Научная статья на тему 'ON ONE BOUNDARY VALUE PROBLEM FOR THE FOURTH-ORDER EQUATION IN PARTIAL DERIVATIVES'

ON ONE BOUNDARY VALUE PROBLEM FOR THE FOURTH-ORDER EQUATION IN PARTIAL DERIVATIVES Текст научной статьи по специальности «Математика»

CC BY
33
12
i Надоели баннеры? Вы всегда можете отключить рекламу.
Ключевые слова
BOUNDARY VALUE PROBLEM / FOURIER METHOD / THE EXISTENCE OF A SOLUTION / THE UNIQUENESS OF A SOLUTION

Аннотация научной статьи по математике, автор научной работы — Kilichov O.Sh., Ubaydullaev A.N.

The initial-boundary problem for the heat conduction equation inside a bounded domain is considered. It is supposed that on the boundary of this domain the heat exchange takes place according to Newton’s law. The control parameter is equal to the magnitude of output of hot air and is defined on a givenmpart of the boundary. Then we determined the dependence T(Θ) on the parameters of the temperature process when Θ is close to critical value.

i Надоели баннеры? Вы всегда можете отключить рекламу.
iНе можете найти то, что вам нужно? Попробуйте сервис подбора литературы.
i Надоели баннеры? Вы всегда можете отключить рекламу.

Текст научной работы на тему «ON ONE BOUNDARY VALUE PROBLEM FOR THE FOURTH-ORDER EQUATION IN PARTIAL DERIVATIVES»

Vestnik ^AUNC. Fiz.-Mat. nauki. 2022. vol. 39. no. 2. P. 32-41. ISSN 2079-6641

MSC 35J15

Research Article

On one boundary value problem for the fourth-order equation in

partial derivatives

4b University str., Tashkent, 100174, Uzbekistan 2 Bukhara State University, M. Ikbol str. 11, Bukhara, 705018, Uzbekistan E-mail: oybek2402@mail.ru

The initial-boundary problem for the heat conduction equation inside a bounded domain is considered. It is supposed that on the boundary of this domain the heat exchange takes place according to Newton's law. The control parameter is equal to the magnitude of output of hot air and is defined on a givenmpart of the boundary. Then we determined the dependence T(0) on the parameters of the temperature process when 0 is close to critical value.

Key words: boundary value problem; Fourier method; the existence of a solution; the uniqueness of a solution.

For citation.Kilichov O. Sh., Ubaydullaev A.N. On one boundary value problem for the fourth-order equation in partial derivatives. Vestnik KRAUNC. Fiz.-mat. nauki. 2022,39: 2, 32-41. DOI: 10.26117/2079-6641-2022-39-2-32-41

The content is published under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/deed.ru)

© Kilichov O. Sh., Ubaydullaev A. N., 2022

Introduction and statement of the Problem

A problem with a high derivative on a part of the domain boundary was studied, for the first time, by A.N. Tikhonov. In [1], he studied the problem of a homogeneous heat equation with the following conditions

O. Sh. Kilichov1, A.N. Ubaydullaev

2

1 V. I. Romanovskiy Institute of Mathematics, Uzbekistan Academy of Sciences,

d DOI: 10.26117/2079-6641-2022-39-2-32-41 Original article submitted: 17.07.2022

Revision submitted: 10.08.2022

in domain (0 < x < oo, t > 0).

Funding. The work was done without financial support.

In [2], in n—dimensional bounded domain D A.V. Bitsadze studied the problem

, ^ . dmu . . A u(x) = 0, dvm = f(x), x e D,

and proved its Fredholm property.

Boundary value problems with boundary conditions, for the Laplace, Poisson and Helmholtz equations in the unit ball, containing high-order derivatives were studied by I.I. Bavrin [3], V.V. Karachik and B.Kh. Turmetov [4], V.V. Karachik [5]-[7], V.B. Sokolovsky [8]. A mixed problem with a high derivative in the initial condition for the heat equation was studied by D. Amanov [9]; for the equation of string vibration, the mixed problem with high derivatives in the initial conditions was studied in [10]-[16]. Authors of [17], [18]-[19] studied boundary value problems for many types of high-order partial differential equations. Mixed problems for fourth-order differential equations were studied in [20]. The difference of this problem from other problems is that in the initial condition the sum of products of higher orders is set. Consider the following equation

9u 34u . ,

at = f(x>l) (1)

in domain O = {(x,t): 0 < x < p, 0 < t < T}, and the boundary value problem with boundary conditions

u(0, t) = u(p, t) = 0, 0 < t < T, (2)

uxx(0,t)= uxx(p,t)= 0,0 < t < T, (3)

and initial conditions

k=0

= p(x), 0 < x < p, (4)

t=0

where k,m— are the fixed natural numbers, f(x,t), cp(x)— are the given functions, continuous in O , and u(x,t) e CX'Jt (O)— is the sought-for function.

Uniqueness of the solution to the problem

Theorem 1. The solution to problem (1)-(4) is unique if it exists.

Proof. Let f(x,t) = 0, p(x) = 0 in O. We show that u(x,t) = 0 in O. Following [21], we consider the following integral

an(t) =

u(x,t)Xn(x)dx, 0 < t < T (5)

where functions

2 nn

Xn(x) ^nAnx, An = —, n = 1,2,... (6)

form a complete orthonormal system in L2(0,p) [22]. Differentiating (5) over t, from homogeneous equation ut + uxxxx = 0 we find

an (t) =

ut(x,t)Xn(x)dx

or

P,4

On (t) =

94u

dX4 (x,t)Xn(x)dx. (7)

0

Integrating four times by parts over x in the integral (7) on the right side, we obtain the following equations:

an (t)+ A^ ■ an(t)= 0. (8)

The general solution to equation (8) is written as

a-n(t) = Cn ■ exp(-Ait). (9)

To find the unknown coefficient Cn , due to conditions (4), we obtain Cn = 0. Then from (9) it follows that an(t) = 0. Since Xn(x) is the complete orthonormal system in L2(0,p), it follows from the completeness of function Xn(x) that u(x,t) = 0 almost everywhere in O. Considering that u(x,t) e Cx,t (O), we obtain u(x,t) = 0 in O. Theorem 1 is proved. □

Existence of a solution to the problem

Solutions of the inhomogeneous differential equation (1) are sought in the form of a Fourier series in sines

/2 00

u(x,t) = W — ^un(t)sin Anx. (10)

V pn=1

Let the functions f(x,t), cp(x) be expanded in a Fourier series

I— 00

f(x,t) = W — Anx, (11)

V —n=1

I— 00

p(x) = W — ^p^nAnx, (12)

V —n=1

where

p

2

fn(t)^/p

f(x,t)sin Anxdx, (13)

Pn =

p

p(x)sin Anxdx.

(14)

Substituting the Fourier series (10) and (11) into the given inhomogeneous differential equation (1), we obtain a first-order ordinary differential equation

un (t)+ An ■ un(t) = fn(t).

m (k)

Its solution satisfies the condition u(nk)( 0) = pn and has the following form

k=0

Un(t) = Wn(t) +

fn(T)e

-A^t-T)

dT.

(15)

Here and hereinafter (... ) = 0 for m < 0,

s=0

-Ait

Wn(t) =

£ (-Atf

s=0

m-1

Pn fnm-1-s)(0^(-A:

nn

s=0

,i=0

Substituting solution (15) into series (10), we find

u(x,t) = W 2 L

vpn=i

iНе можете найти то, что вам нужно? Попробуйте сервис подбора литературы.

Wn(t) +

fn(T)e

-An(t-T)

dT

sinAnx.

Calculating the products of the solution of equation (1), we obtain

(16)

9u 9t

L A'

n=1

-Wn(t)+ A- 4fn(t) -

fn(T)e-An(t-T)dT

sinAnx, (17)

9x4

LA

n=1

Wn(t) +

fn(T)e

-Ai(t—t)

dT

sinAnx,

(18)

9ku

L (-*

n=1

w.

k-1

t(t)+L (-An )s f

s=0

4 s- ( -1-s)

(t)+

fn(T)e

-An(t-T)

'dT

sinAnx. (19)

Let us show that series (16) and (17)-(19) converge absolutely and uniformly. If the series (19) converges then the series (16)-(18) whose terms are less than the corresponding terms of the series (19) converge absolutely and uniformly. Let us show the absolute and uniform convergence of series (19).

k

Lemma 1. If p(x) e W— (O), p(0) = p(p) = 0, then the series

12 f • A ■ Pn ■ e-Ant ,20.

^^smAnX ^-m--(20)

pn= L (-An)s

s—0

converges absolutely and uniformly in O.

Proof. Simplifying series (20), we obtain the following series

L

n=1

AnX

(-Aj)k ■ ■ e-Ant

L (-An)'

s—0

<

L

n—1

-An

k+1

-f-An

(-An)m+1 -1

■ Pn

If we choose the largest value of k (k < m), we have

L

n—1

(-An)m+1 - (-An)1

(-An)m+1 -1

■ Pn

<

Lipni.

n=1

Integrating (14) by parts, we obtain |pn| — a- ■

lpnl, where

pn —

P [x)\l 2cosAnxdx. P

Applying the Cauchy-Bunyakovsky inequality, we find

L iPni—L

n—1 n—1

By virtue of

An

P(1) Pn

Pi11

<

P

oo , \ 1/2 / oo ,N 1/2

Li LIp

nx*— n

(1) n

n—1

n—1

(oo

LK1

1/2

Pn

2

L2(0,p)

, we have

'2^ . . (-An)k■ Pn■ e-Ant

An

X

<

n—1

L (-AJ)'

s—0

iНе можете найти то, что вам нужно? Попробуйте сервис подбора литературы.

V6

Pn

L2(0,p)

Lemma 1 is proved. □

Lemma 2. If f(x,t) G w24k-3,k-1) (Q), f- — f1 — 0, I — , then

the series

An

X

4 )k „-Ai tm-1

-An ) • e

^fnm-1-s)(0) ■(£ (-a:

n—1

-A4n s s—0

s—0

(21)

vi—0

k-1

^sinAnX (-An) f

n—1

s—0

Osfnk-1-s)(t)

(22)

k

1

2

converge absolutely and uniformly in O.

Proof. Let us expand the series (21) and (22); it is easy to see that the term

L Ank-4fn(t)i

(23)

n=1

is the largest of the two series.

If series (23) converges, then series (21) and (22) also converge. Integrating (13) by parts, we obtain

|fn(t)| =

1

A

4k-3

(4k-3,0)

(t)

(24)

where

fn4k-3,0)(t) =

p94k-3f(x,t) /2 . ^n .

9x4k(-3 V p ^ ((4k- 3) 2 + An^ dx.

0

If to apply (24) to (23), due to the Bunyakovsky and Bessel inequalities [23], the series is

I 00 k-1

2 L*mAnxL (-An)Sfnk-1-s)(t) <

n=1

s=0

V6

94k-3f

dx4k-3

A

->4 \ k -a4 t m-1 / s

-An) e Ant p(m-1 -s)

n=1

nx m ln

£ (-An)s s=0

s=0

s)(0) Z(-a

a=0

L2(û) 94k-3f

9x4k-3

L2(0)

Lemma 2 is proved. □

Lemmi the series

Lemma 3. If f(x,t) e W^4k—3,0) (O), = = 0,1 = 0,1,..., (2k—2), then

L sinAnx ■ (-A^ ■ fn(T)e-An(t-T)dT

n=1

(25)

converges absolutely and uniformly in O.

Proof. Introducing representation (24) into series (25), we have

L

n=1

A

4k

fn (T)e-An(t-T)dT

<

L An

n=1

fn4k-3,0)(T) ■ e-An(t-T) dT

<

<

00

L An ■

n=1

2 /1

fn4k-3,0)(T)

dT

-2A^(t-T)

dT I <

ou -, < - y1 ■

2rcz— n

n=1

(4k-3,0)

(T)

dTI .

n

2

e

iНе можете найти то, что вам нужно? Попробуйте сервис подбора литературы.

2

n

oo

L

n=1

This implies the uniform and absolute convergence of series (25) in O. Lemma 3 is proved. □ Theorem 2. Let

i) p(x) e W—(O), p(0) = p(p)= 0

ii) f (x, t) e W(4k-3'k-1) (O), f1 = f1 = 0,1 = 07C—k—2), _

then the series (16)-(19) converge absolutely and uniformly in O. Solution (16) satisfies equation (1) and conditions (2) - (4).

Proof. By virtue of the proved lemmas, it is easy to show that the series (16)-(19) converge absolutely and uniformly. Adding (17) and (18) we make sure that solution (16) satisfies equation (1) in O . Conditions (2) and (3) are satisfied due to the properties of function Xn(x). Theorem 2 is proved. □

Competing interests. The authors declare that there are no conflicts of interest regarding authorship and publication.

Contribution and Responsibility. All authors contributed to this article. Authors are solely responsible for providing the final version of the article in print. The final version of the manuscript was approved by all authors.

Acknowledgments. The authors are deeply grateful to the referee for a number of comments that contributed to the improvement of the article.

References

1. Tikhonov A.N.About the boundary conditions containing derivatives of an order, exceeding an equation order, Mat. Sb., 1950. vol. 26(1), pp. 35-56 (In Russian).

2. Bitsadze A.VOn the Neumann problem for harmonic functions, Dokl. Akad. Nauk SSSR, 1990. vol.311 (1), pp. 11-13 (In Russian).

3. Bavrin 1.1. Operators for harmonic functions and applications, Differential equations, 1985. vol. 21(1), pp. 9-15 (In Russian).

4. Karachik V. V, Turmetov B. H. About a problem for harmonic equation, Izv. Akad. Nauk UzSSR, Ser. Fiz.-Mat. Nauk, 1990. vol.4, pp. 17-21 (In Russian).

5. Karachik V. V. About solvability of a boundary value problem for Helmholtz's equation with high order normal derivative on a boundary, Differ. Uravneniya, 1992. vol. 28 (5), pp. 907-909 (In Russian).

6. Karachik V. V. About a problem for Poisson's equation with high order normal derivative on boundary, Differ. Uravneniya, 1996. vol.32, no.3, pp. 1501-1503 (In Russian).

7. Karachik V. V. Generalized Neumann problem for harmonic functions in space, Differ. Uravneniya, 1999. vol. 35, no. 7, pp. 1-6 (In Russian).

8. Sokolovskiy V. B. On a generalization of Neumann problem, Differ. Uravneniya, 1998. vol. 24, no. 4, pp. 714-716 (In Russian).

9. Amanov D. On a generalization of the first initial-boundary value problem for the heat conduction equation, Contemporary Analysis and Applied Mathematics, 2014. vol. 2, no. 1, pp. 88-97.

10. Amanov D., Ibragimov G., Kilicman A. On a Generalization of the Initial-Boundary Problem for the Vibrating String Equation, Symmetry, 2019. vol. 11, no. 1.

11. Amanov D., Yuldasheva A. Solvability and spectral properties of a self-adjoint problem for a fourth-order equation, Uzbekskii Matem. Zhurnal, 2007. vol. 4, pp. 3-8 (In Russian).

A

4k

fn (T)e-An(t-T)dT

<

2^

9

4k-3

f

9x4k-3

L2(Q)

12. Amanov D., Murzambetova M. A boundary value problem for a fourth order equation with a lower term, Vestnik Udmurtsk. un-ta. Matem. Mekh. Komp'yut. Nauki, 2013. vol.1, pp. 3-10 (In Russian).

13. Amanov D.On a nonlocal problem for the heat equation, Uzbekskii Matem. journal, 2016. vol.2, pp. 21-25.

14. Kilichov O. Sh. On a nonlocal boundary value problem for the equation fourth-order in partial derivatives, Vestnik KRAUNC. Phys.-Mat. Nauki, 2021. vol.37, no. 4, pp. 16-23 (In Russian).

15. Kilichov O. Sh. Nonlocal boundary value problem for the heat conduction equation, Uzbek Mathematical Journal, 2021. vol.2, pp. 110-116.

16. Kilichov O. Sh. A boundary value problem for a fourth-order equation, Bulletin of the Institute of Mathematics, 2021. vol.4, no. 2, pp. 61-69 (In Russian).

17. Ashurov R. R., Mukhiddinova A. T. Initial-boundary value problems for hyperbolic equations with an elliptic operator of arbitrary order, Vestnik KRAUNC. Phys.-Mat. Nauki, 2020. vol.30, no. 1, pp. 8-19 (In Russian).

18. Yuldashev T. K. Nonlocal mixed-value problem for a Boussinesq-type integro-differential equation with degenerate kernel cubature formulas, Ukrainian Mathematical Journal, 2016. vol.68, no. 8, pp. 1278-1296.

19. Yuldashev T. K. Mixed problem for pseudo parabolic integro-differential equation with degenerate kernel, Differential equations, 2017. vol. 53, no. 1, pp. 99-108.

20. Amanov D., Kilichov O. Sh. Boundary value problem for a fourth-order mixed-type equation in a rectangular domain, Bulletin of the Institute of Mathematics, 2018. vol.2, pp. 1-8 (In Russian).

21. Moiseev Y. I. On the solution by a spectral method of a single non-local boundary value problem, Differential Equations, 1999. vol. 8, no. 35, pp. 1094-1100 (In Russian).

22. Il'in V. A., Poznyak E.G. Osnovy matematicheskogo analiza [Fundamentals of mathematical analysis]. Nauka: Moscow, 1973.448 pp. (In Russian)

23. Lyusternik L. A., Sobolev V. I. Elementy funktsional'nogo analiza [Elements of Functional Analysis]. Nauka: Moscow, 1965. 520 pp. (In Russian)

Kilichov Oybek Sharafiddinovich- Doctoral student, Institute of Mathematics of the Academy of Sciences of the Republic of Uzbekistan, Tashkent, Uzbekistan, ORCID 0000-0002-7673-943X.

Ubaydullaev Alisher NematillayevichÄ - Teacher of the Department of Mathematics of Bukhara State University, Bukhara, Uzbekistan, ©ORCID 0000-0002-4219-5155.

Вестник КРАУНЦ. Физ.-мат. науки. 2022. Т. 39. №. 2. С. 32-41. ISSN 2079-6641

УДК 517.95 Научная статья

Об одной краевой задаче для уравнения четвертого порядка в

частных производных

О. Ш. Киличов1, А.Н. Убайдуллаев2

1 Институт математики имени В. И. Романовского Академии наук Узбекистана, 4б ул. Университетская, г. Ташкент, 100174, Узбекистан

2 Бухарский государственный университет, ул. М. Икбол, 11, г. Бухара, 705018, Узбекистан

E-mail: oybek2402@mail.ru

Рассмотрена начально-краевая задача для уравнения теплопроводности внутри ограниченной области. Предполагается, что на границе этой области происходит теплообмен по закону Ньютона. Параметр управления равен величине выхода горячего воздуха и определяется на заданном участке границы. Затем была определена зависимость T(0) от параметров температурного процесса, когда 0 близко к критическому значению.

Key words: краевая задача; метод Фурье; существование решения; единственность решения.

d DOI: 10.26117/2079-6641-2022-39-2-32-41

Поступила в редакцию: 17.07.2022 В окончательном варианте: 10.08.2022

Для цитирования. Kilichov O. Sh., Ubaydullaev A. N. On one boundary value problem for the fourth-order equation in partial derivatives // Вестник КРАУНЦ. Физ.-мат. науки. 2022. Т. 39. № 2. C. 32-41. DOI: 10.26117/2079-6641-2022-39-2-32-41

Конкурирующие интересы. Авторы заявляют, что конфликтов интересов в отношении авторства и публикации нет.

Авторский вклад и ответственность. Все авторы учавствовали в написании статьи и полностью несут ответственность за предоставление ококнчательной версии статьи в печать. Окончательная форма рукописи была одобрена всеми авторами.

Контент публикуется на условиях лицензии Creative Commons Attribution 4.0 International (https://creativecommons.org/licenses/by/4.0/deed.ru)

© Киличов О.Ш., Убайдуллаев А.Н., 2022

Финансирование. Работа выполнена без финансовой поддержки.

Киличов Ойбек Шарафиддинович^ - докторант, Институт математики Академии наук Республики Узбекистан, Ташкент,

Узбекистан, ОЫСГО 0000-0002-7673-943Х.

Убайдуллаев Алишер Нематиллаевич& - Преподаватель кафедры математики Бухарского государственного университета,

Бухара, Узбекистан, СЖСГО 0000-0002-4219-5155.

i Надоели баннеры? Вы всегда можете отключить рекламу.