
c∈L2((0,T)2),
k:Y×R→R, where Y={(t,s):0<s<t<T}, satisfies |k(t,s,x1)-k(t,s,x2)|≤Mk|x1-x2|, for all x1,x2∈R, and almost (t,s)∈Y,
gi∈L2(D),i=1,…,n.
The stochastic integropartial differential equation (3.1) can be written in the abstract form (1.2), where K=H=L2(D), A=Δz, with domain D(A)=H2(D)∪H01((D)). It is well known that A is a closed linear operator which generates a C0-semigroup. We also introduce the mappings F, G, and h defined by, respectively,
F(x)(t,·)=∫0Ta(t,s)f1(s,x(s,z),∫0sk(s,τ,x(τ,z))dτ)ds,G(x)(t,·)=b(t,s)f2(s,x(s,·)),h(x)(·)=x(0,z)=∑i=1ngi(·)x(ti,·)+∫0Tc(s)f3(s,x(s,·))ds.
One can use (H1)–(H8) to verify that F, G, and h satisfy (II)–(IV) in the last section, respectively, with
MF=2Mf1T|a|L2((0,T)2)(1+MkT3)1/2,MG=Mf2,Mh=2∑i=1n∥gi∥L2(D)+Mf3m(D)|G|L2(0,T).
Consequently theorem (2.4) can be applied for (3.1).
KeckD. N.McKibbenM. A.Functional integro-differential stochastic evolution equations in Hilbert space2003162141161MR198958010.1155/S1048953303000108ZBL1031.60061El-BoraiM. M.El-Said El-NadiK.MostafaO. L.AhmedH. M.Semigroup and some fractional stochastic integral equations2006314752BahugunaD.Integro-differential equations with analytic semigroup200316217718910.1155/S1048953303000133BahugunaD.Quasi linear integro-differential equations in Banach spaces19952417518310.1016/0362-546X(94)E0049-MBahugunaD.PaniA. K.Strong solutions to nonlinear integro-differential equations199029–90Canberra, AustraliaAustralian National UniversityEl-BoraiM. M.Semigroups and some nonlinear fractional differential equations20041493823831MR203316510.1016/S0096-3003(03)00188-7ZBL1046.34079El-BoraiM. M.On some fractional differential equations in the Hilbert space2005233240MR2192679ZBL1160.47038El-BoraiM. M.Some probability densities and fundamental solutions of fractional evolution equations2002143433440MR190329510.1016/S0960-0779(01)00208-9ZBL1005.34051El-BoraiM. M.El-Said El-NadiK.MostafaO. L.AhmedH. M.Volterra equations with fractional stochastic integrals200420045453468MR212943510.1155/S1024123X04312020ZBL1081.45007El-BoraiM. M.The fundamental solutions for fractional evolution equations of parabolic type200420043197211MR210283210.1155/S1048953304311020ZBL1081.34053FellerW.1971New York, NY, USAJohn Wiley & Sonsxxiv+6692ndMR0270403GorenfloR.MainardiF.Fractional calculus and stable probability distributions1998503377388MR1648257ZBL0934.35008GovidanT. E.Autonomous semi linear stochastic Volterra itegro-differential equations in Hilbert spaces199435174GovidanT. E.1992New York, NY, USAMarcel-DekkerLecture Notes in Pure and Applied Mathematics, 142GreckschW.TudorC.199585Berlin, GermanyAkademie178Mathematical ResearchMR1353910