Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction

We consider the boundary value problem, where the motion of the object is described by the two-dimensional linear system of partial differential equations of hyperbolic type where a discontinuity is at a point within the interval that defines the phase coordinate x. Using the method of series and La...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Журнал математической физики, анализа, геометрии
Datum:2016
Hauptverfasser: Aliev, F.A., Aliev, N.A., Guliev, A.P.
Format: Artikel
Sprache:Englisch
Veröffentlicht: Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України 2016
Online Zugang:https://nasplib.isofts.kiev.ua/handle/123456789/140549
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Назва журналу:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Zitieren:Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction / F.A. Aliev, N.A. Aliev, A.P. Guliev // Журнал математической физики, анализа, геометрии. — 2016. — Т. 12, № 2. — С. 101-112. — Бібліогр.: 27 назв. — англ.

Institution

Digital Library of Periodicals of National Academy of Sciences of Ukraine
_version_ 1859816243635159040
author Aliev, F.A.
Aliev, N.A.
Guliev, A.P.
author_facet Aliev, F.A.
Aliev, N.A.
Guliev, A.P.
citation_txt Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction / F.A. Aliev, N.A. Aliev, A.P. Guliev // Журнал математической физики, анализа, геометрии. — 2016. — Т. 12, № 2. — С. 101-112. — Бібліогр.: 27 назв. — англ.
collection DSpace DC
container_title Журнал математической физики, анализа, геометрии
description We consider the boundary value problem, where the motion of the object is described by the two-dimensional linear system of partial differential equations of hyperbolic type where a discontinuity is at a point within the interval that defines the phase coordinate x. Using the method of series and Laplace transformation in time t (time-frequency method), an analytical solution is found for the determination of debit Q(2l, t) and pressure P(2l, t), which can be effective in the calculation of the coefficient of hydraulic resistance in the lift at oil extraction by gas lift method where l is the well depth. For the case where the boundary functions are of exponential form, the formulas for P(2l, t) and Q(2l, t) depending only on t are obtained. It is shown that at constant boundary functions, these formulas allow us to determine the coefficient of hydraulic resistance in the lift of gas lift wells, which determines the change in the dynamics of pollution.
first_indexed 2025-12-07T15:22:08Z
format Article
fulltext Journal of Mathematical Physics, Analysis, Geometry 2016, vol. 12, No. 2, pp. 101–112 Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction F.A. Aliev, N.A. Aliev, and A.P. Guliev Institute of Applied Mathematics, Baku State University 23, Z. Khalilov Str., Baku AZ1148, Azerbaijan E-mail: f aliev@yahoo.com Received May 29, 2014, revised November 30, 2015 We consider the boundary value problem, where the motion of the ob- ject is described by the two-dimensional linear system of partial differential equations of hyperbolic type where a discontinuity is at a point within the interval that defines the phase coordinate x. Using the method of series and Laplace transformation in time t (time-frequency method), an analytical so- lution is found for the determination of debit Q(2l, t) and pressure P (2l, t), which can be effective in the calculation of the coefficient of hydraulic re- sistance in the lift at oil extraction by gas lift method where l is the well depth. For the case where the boundary functions are of exponential form, the formulas for P (2l, t) and Q(2l, t) depending only on t are obtained. It is shown that at constant boundary functions, these formulas allow us to determine the coefficient of hydraulic resistance in the lift of gas lift wells, which determines the change in the dynamics of pollution. Key words: hyperbolic equation, boundary problems, method of series, Laplace transformation, time-frequency method, gas lift, coefficient of hy- draulic resistance. Mathematics Subject Classification 2010: 65M38, 35L02, 35L40, 58J45, 58J90. Introduction As known [1–4], only a considerably small part of oil is extracted from the reservoir by gushing method. The rest of oil is extracted by secondary and tertiary methods (gas lifting [2–4], subsurface pumping [1, 2], etc.). As considerable volumes of oil are produced from wells by gas lift method [2, 3, 5–12], researchers have paid a lot of attention to creating mathematical models of this process c© F.A. Aliev, N.A. Aliev, and A.P. Guliev, 2016 F.A. Aliev, N.A. Aliev, and A.P. Guliev lately. However, in [5], the authors, averaging the motion equation (in partial derivatives) with respect to time t and depth E, solved the general problem of control and optimization for the corresponding ordinary differential equations. These simplifications make it possible to obtain “approximate” mathematical models instead of exact ones. Therefore the authors of [13] applying the difference schemes [14] and the method of series by the well depth [15] for the motion equation of gas-liquids mixture in the lift obtained the total number of terms of the series which defines the solution representation to a given accuracy. In the present paper, under some restrictions imposed on the boundary conditions [16– 18], an analytical form of the solution of the motion equation in the considered problem is given. In some partial cases, such a representation (when the volume of the gas supplied ’shoe’ is constant) allows us to define the analytical expression of the coefficient of hydraulic resistance in the pump-compressor pipe and the change of pollution dynamics [19, 20]. 1. The Problem Statement and the Method of Series for Solving a Boundary Value Problem in an Annular Space Let us consider the boundary value problem for the system of equations of hyperbolic type { −F1 ∂P1(x,t) ∂x = ∂Q1(x,t ) ∂t + 2a1Q1(x, t), −F1 ∂P1( x,t) ∂t = c1 ∂Q1(x,t) ∂x , x > 0, t ∈ R, (1) with the boundary conditions { P1(0, t) = P10(t), Q1(0, t) = Q10(t), t ∈ R, (2) where P1(x, t) is a pressure, Q1(x, t) is a gas volume, 2a1 = g ωc1 + λωc1 2D1 , (3) ωc1 is an average velocity of a flow with respect to the length of the annular pipeline, λ is the coefficient of hydraulic resistances, D1 is the effective diameter of the annular space, F1 is the area of the cross-section of pump-compressor pipes, which is constant along the axis x, c1 is the sound speed in gas, g is the gravity acceleration, P10(t) and Q10(t) are given continuous functions specifying the initial pressure and the volume of the gas supplied to the “shoe”. Thus, it is required to find P (x, t) and Q(x, t) which satisfy conditions (1), (2), where P (x, t) is the pressure and Q(x, t) is the gas volume at each point (x, t) in the pipe. 102 Journal of Mathematical Physics, Analysis, Geometry, 2016, vol. 12, No. 2 Time Frequency Method of Solving One Boundary Value Problem... We will search the solution of problem (1)–(3) in the form of [15, 21–23] as following from [24], { P1(x, t) = ∑∞ k=0 P1k(t)xk k! , Q1(x, t) = ∑∞ k=0 Q1k(t)xk k! , x > 0, t ∈ R, (4) which satisfies the boundary condition (2). Differentiating (4) and substituting it into (1), for the unknown functions P1k(t) and Q1k(t), for k > 0, we have:    P1,2k(t) = ( d dt + 2a1 )k P (k) 10 (t) ck 1 , Q1,2k(t) = ( d dt + 2a1 )k Q (k) 10 (t) ck 1 , k ≥ 0, (5) and    P1,2k+1(t) = − ( d dt + 2a1 )k+1 Q (k) 10 (t) F1 ck 1 , Q1,2k+1(t) = − ( d dt + 2a1 )k F1P (k+1) 10 (t) ck+1 1 , k ≥ 0. (6) Introduce the Laplace transforms [25]: { P̃10(ρ) = ∫∞ 0 e−ρtP10(t)dt, Q̃10(ρ) = ∫∞ 0 e−ρtQ10(t)dt, (7) allowing one to recover P10 and Q10 by the the formulas { P10(t) = 1 2πi ∫ L1 eρtP̃10(ρ)dρ, Q10(t) = 1 2πi ∫ L1 eρtQ̃10(ρ)dρ, (8) where L1 = ρ : Reρ = σ1 > 0. Then, taking into account (8), in (5) and (6) we obtain { P1,2k(t) = 1 2πi ∫ L1 eρt(ρ + 2a1)kρkP̃10(ρ)dρ ck 1 , Q1,2k(t) = 1 2πi ∫ L1 eρt(ρ + 2a1)kρkQ̃10(ρ)dρ ck 1 (9) and    P1,2k+1(t) = − 1 2πi ∫ L1 eρt(ρ + 2a1)k+1ρkQ̃10(ρ) dρ F1ck 1 , Q1,2k+1(t) = − 1 2πi ∫ L1 eρt(ρ + 2a1)kF1ρ k+1P̃10(ρ) dρ ck+1 1 . (10) Thus, in (9), (10) and (4), we have the following analytic expressions for the solution of equation (1): P1(x, t) = ∑∞ k=0 P1,2k(t) x2k (2k)! + ∑∞ k=0 P1,2k+1(t) x2k+1 (2k+1)! = 1 2πi ∫ L1 eρt { P̃10(ρ)ch (√ ρ(ρ+2a1) c1 x ) − Q̃10(ρ) F1 √ c1(ρ+2a1) ρ sh (√ ρ(ρ+2a1) c1 x )} dρ, (11) Journal of Mathematical Physics, Analysis, Geometry, 2016, vol. 12, No. 2 103 F.A. Aliev, N.A. Aliev, and A.P. Guliev Q1(x, t) = ∑∞ k=0 Q1,2k(t) x2k (2k)! + ∑∞ k=0 Q1,2k+1(t) x2k+1 (2k+1)! = 1 2πi ∫ L1 eρt { Q̃10(ρ)ch (√ ρ(ρ+2a1) c1 x ) − P̃10(ρ) F1 √ ρ c1(ρ+2a1) sh (√ ρ(ρ+2a1) c1 x )} dρ, (12) which can be formulated as the theorem below. Theorem 1. If F1, a1, c1 are constant, P10(t) and Q10(t) are infinitely dif- ferentiable functions, then BVP (1)-(2) has a solution analytically represented in (11) and (12). The proof of Theorem 1 follows from (11) and (12). From (11) and (12) at x = l, we define P1(l − 0, t) and Q1(l − 0, t) in the form:    P1(l − 0, t) = P1(l, t) = 1 2πi ∫ L1 eρt { P̃10(ρ)ch(µl)− Q̃10(ρ) F1 √ c1(ρ+2a1) ρ sh(µl) } dρ, Q1(l − 0, t) = Q1(l, t) = 1 2πi ∫ L1 eρt { Q̃10(ρ)ch(µl)−P̃10(ρ)F1 √ ρ c1(ρ+2a1) sh(µl) } dρ, (13) where (µ) = (√ ρ(ρ+2a1) c1 l ) . 2. The Solution of Equation (1) in the Lift x > l Now we pass from the end of the first stage to the initial position of the second stage, i.e., let the hyperbolic equation (1) at (l, t) have discontinuity [26], { P2(l + 0, t)=P2(l, t) = F 1 δ P1(l − 0, t) + χ1(P1(l − 0, t), α1, α2, α3)P̄ (t) ≡ P20(t), Q2(l + 0, t)=Q2(l, t) = F 2 δ Q1(l − 0, t) + χ2(Q1(l − 0, t), β1, β2, β3)Q̄(t)≡Q20(t), (14) where χ1(·), χ2(·) are continuous functions of their arguments, P̄ (t), Q̄(t) are given continuous functions that are a possible balance of oil products [9], F i δ(i = 1, 2), αk, βk(k = 1, 3) are real numbers to be determined by using the corre- sponding identification problems for creation of gas-liquid mixture in the well bottom zone. Now we pass to the second stage. Let us consider the boundary value problem (1)–(3) for a system of equations of hyperbolic type { −F2 ∂P2(x,t) ∂x = ∂Q2(x,t ) ∂t + 2a2Q2(x, t), −F2 ∂P2( x,t) ∂t = c2 ∂Q2(x,t) ∂x , x > l, t ∈ R, (15) with the boundary conditions { P2(l, t) = P20(t), Q2(l, t) = Q20(t), t ∈ R, (16) 104 Journal of Mathematical Physics, Analysis, Geometry, 2016, vol. 12, No. 2 Time Frequency Method of Solving One Boundary Value Problem... where P20(t) and Q20(t) are given in the form of (14). Then system (15) and unknowns P2(·), Q2(·) are similar to those of the previous case, and a2, similarly to k in (3), will be 2a2 = g ω2 + λω2 2D2 . (17) Analogously to the previous case from Sec. 2, the solution of the boundary problem (15)–(17) will be sought in the form of [15]: { P2(x, t) = ∑∞ k=0 P2k(t)xk k! , Q2(x, t) = ∑∞ k=0 Q2k(t)xk k! , x > l, t ∈ R, (18) for which the boundary conditions (16) are satisfied automatically, and P2k(t) and Q2k(t) are to be defined for k > 0. If not to repeat the calculations given above for solving (18), we can get the following expressions:    P2(x, t) = 1 2πi ∫ L2 ert { P̃20(r)ch(µx)− Q̃20(r) F2 √ c2(r+2a2) r sh(µx) } dr, Q2(x, t) = 1 2πi ∫ L2 ert { Q̃20(r)ch(µx)− P̃20(r)F2 √ r c2(r+2a2) sh(µx) } dr, (19) where µ = (√ r(r+2a2) c2 x, ) , which, for x = 2l, have the form    P2(2l, t) = 1 2πi ∫ L2 ert { P̃20(r)ch(µ2l)− Q̃20(r) F2 √ c2(r+2a2) r sh(µ2l) } dr, Q2(2l, t) = 1 2πi ∫ L2 ert { Q̃20(r)ch(µ2l)− P̃20(r)F2 (√ r c2(r+2a2) ) sh(µ2l) } dr, (20) where µ = (√ r(r+2a2) c2 2l ) , L2, like L1, is a Laplace line, i.e., if r ∈ L2, then Rer = σ2 > 0. Thus, the following theorem is proved: Theorem 2. Under the conditions of Theorem 1, if F2, a2, c2 are con- stant numbers, P20(t) and Q20(t) are infinitely differentiable functions, then BVP (15)–(16) has the solutions represented in the analytical form (19). 3. Simplified Scheme It should be noted that if the boundary functions P0(t), Q0(t) have an expo- nential form, then the solutions (11), (12) and (19) become much more simplified. In fact, let { P10(t) = α1e ν1t, Q10(t) = β1e µ1t, t ∈ R. (21) Journal of Mathematical Physics, Analysis, Geometry, 2016, vol. 12, No. 2 105 F.A. Aliev, N.A. Aliev, and A.P. Guliev Then P̃10(ρ) = ∞∫ 0 e−ρtP10(t)dt = ∞∫ 0 α1e ν1t e−ρtdt = α1 ∞∫ 0 e−(ρ−ν1)tdt, where the constants α1, β1, ν1, µ1 are given real numbers. If Reρ = σ1 > ν1, then { P̃10(ρ) = α1 e−(ρ−ν1)t −(ρ−ν1) ∣∣∣ ∞ t=0 = α1 ρ−ν1 , Q̃10(ρ) = β1 ρ−µ1 . (22) Substituting (22) into (11), (12) and taking into account the analyticity of the functions ch (√ ρ(ρ+2a1) c1 x ) and √ c1(ρ+2a1) ρ sh (√ ρ(ρ+2a1) c1 x ) , via the variable ρ we find    P1(x, t) = α1e ν1tch (√ ν1(ν1+2a1) c1 x ) − β1e µ1t 1 F1 √ c1(µ1+2a1) µ1 sh (√ µ1(µ1+2a1) c1 x ) , Q1(x, t) = β1e µ1tch (√ µ1(µ1+2a1) c1 x ) − α1e ν1tF1 √ ν1 c1(ν1+2a1) sh (√ ν1(ν1+2a1) c1 x ) . (23) Thus, expression (13) finally takes the form    P1(l, t) = α1e ν1tch (√ ν1(ν1+2a1) c1 l ) − β1e µ1t 1 F1 √ c1(µ1+2a1) µ1 sh (√ µ1(µ1+2a1) c1 l ) , Q1(l, t) = β1e µ1tch (√ µ1(µ1+2a1) c1 l ) − α1e ν1tF1 √ c1(ν1+2a1) ν1 sh (√ ν1(ν1+2a1) c1 l ) . (24) Now we suggest that from (14) we have { P20(t) = P 2 1 (l, t) · P2, Q20(t) = −Q2 1(l, t) Q2, (25) where P2 > 0 and Q2 > 0 are constant numbers. Then P̃20(r) = ∫∞ 0 e−rtP20(t)dt = P2α 2 1ch 2 (√ ν1(ν1+2a1) c1 l ) 1 r−2ν1 −2α1β1P2 1 F1 √ c1(µ1+2a1) µ1 sh (√ µ1(µ1+2a1) c1 l ) ch (√ ν1(ν1+2a1) c1 l ) × 1 r−(ν1+µ1) + P2β 2 1 1 F 2 1 c1(µ1+2a1) µ1 sh2 (√ µ1(µ1+2a1) c1 l ) 1 r−2µ1 , (26) 106 Journal of Mathematical Physics, Analysis, Geometry, 2016, vol. 12, No. 2 Time Frequency Method of Solving One Boundary Value Problem... Q̃20(r) = ∫∞ 0 e−rtQ20(t)dt = −Q2β 2 1ch2 (√ µ1(µ1+2a1) c1 l ) 1 r−2µ1 +2α1β1Q2F1 √ ν1 c1(ν1+2a1) sh (√ ν1(ν1+2a1) c1 l ) ch (√ µ1(µ1+2a1) c1 l ) 1 r−(ν1+µ1) −Q2α 2 1F 2 1 ν1 c1(µ1+2a1) sh2 (√ ν1(ν1+2a1) c1 l ) 1 r−2ν1 . (27) Here it is supposed that Re r > max {2ν1, (ν1 + µ1), 2µ1} . At last, from (20), we obtain P2(2l, t) = P2α 2 1ch (√ 2ν1 2(ν1+a2) c2 2l ) ch2 (√ ν1(ν1+2a1) c1 l ) e2ν1t −2α1β1P2 1 F1 √ c1(µ1+2a1) µ1 sh (√ µ1(µ1+2a1) c1 l ) ch (√ ν1(ν1+2a1) c1 l ) ×ch (√ (ν1+µ1)(ν1+µ1+2a1) c2 2l ) e(ν1+µ1)t + P2β 2 1 1 F 2 1 c1(µ1+2a1) µ1 ×sh2 (√ µ1(µ1+2a1) c1 l ) ch (√ 2µ1 2(µ1+a2) c2 2l ) e2µ1t + Q2β 2 1 1 F2 √ c2(µ1+a2) µ1 ×sh (√ 2µ1 2(µ1+a2) c2 2l ) ch2 (√ µ1(µ1+2a1) c1 l ) e2µ1t − 2α1β1Q2 F1 F2 × √ ν1 c1(ν1+2a1) sh (√ ν1(ν1+2a1) c1 l ) ch (√ µ1(µ1+2a1) c1 l )√ c2(ν1+µ1+2a2) ν1+µ1 ×sh (√ (ν1+µ1)(ν1+µ1+2a2) c2 2l ) e(ν1+µ1)t + Q2α 2 1F 2 1 ν1 c1(ν1+2a1) ×sh2 (√ ν1(ν1+2a1) c1 l ) 1 F2 √ c2(ν1+a2) ν1 sh (√ 2ν1 2(ν1+a2) c2 2l ) e2ν1t, (28) Q2(2l, t) = −Q2β 2 1ch2 (√ µ1(µ1+2a1) c1 l ) ch (√ 2µ1 2(µ1+a2) c2 2l ) e2µ1t +2α1β1Q2F1 √ ν1 c1(ν1+2a1) sh (√ ν1(ν1+2a1) c1 l ) ch (√ µ1(µ1+2a1) c1 l ) ×ch (√ (ν1+µ1)(ν1+µ1+2a2) c2 2l ) e(ν1+µ1)t −Q2α 2 1F 2 1 ν1 c1(ν1+2a1) ×sh2 (√ ν1(ν1+2a1) c1 l ) ch (√ 2ν1 2(ν1+a2) c2 2l ) e2ν1t − P2α 2 1 ×ch2 (√ ν1(ν1+2a1) c1 l ) F2 (√ ν1 c2(ν1+a2) ) sh (√ 2ν1 2(ν1+a2) c2 2l ) e2ν1t +2α1β1P2 1 F1 √ c1(µ1+2a1) µ1 sh (√ µ1(µ1+2a1) c1 l ) ch (√ ν1(ν1+2a1) c1 l ) ×F2 (√ ν1+µ1 c2(ν1+µ1+2a2) ) sh (√ (ν1+µ1)(ν1+µ1+2a2) c2 2l ) e(ν1+µ1)t Journal of Mathematical Physics, Analysis, Geometry, 2016, vol. 12, No. 2 107 F.A. Aliev, N.A. Aliev, and A.P. Guliev −P2β 2 1 1 F 2 1 c1(µ1+2a1) µ1 sh2 (√ µ1(µ1+2a1) c1 l ) F2 (√ µ1 c2(µ1+a2) ) ×sh (√ 2µ1 2(µ1+a2) c2 2l ) e2µ1t, (29) which proves the theorem below. Theorem 3. If Fi, ai, ci, i = 1, 2, are constant numbers, P10(t) and Q10(t) have the form of (21), P20(t) and Q20(t) have the form of (25), α1, β1, ν1, µ1, P2, Q2 are constant numbers, then (28) and (29) hold true for solving BVP (1)–(3), (14)–(17). 4. Determination of Coefficient of Hydraulic Resistance [19, 20] Now we use the relationships (28) and (29) to define λ, the coefficient of hydraulic resistance [19, 20] (CHR), from (16), i.e., to define the pollution dy- namics in the lift. Note that this problem is essentially the inverse problem to that considered for parabolic systems in [27]. Let statistical data for the volume of the supplied gas Q̃10(t) from (2) and debit Q̃(2l, t) be given. Then writing the functional in the form RQ(λ) = T∫ 0 [ Q(2l, t)− Q̃(2l, t) ]2 dt, (30) we have to find λ∗ such that the functional R(λ) gets its minimum. Therefore, we calculate the first variation of the functional (30) and make it equal zero T∫ 0 [ Q(2l, t)− Q̃(2l, t) ] Q′ λ(2l, t)dt = 0. (31) Taking into account (29) and (31), for defining the CHR λ we can obtain a transcendent equation which allows us to find a numerical meaning of the optimal λ∗. E x a m p l e. Let us consider BVP (1), (2) (see, [8]), where the boundary data P10(t) = P0, Q10(t) = Q0 from (2) are constant. While in [15], in the domain [0, l]× [0, T ], P (x, t) = P0 − 2 a1 F1 Q0x, Q(x, t) = Q0 (32) and at the point (l − 0, t), P (l − 0, t) = P0 − 2 a1 F1 Q0l, Q(l − 0, t) = Q0. 108 Journal of Mathematical Physics, Analysis, Geometry, 2016, vol. 12, No. 2 Time Frequency Method of Solving One Boundary Value Problem... Suggesting that in (14) P̄ (t) = P̄ , Q̄(t) = Q̄, α1 = −1, α2 = α3 = 0, β1 = 1, β2 = β3 = 0, F 1 δ = F 2 δ = 1, χ1 = P 2(l − 0, t), χ2 = −Q2(l − 0, t), at the point (l + 0, t) we have (boundary layer [8]) P (l + 0, t) = P (l − 0, t) + P 2(l − 0, t)P̄ = Pl+0 = const, Q(l + 0, t) = Q(l − 0, t)−Q2(l − 0, t)Q̄ = Ql+0 = const. (33) Similarly to the above, it is easy to show that in the domain [l, 2l]× [0, T ] , P (x, t) = Pl+0 − 2a2 F2 Ql+0x, Q(x, t) = Ql+0. (34) According to (34), Q(x, t) is constant on [l, 2l]× [0, T ]. This makes it difficult to find the CHR λ. Besides, the expression for P (x, t) contains a2(λ). This requires the necessity to have an additional information on the history of the pressure both at x = 0 (gas supply) and at x = 2l (gas measurement). Let P̃ (0, t) = P̃0(0) = P̃0 , P̃ (2l, t) = P̃2l . (35) Let us write the functional (30) for the pressure P (x, t), RP (λ) = T∫ 0 [ P (2l, t)− P̃ (2l, t) ]2 dt, (36) where at the point x = 2l from (34) we have P (2l, t) = Pl+0 − 2a2 F2 Ql+0 2l. (37) Taking into account that there is no dependence from t either in (35) or (37), the functional (36) takes the form RP (λ) = (P (2l, t)− P̃2l)2T = (Pl+0 − 2a2(λ) F2 Ql+0 l − P̃2l)2T. (38) Now, similarly as in (31), we have the gradient of the functional in the form a2(λ) = F2 4lQ(l + 0) (Pl+0 − P̃2l). (39) Taking into account (39), in (17), for λ, we have λ = −2D2g ω0 ω2 + F2D2 2ω2lQ(l + 0) (Pl+0 − P̃2l). (40) Journal of Mathematical Physics, Analysis, Geometry, 2016, vol. 12, No. 2 109 F.A. Aliev, N.A. Aliev, and A.P. Guliev Now we will show the above on the numerical example. Let for 0 ≤ x < l, l = 1485m, c1 = 331m/c, ρ1 = 0, 717 : 3/m3, F1 = 0.006 , λ1 = 0.01, D1 = 0.0876, g = 9.81,ω1 = 48.6390, and for l ≤ x < 2l, c = 850, D2 = 0.073, F2 = 0.0042, ω2 = 0.0546, λ2 = 0.23, Q̄ = 100, P̄ = 100. (41) Performing the corresponding calculations using the values of the latter parame- ters, we have: P0 − 2a1 F1 Q0l = −2.466 · 103, Pl+0 = Pl−0 + P 2 l−0 · P̄ = 1.3494 · 1015, Pl−0 = P0 − 2 a1 F1 Q0l = −3.6734 · 106, Ql+0 = Q0 −Q2 0Q̄ = −2495, Q0 = 0.001, P̄ = 0.00135, P̃ (2l, t) = P̃2l = 1.34 · 1015. Then, for the value P̃2l = 1.34 · 1015 from (40), we can find λ∗ in the form λ∗ = 2.2999 · 10−1 which coincides with λ2 from (41) with an accuracy of 10−3. Thus, the stated problems can be extended to the more difficult problem (1)–(3). References [1] V.I. Shurov, Technology and Techniques of Oil Recovery. Nedra, Moscow, 1983. [2] E. Camponogara, A. Plucenio, A.F. Teixeira, and S.R.V. Campos, An Automation System for Gas-Lifted Oil Wells: Model Identification Control and Optimization. — J. Petroleum Sci. and Engineering 70 (2010), 157–167. [3] F.A. Aliev, N.A. Ismailov, and N.S. Mukhtarova, Algorithm to Determine the Optimal Solution of a Boundary Control Problem. — Automation and Remote Control 76 (2015), No. 4, 627–633. [4] F.A. Aliev, M.Kh. Ilyasov, and M.A. Dzamalbekov, Modeling of Work of Gas-Lift Wells. — Rep. of NAS of Azerbaijan 4 (2008), 30–41. [5] F.A. Aliev, M.Kh. Ilyasov, and N.B. Nuriev, The Problems of Modeling and Opti- mal Stabilization of Gas Lift Process. — Appl. Mechanics (2010), No. 6, 115–122. [6] F.A. Aliev, M.M. Mutallimov, I.M. Askerov, and I.S. Raguimov, Asymptotic Method of Solution for a Problem of Construction of Optimal Gas-lift Process Modes. Mathematical Problems in Engineering (2010), Article ID 191053, 10 pages. 110 Journal of Mathematical Physics, Analysis, Geometry, 2016, vol. 12, No. 2 Time Frequency Method of Solving One Boundary Value Problem... [7] F.A. Aliev, N.A. Aliev, K.G. Hasanov, A.K. Turarov, A.P. Guliev, and G.V.Isaeva, Numerical-Analytical Method for Solving of the First Order Partial Quasi-Linear Equations, TWMS J. — Pure Appl. Math. 6 (2015), No. 2, 158–164. [8] M. Ashraf, S. Asghar, and M.A. Hossain, The Computational Study of the Effects of Magnetic Field and Free Stream Velocity Oscillation on Boundary Layer Flow Past a Magnetized Vertical Plate. — Appl. Comput. Math. 13 (2014), No. 2, 175–193. [9] A.Kh. Mizadjanzadeh, I.M. Ametov, A.M. Khasaev, and V.I. Gusev, Technology and Technique of Oil Extractions. Nedra, Moscow, 1986. [10] F.A. Aliev, M.M. Mutallimov, N.A. Ismailov, and M.F. Racabov, An Algorithm for Constructing Optimal Controllers for Gaslift Operation. — Automation and Remote Control 73 (2012), No. 8, 12791289. [11] F.A. Aliev, M.A. Djamalbekov, and M.Kh. Ilyasov, Mathematical Simulation and Control of Gas-Lift. — J. Computer and Systems Sciences International 50 (2011), No. 1, 805–814. [12] F.A. Aliev and N.A. Ismailov, Problems of Optimization with the Periodic Bound- ary Condition and Boundary Control in Gas Lift Wells. — Nonlinear Vibrations (2014), No. 2, 216–227. [13] A.P. Guliev, M.Kh. Ilyasov, N.A. Aliev, and F.A. Aliev, The Algorithm for Solving the Problem of Determining the Motion of Space Process. — Proc. IAM 2 (2013), No. 1, 91–97. [14] A.O. Gelfand, Calculation of Finite Differences. Nauka, Moscow, 1967. [15] N.A. Aliev, F.A. Aliev, A.P. Guliev, and M.Kh. Ilyasov, Method of Series in the Solution of One Boundary Problem for the System of Hyperbolic Type Equations, Arising in the Oil Production. — Proc. IAM 2 (2013), No. 2, 113–136. [16] S. Alkan and A. Secer, Solution of Nonlinear Fractional Boundary Value Problems with Nonhomogeneous Boundary Conditions. — Appl. Comput. Math. 14 (2015), No. 3, 284–295. [17] J. Rashidinia, M. Khazaei, and H. Nikmarvani, Spline Collocation Method for Solution of Higher Order Linear Boundary Value Problems, TWMS J. — Pure Appl. Math. 6 (2015), No. 1, 38–47. [18] L.K. Vashisht, Brief Paper. Banach Frames Generated by Compact Operators As- sociated with a Boundary Value Problem, TWMS J. — Pure Appl. Math. 6 (2015), No. 2, 254–258. [19] F.A. Aliev and N.A. Ismailov, Inverse Problem to Determine the Hydraulic Resis- tance Coefficient in the Gas Lift Process. — Appl. Comput. Math. 12 (2013), No. 3, 306–313. [20] F.A. Aliev, N.A. Ismailov, and A.A. Namazov, Asymptotic Method for Finding the Coefficient of Hydraulic Resistance in Lifting of Fluid on Tubing. — Inverse and Ill Posed Problems 23 (2015), No. 5, 511–518. Journal of Mathematical Physics, Analysis, Geometry, 2016, vol. 12, No. 2 111 F.A. Aliev, N.A. Aliev, and A.P. Guliev [21] G.D. Birkhoff, On the Asymptotic Character of the Solutions of Certain Linear Differential Equations Containing a Parameter. — Trans. Am. Math Soc. 9 (1908), 219–231; Boundary Value and Expansion Problems of Ordinary Linear Differential Equations. AMS 9 (1908), No. 4. [22] A. Erdeyi, Asymptotic Decompositions. Moscow, 1962. [23] S. Tiwari and M. Kumar, An Initial Value Technique to Solve Two-Point Non- Linear Singularly Perturbed Boundary Value Problems. — Appl. Comput. Math. 14 (2015), No. 2, 150–157. [24] Hector Vazquez-Leal, The Enhanced Power Series Method to Find Exact or Ap- proximate Solutions of Nonlinear Differential Equations. — Appl. Comput. Math. 14 (2015), No. 2, 168–179. [25] M.A. Lavrentyev and V.V. Shabat, Methods of Theory of Functions of Complex Variables. Nauka, Moscow, 1973. [26] Z.O. Melnik, Solution of a Mixed Problem for Hyperbolic Equations with Discon- tinuous Coefficients. — Differ. Eqs. 2 (1966), No. 4, 560–570. [27] M. Mohammadi, R. Mokhtari, and H. Panahipour, Solving Two Parabolic Inverse Problems with a Nonlocal Boundary Condition in the Reproducing Kernel Space. — Appl. Comput. Math. 13 (2014), No. 1, 91–106. 112 Journal of Mathematical Physics, Analysis, Geometry, 2016, vol. 12, No. 2
id nasplib_isofts_kiev_ua-123456789-140549
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
issn 1812-9471
language English
last_indexed 2025-12-07T15:22:08Z
publishDate 2016
publisher Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
record_format dspace
spelling Aliev, F.A.
Aliev, N.A.
Guliev, A.P.
2018-07-10T13:53:53Z
2018-07-10T13:53:53Z
2016
Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction / F.A. Aliev, N.A. Aliev, A.P. Guliev // Журнал математической физики, анализа, геометрии. — 2016. — Т. 12, № 2. — С. 101-112. — Бібліогр.: 27 назв. — англ.
1812-9471
DOI: doi.org/10.15407/mag12.02.101
Mathematics Subject Classification 2000: 65M38, 35L02, 35L40, 58J45, 58J90
https://nasplib.isofts.kiev.ua/handle/123456789/140549
We consider the boundary value problem, where the motion of the object is described by the two-dimensional linear system of partial differential equations of hyperbolic type where a discontinuity is at a point within the interval that defines the phase coordinate x. Using the method of series and Laplace transformation in time t (time-frequency method), an analytical solution is found for the determination of debit Q(2l, t) and pressure P(2l, t), which can be effective in the calculation of the coefficient of hydraulic resistance in the lift at oil extraction by gas lift method where l is the well depth. For the case where the boundary functions are of exponential form, the formulas for P(2l, t) and Q(2l, t) depending only on t are obtained. It is shown that at constant boundary functions, these formulas allow us to determine the coefficient of hydraulic resistance in the lift of gas lift wells, which determines the change in the dynamics of pollution.
en
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
Журнал математической физики, анализа, геометрии
Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction
Article
published earlier
spellingShingle Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction
Aliev, F.A.
Aliev, N.A.
Guliev, A.P.
title Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction
title_full Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction
title_fullStr Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction
title_full_unstemmed Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction
title_short Time Frequency Method of Solving One Boundary Value Problem for a Hyperbolic System and Its Application to the Oil Extraction
title_sort time frequency method of solving one boundary value problem for a hyperbolic system and its application to the oil extraction
url https://nasplib.isofts.kiev.ua/handle/123456789/140549
work_keys_str_mv AT alievfa timefrequencymethodofsolvingoneboundaryvalueproblemforahyperbolicsystemanditsapplicationtotheoilextraction
AT alievna timefrequencymethodofsolvingoneboundaryvalueproblemforahyperbolicsystemanditsapplicationtotheoilextraction
AT gulievap timefrequencymethodofsolvingoneboundaryvalueproblemforahyperbolicsystemanditsapplicationtotheoilextraction