HAL Id: hal-02548520
https://hal.archives-ouvertes.fr/hal-02548520
Submitted on 20 Apr 2020
HAL
is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire
HAL, estdestinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Low-regret control for a fractional wave equation with incomplete data
Dumitru Baleanu, Claire Joseph, Gisèle Mophou
To cite this version:
Dumitru Baleanu, Claire Joseph, Gisèle Mophou. Low-regret control for a fractional wave equa-
tion with incomplete data. Advances in Difference Equations, Hindawi Publishin Corporation /
SpringerOpen, 2016, �10.1186/s13662-016-0970-8�. �hal-02548520�
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_
DOI 10.1186/s13662-016-0970-8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47
R E S E A R C H Open Access
Low-regret control for a fractional wave equation with incomplete data
Dumitru Baleanu1,2, Claire Joseph3and Gisèle Mophou3,4*
*Correspondence:
gmophou@univ-ag.fr
3Université des Antilles et de la Guyane, Campus Fouillole, 97159, Pointe-à-Pitre, Guadeloupe (FWI), France
4Laboratoire MAINEGE, Université Ouaga 3S, 06 BP 10347, Ouagadougou 06, Burkina Faso Full list of author information is available at the end of the article
Abstract
We investigate in this manuscript an optimal control problem for a fractional wave equation involving the fractional Riemann-Liouville derivative and with missing initial condition. For this purpose, we use the concept of no-regret and low-regret controls.
Assuming that the missing datum belongs to a certain space we show the existence and the uniqueness of the low-regret control. Besides, its convergence to the no-regret control is discussed together with the optimality system describing the no-regret control.
Keywords: Riemann-Liouville fractional derivative; Caputo fractional derivative;
optimal control; no-regret control; low-regret control
1 Introduction
Let us considerN∈N∗anda bounded open subset ofRNpossessing the boundary∂
of classC. When the timeT> , we considerQ=×],T[ and6=∂×],T[ and we discuss the fractional wave equation:
DαRLy(x,t) –1y(x,t) =v(x,t), (x,t)∈Q, y(σ,t) = , (σ,t)∈6,
I–αy(x, +) =y, x∈,
∂
∂tI–αy(x, +) =g, x∈,
()
such that / <α< ,y∈H()∩H(),I–αy(x, +) =limt→I–αy(x,t) and ∂t∂I–αy(x,
+) =limt→∂t∂I–αy(x,t) where the fractional integralIα of orderα and the fractional derivativeDαRL of orderαare within the Riemann-Liouville sense. The functiong is un- known and belongs toL() and the controlv∈L(Q).
Since the initial condition is unknown, the system () is a fractional wave equations with missing data. Such equations are used to model pollution phenomena. In this systemg represents the pollution term.
According to the data, we know that system () admits a unique solution y(v,g) = y(x,t;v,g) inL((,T);H())⊂L(Q) []. Hence, we can define the following functional:
J(v,g) =
y(v,g) –zd
L(Q)+NkvkL(Q), ()
wherezd∈L(Q) andN> .
©2016 Baleanu et al. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, pro- vided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 2 of 20
48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94
In this manuscript, we discuss the optimal control problem, namely inf
v∈L(Q)J(v,g), ∀g∈L(). ()
If the functiongis given, namelyg=g∈L(), then system () is completely determined and problem () becomes a classical optimal control problem []. Such a problem was studied by Mophou and Joseph [] with a cost function defined with a final observation.
Actually, the authors proved that one can approach the fractional integral of order <
–α< / of the state at final time by a desired state by acting on a distributed control.
For more literature on fractional optimal control, we refer to [–] and the references therein.
Since the functiongis unknown, the optimal control problem () has no sense because L() is of infinite dimension. So, to solve this problem, we proceed as Lions [, ] for the control of partial differential equations with integer time derivatives and missing data.
This means that we use the notions of no-regret and low-regret controls. There are many works using these concepts in the literature. In [] for instance, Nakoulimaet al.utilized these concepts to control distributed linear systems possessing missing data. A generaliza- tion of this approach can be found in [] for some nonlinear distributed systems possess- ing incomplete data. Jacob and Omrane used the notion of no-regret control to control a linear population dynamics equation with missing initial data []. Recently, Mophou [] used these notions to control a fractional diffusion equation with unknown bound- ary condition. For more literature on such control we refer to [–] and the references therein.
In our paper, we show that the low-regret control problem associated to () admits a unique solution which converges toward the no-regret control. We provide the singular optimality system for the no-regret control.
Below we present the organization of our manuscript. In the following section, we show briefly some results about fractional derivatives and preliminary results on the existence and uniqueness of solution to fractional wave equations. In Section , we investigate the no-regret and low-regret control problems corresponding to ().
2 Preliminaries
Below, we give briefly some results about fractional calculus and some existence results about fractional wave equations.
Definition .[, ] Iff :R+→Ris a continuous function onR+, andα> , then the expression of the Riemann-Liouville fractional integral of orderαis
Iαf(t) = Ŵ(α)
Z t
(t–s)α–f(s)ds, t> .
Definition .[, ] The form of the left Riemann-Liouville fractional derivative of order ≤n– <α<n,n∈Noff is given by
DαRLf(t) =
Ŵ(n–α)· dn dtn
Z t
(t–s)n––αf(s)ds, t> .
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 3 of 20
95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141
Definition .[, ] The left Caputo fractional derivative of order ≤n– <α<n, n∈Noff is given by
DαCf(t) = Ŵ(n–α)
Z t
(t–s)n––αf(n)(s)ds, t> . ()
We mention that in the above two definitions we considerf :R+→R.
Definition .[–] Letf :R+→R, ≤n– <α<n,n∈N. Then the right Caputo fractional derivative of orderαoff is
DαCf(t) = (–)n Ŵ(n–α)
Z T t
(s–t)n––αf(n)(s)ds, <t<T. ()
In all above definitions we assume that the integrals exist.
Lemma .[] Let y∈C∞(Q)andϕ∈C∞(Q).Then we have Z
Q
DαRLy(x,t) –1y(x,t)
ϕ(x,t)dx dt
= Z
ϕ(x,T)∂
∂tI–αy(x,T)dx– Z
ϕ(x, )∂
∂tI–αy x, + dx –
Z
I–αy(x,T)∂ϕ
∂t(x,T)dx+ Z
I–αy(x, )∂ϕ
∂t(x, )dx +
Z
6
y(σ,s)∂ϕ
∂ν(σ,s)dσdt– Z
6
∂y
∂ν(σ,s)ϕ(σ,s)dσdt +
Z
Q
y(x,t) DCαϕ(x,t) –1ϕ(x,t)
dx dt. ()
In the following we give some results that will be use to prove the existence of the low- regret and no-regret controls.
Theorem .[] Let/ <α< ,y∈H()∩H(),y∈L()and v∈L(Q).Then the problem
DαRLy(x,t) –1y(x,t) =v(x,t), (x,t)∈Q, y(σ,t) = , (σ,t)∈6,
I–αy(x, +) =y, x∈,
∂
∂tI–αy(x, +) =y, x∈
()
has a unique solution y∈L((,T);H()).Moreover,the following estimates hold:
kykL((,T);H())≤1 y
H()+ y
L()+kvkL(Q)
, () I–αy
C([,T];H())≤5 y
H()+ y
L()+kvkL(Q)
, ()
∂
∂tI–αy
C([,T];L())
≤2 y
H()+ y
L()+kvkL(Q)
, ()
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 4 of 20
142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
with
1=max
C s
Tα–
(α– ),C s
Tα–
(α– ),C s
Tα α(α– )
,
5=sup
C√
,C√
T–α,C
sT–α ( –α)
,
and
2=max √
CTα–,√
C .
Consider the fractional wave equation involving the left Caputo fractional derivative of order / <α< :
DαCy(x,t) –1y(x,t) =f, (x,t)∈Q, y(σ,t) = , (σ,t)∈6,
y(x, ) = , x∈,
∂y
∂t(x, ) = , x∈,
()
wheref ∈L(Q).
Theorem . Let f∈L(Q).Then problem()has a unique solution y∈C([,T];H()).
Moreover,∂y∂t ∈C([,T];L())and there exists C> in such a way that
kykC([,T];H())≤C rTα–
α– kfkL(Q) ()
and
∂y
∂t
C([,T];L())
≤C rTα–
α– kfkL(Q). ()
Proof Below we proceed as was mentioned in [].
Corollary . Let/ <α< andφ∈L(Q).Consider the fractional wave equation:
DCαψ(x,t) –1ψ(x,t) =φ, (x,t)∈Q, ψ(σ,t) = , (σ,t)∈6,
ψ(x,T) = , x∈,
∂ψ
∂t(x,T) = , x∈,
()
whereDCαis the right Caputo fractional derivative of order <α< .Then()has a unique solutionψ∈C([,T];H()).Moreover, ∂ψ∂t ∈C([,T];L()),and there exists C> ful- filling
kψkC([,T];H())≤C rTα–
α– kφkL(Q) ()
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 5 of 20
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235
and
∂ψ
∂t
C([,T];L())≤C rTα–
α– kφkL(Q). ()
Proof If we make the change of variablet→T–tin (), then we conclude thatψˆ(t) = ψ(T–t) verifies
DαCψˆ –1ψˆ =φˆ inQ, ˆ
ψ= on6, ˆ
ψ() = in,
∂ψˆ
∂t() = in,
()
whereφˆ(t) =φ(T–t) andDαCis the left Caputo fractional derivative of order / <α< .
BecauseT–t∈[,T] whent∈[,T], we say thatφˆ∈L(Q) due to the fact thatφ∈L(Q).
It then suffices to use Theorem . to conclude.
We also need some trace results.
Lemma .[] Let f ∈L(Q)and y∈L(Q)such that DαRLy–1y=f.Then:
(i) y|∂and∂ν∂y|∂exist and belong toH–((,T);H–/(∂))and H–((,T);H–/(∂))respectively.
(ii) I–αy∈C([,T];L()).
(iii) ∂t∂I–αy∈C([,T];H–()).
3 Existence and uniqueness of no-regret and low-regret controls
Below, we show the existence and the uniqueness of the no-regret control and the low- regret control problem for system ().
Lemma . Let v∈L(Q)and g∈L().Then we have J(v,g) =J(,g) +J(v, ) –J(, )
+ Z
Q
y(,g) –y(, )
y(v, ) –y(, )
dt dx. ()
Here J denotes the functional given by()and y(v,g) =y(x,t;v,g)∈L(,T;H())⊂L(Q) is the solution of ().
Proof Let us considery(v, ) =y(x,t;v, ),y(,g) =y(x,t; ,g), andy(, ) =y(x,t; , ) be the solutions of
DαRLy(v, ) –1y(v, ) =v inQ, y= on6,
I–αy(;v, ) =y in,
∂
∂tI–αy(;v, ) = in,
()
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 6 of 20
236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282
DαRLy(,g) –1y(,g) = inQ, y= on6,
I–αy(; ,g) =y in,
∂
∂tI–αy(; ,g) =g in,
()
and
DαRLy(, ) –1y(, ) = inQ, y= on6,
I–αy(; , ) =y in,
∂
∂tI–αy(; , ) = in,
()
whereI–αy(;v,g) =limt→+I–αy(x,t;v,g) and ∂t∂I–αy(;v,g) =limt→+ ∂
∂tI–αy(x,t;v,g).
Sincev∈L(Q),y∈H()∩H() andg∈L(), we see from Theorem . thaty(v, ), y(,g) andy(, ) belong toL((,T);H()).
Observing that J(v, ) =
Z
Q
y(v, ) –zd
y(v, ) –zd
dt dx+NkvkL(Q), (a) J(,g) =
Z
Q
y(,g) –zd
y(,g) –zd
dt dx, (b)
J(, ) = Z
Q
y(, ) –zd
y(, ) –zd
dt dx, (c)
and using the fact that
y(v,g) =y(v, ) +y(,g) –y(, ), we have
J(v,g) =
y(v,g) –zd
L(Q)+NkvkL(Q)
=
y(v, ) +y(,g) –y(, ) –zd
L(Q)+NkvkL(Q)
=J(v, ) + Z
Q
y(v, ) –zd
y(,g) –y(, ) dt dx +
y(,g) –y(, ) –zd
L(Q)
=J(v, ) + Z
Q
y(v, ) –y(, )
y(,g) –y(, ) dt dx +
Z
Q
y(, ) –zd
y(,g) –y(, )
dt dx+
y(,g) –y(, ) –zd
L(Q). Using
y(,g) –y(, ) –zd
L(Q)=J(,g) +J(, ) –
Z
Q
y(, ) –zd
y(,g) –y(, )
dt dx– J(, ),
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 7 of 20
283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329
we conclude that
J(v,g) =J(,g) +J(v, ) –J(, ) +
Z
Z T
y(v, ) –y(, )
y(,g) –y(, )
dt dx.
Lemma . Let v∈L(Q)and g∈L().Then we have J(v,g) =J(,g) +J(v, ) –J(, ) +
Z
gζ(x, ;v)dx, ()
whereζ(v) =ζ(x,t;v)∈C([,T];H())be solution of
DCαζ(v) –1ζ(v) =y(v, ) –y(, ) in Q, ζ= on6,
ζ(x,T;v) = in,
∂ζ
∂t(x,T;v) = in.
()
Proof Sincey(v, ) –y(, )∈L(Q), from Proposition ., we know that the system () admits a unique solutionζ(v)∈C([,T];H()). Also, there existsC> such that
ζ(v)
C([,T];H())≤C rTα–
α–
y(v, ) –y(, )
L(Q) () and
∂ζ
∂t(v) C
([,T];L())
≤C rTα–
α–
y(v, ) –y(, )
L(Q). () Setz=y(g, ) –y(, ). Thenzverifies
DαRLz–1z= inQ, z= on6,
I–αz() = in,
∂
∂tI–αz() =g in.
()
Sinceg∈L(), it follows from Theorem . thatz∈L((,T);H()),I–αz∈C([,T], H()), and ∂t∂I–αz∈C([,T],L()). So, if we multiply the first equation of () byz utilizing the fractional integration by parts provided by Lemma ., we conclude
Z
Q
y(v, ) –y(, ) z dt dx
= Z
Q
DαCζ(v) –1ζ(v) z dt dx
= Z
ζ(x, ;v)∂
∂tI–αz()dx. ()
Thus, replacingzby (y(,g) –y(, )), we obtain Z
Q
y(v, ) –y(, )
y(,g) –y(, ) dt dx=
Z
ζ(x, ;v)g dx,
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 8 of 20
330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376
and () becomes
J(v,g) –J(,g) =J(v, ) –J(, ) + Z
ζ(x, ;v)g dx.
Now we consider the no-regret control problem:
inf
v∈L(Q)
sup
g∈L()
J(v,g) –J(,g)
. ()
From (), this problem is equivalent to the following one:
inf
v∈L(Q)
sup
g∈L()
J(v, ) –J(, ) + Z
ζ(x, ;v)g dx
. ()
As the spaceL() is a vector space, the no-regret control exists only if sup
g∈L()
Z
ζ(x, ;v)g dx
= . ()
This implies that the no-regret control belongs toUdefined by U=
v∈L(Q)
Z
ζ(x, ;v)g dx
= ,∀g∈L()
.
As a result such control should be carefully investigated. So, we proceed by penalization.
For allγ > , we discuss the low-regret control problem:
inf
v∈L(Q) sup
g∈L()
J(v,g) –J(,g) –γkgkL()
. ()
According to (), the problem () is equivalent to the following problem:
inf
v∈L(Q)
J(v, ) –J(, ) + sup
g∈L()
Z
ζ(x, ;v)g dx–γ
kgkL()
.
Using the Legendre-Fenchel transform, we conclude
γ sup
g∈L()
Z
γζ(x, ;v)g dx– γ γ
kgkL()
= γ
ζ(·, ;v)
L(), and problem () becomes: For anyγ > , finduγ∈L(Q) such that
Jγ uγ
= inf
v∈L(Q)Jγ(v), ()
where
Jγ(v) =J(v, ) –J(, ) + γ
ζ(·, ;v)
L(). ()
Proposition . Letγ > .Then()has a unique solution uγ,called a low-regret control.
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 9 of 20
377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423
Proof We recall that
Jγ(v) =J(v, ) –J(, ) + γ
ζ(·, ;v)
L()≥–J(, ).
Thus, we can say thatinfv∈L(Q)Jγ exists. Let (vn)∈L(Q) be a minimizing sequence such that
n→+∞lim Jγ(vn) = inf
v∈L(Q)Jγ(v). ()
Thenyn=y(x,t;vn, ) is a solution of () andynsatisfies
DαRLyn(x,t) –1yn(x,t) =vn(x,t) inQ, (a)
yn(x,t) = on6, (b)
I–αyn(x, ) =y in, (c)
∂
∂tI–αyn(x, ) = in. (d)
It follows from () that there existsC(γ) > independent ofnsuch that
≤J(vn, ) + γ
ζ(·, ;vn)
L()≤C(γ) +J(, ) =C(γ).
From the definition ofJ(vn, ) we obtain
kvnkL(Q)≤C(γ), (a)
ζ(·, ;vn)
L()≤√γC(γ). (b)
Therefore, from Theorem ., we know that there exists a constantCindependent ofn such that
kynkL((,T);H())≤C(γ), (a)
I–αyn
L((,T);H())≤C(γ), (b)
∂
∂tI–αyn
L((,T);L())≤C(γ). (c)
Moreover, from (a) and (a), we have DαRLyn–1yn
L(Q)≤C(γ). () Consequently, there existuγ ∈L(Q),yγ ∈L((,T);H()),δ∈L(Q), η∈L((,T);
H()),θ∈L((,T);L()) and we can extract subsequences of (vn) and (yn) (still called (vn) and (yn)) such that:
vn⇀uγ weakly inL(Q), (a)
DαRLyn–1yn⇀ δ weakly inL(Q), (b)
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 10 of 20
424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470
yn⇀yγ weakly inL (,T);H()
, (c)
I–αyn⇀ η weakly inL [,T],H()
, (d)
∂
∂tI–αyn⇀ θ weakly inL (,T);L()
. (e)
The remaining part of the proof contains three steps.
Step: We show that (uγ,yγ) fulfills ().
SetD(Q), the set ofC∞function onQwith compact support and denote byD′(Q) its dual. Multiplying (a) byϕ∈D(Q) and using Lemma ., (a), and (c), we prove as in [] that
DαRLyn–1yn⇀DαRLyγ –1yγ weakly inD′(Q).
From (b) and the uniqueness of the limit, we conclude
DαRLyγ –1yγ=δ∈L(Q). ()
Hence,
DαRLyn–1yn⇀DαRLyγ –1yγ weakly inL(Q). () Then passing to the limit in (a) and using () and (a), we obtain
DαRLyγ(x,t) –1yγ(x,t) =uγ(x,t), (x,t)∈Q. () On the other hand, we have
Z
Q
I–αyn(x,t)ϕ(x,t)dt dx
= Z
Z T
yn(x,s)
Ŵ( –α) Z T
s
(t–s)–αϕ(x,t)dt
ds dx, ∀ϕ∈D(Q).
Thus using (c) and (d), while passing to the limit, we get Z
Q
ηϕ(x,t)dt dx= Z
Z T
yγ(x,s)
Ŵ( –α) Z T
s
(t–s)–αϕ(x,t)dt
ds dx
= Z
Q
I–αyγ(x,t)ϕ(x,t)dt dx, ∀ϕ∈D(Q).
This implies that I–αyγ(x,t) =η inQ.
Thus, (d) becomes
I–αyn⇀I–αyγ weakly inL [,T],H()
. ()
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 11 of 20
471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517
In view of (), we have
∂
∂tI–αyn⇀ ∂
∂tI–αyγ weakly inD′(Q), and as we have (e), we obtain
∂
∂tI–αyn⇀ ∂
∂tI–αyγ =θ weakly inL(Q). ()
Sinceyγ ∈L(Q) andDαRLyγ–1yγ ∈L(Q), in view of Lemma ., we know thatyγ|∂and
∂yγ
∂v|∂exist and belong toH–((,T);H–/(∂)) andH–((,T);H–/(∂)), respectively.
Moreover, we haveI–αyγ ∈C([,T];L()) and∂t∂I–αyγ∈C([,T];H–()).
Now multiplying (a) by a function ϕ ∈C∞(Q) such that ϕ|∂ = and ϕ(x,T) =
∂ϕ
∂t(x,T) = in, and integrating by parts overQ, we obtain Z
Q
vn(x,t)ϕ(x,t)dx dt= Z
Q
DαRLyn(x,t) –1yn(x,t)
ϕ(x,t)dx dt
= Z
y∂ϕ
∂t(x, )dx +
Z
Q
yn(x,t) DCαϕ(x,t) –1ϕ(x,t) dx dt
because we have (c) and (d). Thus, using (a) and (c) while passing to the limit in the latter identity, we get
Z
Q
uγ(x,t)ϕ(x,t)dx dt= Z
y∂ϕ
∂t(x, )dx +
Z
Q
yγ(x,t) DCαϕ(x,t) –1ϕ(x,t) dx dt,
∀ϕ∈C∞(Q) such thatϕ|∂= ,ϕ(T) =∂ϕ∂t(T) = in, which, according to Lemma ., can be rewritten as
Z
Q
uγ(x,t)ϕ(x,t)dx dt= Z
y∂ϕ
∂t(x, )dx +
Z
Q
DαRLyγ(x,t) –1yγ(x,t)
ϕ(x,t)dx dt +
ϕ(x, ), ∂
∂tI–αyγ(x, )
H(),H–()
– Z
∂ϕ
∂t(x, ),I–αyγ(x, )dx –
yγ(σ,t),∂ϕ
∂ν(σ,t)
H–((,T);H–/(∂)),H((,T);H/(∂))
,
∀ϕ∈C∞(Q) such thatϕ|∂= ,ϕ(T) =∂ϕ∂t(T) = in.
Using (), we obtain
= Z
y∂ϕ
∂t(x, )dx
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 12 of 20
518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564
+
ϕ(x, ), ∂
∂tI–αyγ(x, )
H(),H–()
– Z
∂ϕ
∂t(x, ),I–αyγ(x, )dx –
yγ(σ,t),∂ϕ
∂ν(σ,t)
H–((,T);H–/(∂)),H((,T);H/(∂))
, ()
∀ϕ∈C∞(Q) such thatϕ|∂= ,ϕ(T) =∂ϕ∂t(T) = in.
Choosing successively in ()ϕ such thatϕ(x, ) =∂ϕ∂t(x, ) = andϕ(x, ) = , we de- duce that
yγ(x,t) = , (x,t)∈6, ()
I–αyγ(x, ) =y, x∈, ()
and then
∂
∂tI–αyγ(x, ) = , x∈. ()
In view of (), (), (), and (), we see thatyγ =yγ(x,t;uγ, ) is a solution of ().
Step: We showζn=ζ(x,t;vn) converges toζγ =ζ(x,t;uγ).
In view of (),ζn=ζ(x,t;vn) verifies
DCαζn–1ζn=y(vn, ) –y(, ) inQ, ζn= on6,
ζn(T) = in,
∂
∂tζn(T) = in.
()
Setzn=y(vn, ) –y(, ). In view of () and (),znverifies
DαRLzn–1zn=vn inQ, zn= on6,
I–αzn() = in,
∂
∂tI–αzn() = in.
It follows, from Theorem . and (a), that kznkL((,T);H())=
yn(vn, ) –y(, )
L((,T);H())≤C(γ).
Hence, from Corollary ., we deduce that
kζnkC([,T];H())≤C(γ), ()
∂
∂tζn
C([,T];L())≤C(γ). ()
Since the embedding of C([,T];H()) into L((,T);H()) and the embedding of C([,T];L()) into L(Q) are continuous, we can conclude that there exists ζγ ∈
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 13 of 20
565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611
L((,T);H()) such that
ζn⇀ ζγ weakly inL (,T);H()
. ()
Therefore,
∂
∂tζn⇀ ∂
∂tζγ weakly inD′(Q) and, consequently,
∂
∂tζn⇀ ∂
∂tζγ weakly inL(Q). ()
Sinceζγ ∈L((,T);H()) and ∂t∂ζγ ∈L(Q), we see that ζγ() and ζγ(T) belongs to L(). In view of (), we have
ζγ(T) = in ()
and in view of ()and (), we set
∂
∂tζγ(T) = in. ()
From (b), we deduce that there existsρ∈L() such that
ζ(·, ;vn)⇀ ρ weakly inL(). ()
Multiplying the first equation of () byφ∈D(Q) then, using the integration by parts given by Lemma ., we obtain
Z
Q
y(x,t;vn, ) –y(x,t; , )
φ(x,t)dt dx
= Z
Q
DαRLφ(x,t) –1φ(x,t)
ζn(x,t)dt dx.
Hence, using (c) and () while passing to the limit in the latter identity, we have Z
Q
y x,t;uγ,
–y(x,t; , )
φ(x,t)dt dx
= Z
Q
DαRLφ(x,t) –1φ(x,t)
ζγ(x,t)dt dx, ∀φ∈D(Q), () which by using again Lemma . gives
Z
Q
y x,t;uγ,
–y(x,t; , )
φ(x,t)dt dx
= Z
Q
DαCζγ(x,t) –1ζγ(x,t)
φ(x,t)dt dx, ∀φ∈D(Q).
«ADE layout: Onecolumn v.. file: ade.tex (Aurimas) class: bmc-onecol-v v.// Prn://; : p. /»
« doctopic: OriginalPaper numbering style: ContentOnly reference style: mathphys»
Baleanu et al.Advances in Difference Equations_#####################_ Page 14 of 20
612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658
This implies that Dα
Cζγ–1ζγ=y uγ,
–y(, ) inQ. ()
Now, if we multiply the first equation of () byφ∈C∞(Q) withφ|∂= andI–αφ() = inand integrating by parts overQ, we obtain
Z
Q
yn(x,t) –y(x,t; , )
φ(x,t)dt dx
= Z
Q
DαCζn(x,t) –1ζn(x,t)
φ(x,t)dt dx
= Z
Q
DαRLφ(x,t) –1φ(x,t)
ζn(x,t)dt dx+ Z
ζ(x, ,vn)∂
∂tI–αφ()dx.
Using (c), (), and () while passing the latter identity to the limit, we obtain Z
Q
yγ(x,t) –y(x,t; , )
φ(x,t)dt dx
= Z
Q
DαRLφ(x,t) –1φ(x,t)
ζγ(x,t)dt dx+ Z
ρ∂
∂tI–αφ()dx,
∀φ∈C∞(Q) such thatφ|∂= ,I–αφ() = in, () which by using again Lemma ., (), (), and () gives
– Z
6
∂
∂νφ(σ,t)ζγ(σ,t)dσdt+ Z
ρ(x)∂
∂tI–αφ()dx
= Z
ζγ()∂
∂tI–αφ()dx
∀φ∈C∞(Q) such thatφ|∂= ,I–αφ() = in.
Hence, choosingφ∈C∞(Q), such thatφ|∂= ,I–αφ() =∂t∂I–αφ() = , we get
ζγ = on6, ()
and then
ζγ() =ρ in. ()
In view of (), (), (), and (), we see thatζγ=ζ(uγ) is a solution of
DCαζγ –1ζγ =y(uγ, ) –y(, ) inQ, ζγ= on6,
ζγ(T) = in
∂ζγ
∂t (T) = in.
()
Moreover, using (), equation () becomes ζ(·, ;vn)⇀ ζγ() =ζ ·, ;uγ
weakly inL(). ()