Authors

  • Allanazar Ro‘ziyev
    Termiz State University
  • Chori Normurodov
    Termiz State University

DOI:

https://doi.org/10.71337/inlibrary.uz.ijai.97464

Abstract

This study addresses the numerical solution of differential equations that contain a small parameter in front of the highest-order derivative. Such equations frequently arise in various physical modeling scenarios, such as heat conduction, diffusion, and fluid dynamics. These equations exhibit special characteristics, where the presence of a small parameter leads to the formation of boundary layers and sharp changes in the solution. The initial integration method is a powerful approach for obtaining stable and highly accurate results in solving such problems numerically. This research analyzes the application features of the method, its advantages, and its impact on the sensitive zones of the solution. The results demonstrate that the method is an effective tool for reliably solving complex problems involving small parameters.

 

 

background image

INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE

ISSN: 2692-5206, Impact Factor: 12,23

American Academic publishers, volume 05, issue 05,2025

Journal:

https://www.academicpublishers.org/journals/index.php/ijai

page 303

APPLICATION OF THE INITIAL INTEGRATION METHOD TO THE

NUMERICAL SOLUTION OF DIFFERENTIAL EQUATIONS WITH A SMALL

PARAMETER IN FRONT OF THE HIGHEST DERIVATIVE

Ro‘ziyev Allanazar Yuldosh ugli

2nd-year Master's Student in Applied Mathematics, Termiz State University

Normurodov Chori Begaliyevich

Scientific Advisor: PhD Professor

Abstract:

This study addresses the numerical solution of differential equations that contain a

small parameter in front of the highest-order derivative. Such equations frequently arise in

various physical modeling scenarios, such as heat conduction, diffusion, and fluid dynamics.

These equations exhibit special characteristics, where the presence of a small parameter leads

to the formation of boundary layers and sharp changes in the solution. The initial integration

method is a powerful approach for obtaining stable and highly accurate results in solving

such problems numerically. This research analyzes the application features of the method, its

advantages, and its impact on the sensitive zones of the solution. The results demonstrate that

the method is an effective tool for reliably solving complex problems involving small

parameters

.

Keywords:

Initial integration method, small parameter, higher-order differential equation.

Modeling of Physical and Technical Processes Expressed by Differential

EquationsModeling of physical and technical processes using differential equations is one of

the key directions in applied mathematics. Most real-world models include higher-order

derivatives, and in some cases, a small parameter appears in front of these derivatives. For

example:\varepsilon y^{(n)}(x) + a_{n-1}(x)y^{(n-1)}(x) + \ldots + a_0(x)y(x) = f(x), \quad

0 < \varepsilon \ll 1The solutions to such equations differ significantly from those of standard

equations and exhibit singular behavior. That is, boundary layer zones appear in the

solution—regions where the solution changes rapidly—while the rest of the interval remains

smooth. Traditional numerical approaches often fail to provide sufficient accuracy in these

cases, or they incur very high computational costs. Therefore, the method of initial

integration becomes especially relevant. This method, which takes into account the sensitive

structure of such equations, allows for stable and efficient solutions. The essence of the initial

integration method lies in integrating the equation beforehand, thereby reducing higher-order

derivatives to lower-order ones. As a result, the influence of the small parameter is

diminished, and the solution becomes smoother. Particularly, if the differential equation is

expressed using integrals, this method naturally captures the boundary layer zones in the

solution. In our research, we comprehensively cover the classical analysis of high-order

singularly perturbed equations, the algorithmic application of the initial integration method,

the accuracy and stability of numerical solutions, and comparisons with other methods.


background image

INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE

ISSN: 2692-5206, Impact Factor: 12,23

American Academic publishers, volume 05, issue 05,2025

Journal:

https://www.academicpublishers.org/journals/index.php/ijai

page 304

Additionally, we demonstrate the method’s applicability in physical models (such as heat

flow and diffusion models) through examples. Thus, this work highlights the advantages of

new approaches in mathematical modeling and provides methodologically and practically

significant conclusions in solving singular differential equations.

^c • n ? j. A(r, e) + eC(r, e)-^ + b(r, e)x = P(r, E)el6(t-£) (1) look at the question of integrating

a system of second-order Linear Differential Equations dependent on the parameter, where

x(t,£) is an n-dimensional vector R = Et - slowly changing time e < 1-small real parameter;

6(t, e) is a scalar function, i = V - l; A(t, e), B(t, e), C(T, E)—(N x n) matrices p(r, e) — n -

dimensional vector, let the conditions below Be Satisfied: A(R, E), B(T, E), C(T, E)

matrices,and P(T, E)-vector function given r e [0, l] in the range, let e extend to the

converging series by the degrees of parameter: Co Co C(T, = ^ ESCs (t), p(t, = ^ ESPs (t), (2)

s=0 s=0 CC A(x, e) = ^ ESAs (t), b(t,e) = ^ ESBs (T) S=0 s=0 VTE [0, l], deta0(r) * 0(3) DQ

d(T, E) the-derivative of a function is a slow — changing function, i.e. dO Tt = m() (2) series

coefficients As(t), Bs(t), C(t), Ps(t)s = 0.1.2,... and given a function k(r) [0,L] in the study of

an infinitesimally differentiable system in a section, its character is

fs (r) = Bivs-I - ^ ^ ^ Xixjagvs—1—i—j - ^ ¿0¿iajvs-i-j - i=1 i=1 c=0 k=0 i=1 c=0 s k-1s-1—

i s-1 s-1-I - ^¿0a0vs—I-^ ^ ¿iccvk—1—i—j-0-2^ ^ Xiajv'k-1-i-j- i=1 i=0 c=0 i=0 c=0 s-2 s-2

- ^ Civ's—2—I - ^ AIV'sl-2-i s = 1,2,3,... (14) I=1 i=1 based on the condition of the theorem

these roots are simple, then to each of them one p¿r) corresponding to the specific vector (B0-

WiA0) (i = 0 satisfies the relation and is defined in arbitrary scalar product accuracy, which is

nonzero. In this case, the vector A0 added to The Matrix B0 will not exist. In this case (B0 -

WiA0)Z = a0pb (i = l/n) (LS) has no equation solution. We look at the system of equations

(11), (12). Equation (11) has a nonzero solution only and only when ¿2 = -Wi, (i = 1/n), from

this we define 2N different Ä0(r) : Yao (r) = ±i^Wi (r), (i = %n) (16 Then n different v0 (r)

vector functions from(11) are defined: v0(r) = Pi(r), (i = ln) (l7) where(pi(r), A0(r) are the hos

values of the Matrix B0 (r) with respect to The Matrix. (16) munoit plays an important role in

the modeling of many other natural processes. The underlying complexity is the layer zones

generated by the presence of a small parameter and is manifested in the drastically changing

nature of the solution. This makes it difficult for traditional numerical methods to be used, or

they will not have sufficient accuracy as a result. The study initially explored both

theoretically and practically the application of the integralization method to solving such

equations. Based on analyzes and numerical experiments, the following main conclusions

were drawn: 1. Initially, the integrable method takes into account the sensitive zones of small

parametric equations in a natural way. This plays an important role in capturing the drastic

change in solution, especially in the layer zone.2. This method is highly effective in terms of

stability and accuracy. Reduces the uncertainties associated with a small parameter,

smoothing the points of view in the solution.3. In relation to traditional methods, the

computer resource perspective

References:

1. Vazov V. Asymptotic expansions of solutions of ordinary differential equations, Moscow:

Mir, 1973. 464c. 2. Bogolyubov N.N., Mitropolsky Yu.A. The asymptotic method in the

theory of nonlinear oscillations, Moscow: Fizmatgiz, 1963, 410 p. 3. Feshenko S.F.,


background image

INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE

ISSN: 2692-5206, Impact Factor: 12,23

American Academic publishers, volume 05, issue 05,2025

Journal:

https://www.academicpublishers.org/journals/index.php/ijai

page 305

Shkil N. I., Nikolenko L. D. Asymptotic methods in theory of linear differential

equations. Moscow: Nauk dumka, 1966. 252 p. 4. Vasilyeva A. B., Butuzov V. F.,

Asymptotic decomposition of singularly perturbed equations. Moscow: Nauka, 1973,

272 p.

References

Vazov V. Asymptotic expansions of solutions of ordinary differential equations, Moscow: Mir, 1973. 464c. 2. Bogolyubov N.N., Mitropolsky Yu.A. The asymptotic method in the theory of nonlinear oscillations, Moscow: Fizmatgiz, 1963, 410 p. 3. Feshenko S.F., Shkil N. I., Nikolenko L. D. Asymptotic methods in theory of linear differential equations. Moscow: Nauk dumka, 1966. 252 p. 4. Vasilyeva A. B., Butuzov V. F., Asymptotic decomposition of singularly perturbed equations. Moscow: Nauka, 1973, 272 p.