ORIENTAL UNIVERSITETIQidiruvПоискSearch
Oriental universiteti nashriyotiIlmiy jurnallar platformasiПлатформа научных журналовScholarly journals platform

Toʻliq matnПолный текстFull text

Mathematical models of turbulent gas flow with combustion reactions

Aminov Xayotjon Xalimjonovich, Hamdamov Muzaffar Muxiddinovich

Matn PDF dan avtomatik ajratilgan: ekran oʻqiydigan dasturlar, telefon va qidiruv uchun. Rasmiy nusxa — PDF.Текст извлечён из PDF автоматически — для программ чтения с экрана, телефонов и поиска. Официальная версия — PDF.Text was extracted from the PDF automatically for screen readers, phones and search. The PDF remains the version of record.

PDF ni oʻqishЧитать PDFRead the PDF → Nashr kartasiКарточка изданияRecord

MATHEMATICAL MODELS OF TURBULENT GAS FLOW WITH
COMBUSTION REACTIONS

YONISH REAKSIYALARI BILAN TURBULENT GAZ OQIMINING
MATEMATIK MODELLARI

МАТЕМАТИЧЕСКИЕ МОДЕЛИ ТУРБУЛЕНТНОГО ТЕЧЕНИЯ
ГАЗА С РЕАКЦИЯМИ ГОРЕНИЯ

Aminov Xayotjon Xalimjonovich,
Oriental universiteti,

Aniq fanlar kafedrasi mudiri, PhD, dots v.b.
Tel:+998 99 909-72-36

hayotjonaminov@gmail.com

Hamdamov Muzaffar Muxiddinovich,
Oriental universiteti,

Aniq fanlar kafedrasi o‘qituvchisi, PhD, dots.
Tel: +998 97 738-40-62
mmhamdamov@mail.ru

Annotation: In this work, numerical methods are encountered for solving the
problems of propagation and combustion of methane with a finite velocity in a
cocurrent turbulent flow of an oxidizer. In the mathematical modeling of the jet flow,
the full equations of the turbulent boundary composition of a multicomponent reacting
gas are used. The process of one-stage irreversible combustion is described by the
Arrhenius model with constants extracted in the work of C.A. Frolova and recorded.
The flow turbulence is described by Sekundov t  92 equations. Methane is considered
as a fuel, air was supplied with a cocurrent flow. The number of component
conservation equations has been reduced with the use of ordinary and reduced Schwab-
Zeldovich functions. Numerical calculations were carried out in dimensionless Mises
variables. The used implicit finite-difference scheme provided the second order of
approximation accuracy in longitudinal and transverse coordinates. Due to the
nonlinearity of the differential equations of substance conservation, internal (in terms
of speed) and external iterative processes were organized. Some results of the
computational experiment are presented. The calculation results for the temperature
change on the flow axis are compared with the experimental data.

Keywords: mathematical model, turbulent jet, chemical reaction rate, total
enthalpy, Arrhenius law, finite differences, numerical algorithm, computational
experiment.

Annotatsiya: Maqolada oksidlovchining birgalikdagi aylanma oqimda cheklangan
tezlikda metanning tarqalishi va yonishi masalalarini yechishning raqamli usullari
keltirilgan. Reaktiv oqimni matematik modellashtirishda ko‘p komponentli reaksiyaga
kirishuvchi gaz aylanish me’yorining to‘liq tenglamalari qo‘llanildi. Bir bosqichli
qaytarilmas yonish jarayoni Arrenius modelida hamda S. Frolova ishlarida alohida yozib
olinib tasvirlangan. Oqimning aylanma harakati Sekundov tenglamalari bilan tavsiflanadi.
Ma’lumki, metan yonilg‘i hisoblanadi. Jarayonda ishlatiladigan havo to‘g‘ridan-to‘g‘ri
oqim bilan ta’minlandi. Oddiy va kichraytirilgan Shvab-Zeldovich funksiyalaridan
foydalanish orqali saqlanish tenglamalarining komponentlari soni kamayadi. Raqamli
hisob-kitoblar Mises usulida o‘lchamsiz o‘zgaruvchanlikda amalga oshirildi.
Qo‘llaniladigan yashirin farq sxemasi bo‘ylama va ko‘ndalang koordinatalarda
yaqinlashishning ikkinchi darajali aniqligini ta’minladi. Moddaning saqlanishi uchun
differensial tenglamalarning chiziqli bo‘lmaganligi sababli ichki (tezlik bo‘yicha) va
tashqi iteratsiya jarayonlari tashkil etilgan. Hisoblash tajribasining ba’zi natijalari
maqolada keltirilgan. Oqim yo‘nalishi bo‘yicha harorat o‘zgarishini hisoblash natijalari
eksperimental ma’lumotlar bilan taqqoslanadi.

Kalit so‘zlar: matematik model, aylanma reaktiv, kimyoviy reaksiya tezligi,
umumiy entalpiya, Arrenius qonuni, chekli farqlar, sonli algoritm, hisoblash tajribasi.

Аннотация: В данной работе встречаются численные методы решения задач
о распространении и горении метана с конечной скоростью в спутном
турбулентном потоке окислителя. При математическом моделировании струйного
течения используются полные уравнения турбулентного граничного состава
многокомпонентного реагирующего газа. Процесс одностадийного необратимого
горения описывается моделью Аррениуса с константами, извлеченными в работе
С. Фроловой и записанными. Турбулентность потока описывается уравнениями
Секундова. В качестве топлива рассматривался метан, воздух подавался
прямотоком. Число компонент уравнений сохранения сокращено за счет
использования обычных и приведенных функций Шваба-Зельдовича. Численные
расчеты проводились в безразмерных переменных Мизеса. Использованная
неявная разностная схема обеспечила второй порядок точности аппроксимации в
продольных и поперечных координатах. За счет нелинейности дифференциальных
уравнений сохранения вещества были организованы внутренние (по скорости) и
внешние итерационные процессы. Представлены некоторые результаты
вычислительного эксперимента. Результаты расчета изменения температуры на оси
потока сравниваются с экспериментальными данными.

Ключевые слова: математическая модель, турбулентная струя,
скорость химической реакции, полная энтальпия, закон Аррениуса, конечные
разности, численный алгоритм, вычислительный эксперимент.

INTRODUCTION
Combustion is a physical and chemical process, which is a chemical oxidation
reaction, accompanied by the release of heat and a bright glow. In the world, jet streams
of reacting media are widely used in daily life in the production of thermal and
electrical energy, in internal combustion chambers, in chemical lasers, in the
production of mercury and building materials, in rocket engine power plants, and in
many other branches of technology and production. Depending on the purpose of the
organization and the design of gas combustion devices, various problems of facility
management arise. In particular, the gas component in the global energy balance is
about 70%. Despite the constant improvement in the performance of gas combustion
devices, their efficiency is still very low.
The main problems in the mathematical modeling of combustion processes relate
to the description of the turbulence of the flow and the kinetics of the combustion
process. Let us dwell on the description of the turbulence of the flow of reacting media.
The choice of the dissipation rate of turbulent energy as the second independent
variable for describing the properties of turbulence is very convenient, since, on the
one hand, it is possible to obtain a constitutive equation for it under fairly clear physical
assumptions, and, on the other hand, this quantity is included in the constitutive
equation for the energy of turbulent pulsations. The family of these models is called
k  turbulence models, with the help of which individual problems of isothermal jets
and jets of reacting gas mixtures are currently solved. (Баев В., Головичев В.,
Третьякова П., 1984: 2)
The most general analysis of the combustion of unmixed gases was carried out by
Y.B. Zeldovich. (Иевлев В., 1975) One of the first serious attempts to construct a
theory of a turbulent plume belongs to Schwab, who was the first to develop a general
scheme of the process from the standpoint of the theory of free jets. (Шваб А., 1948)
Subsequently, diffusion combustion was studied by the authors and others. (Вулис Л.,
Ярин Л., 1978)
The flame theory, based on the assumption of the generality of the transport
mechanism in turbulent jets and in a burning flame is described in the works of G.
Abramovich. (Абрамович Г., 1984) The results of computational-theoretical and
experimental studies of the aerodynamics of an axisymmetric turbulent plume are
presented, which make it possible to determine the location of the flame front, the
distribution of temperature, concentrations and velocity in the cross-sections of the
flame. A quantitative assessment and a physical explanation are given for such facts
known from experiment as an increase in the degree of turbulence in the flame front,
and the difference from unity in the values of the Pr and Sc numbers in shear flows.
The paper written by L. Artyuks, V. Kashkarov, and M. Tyshkanbayeva describes a

technique and presents some results of a numerical solution to the problem of a

turbulent diffusion plume, taking into account the finite reaction rate. Numerical

integration of the system of equations of the boundary layer was carried out in the

Mises-Mager-Libby plane and using an explicit four-point finite difference scheme.

The momentum equation is linearized. Despite these assumptions, the calculation

results agree satisfactorily with the experimental data. (Артюкс Л., Кашкаров В.,

Тишканбаева М., 1983)

In the article of Bush William and Fink Stantion a mathematical model was

developed that makes it possible to calculate the characteristics of a turbulent diffusion

flame propagating in an isobaric subsonic flow. (Bush William, Fink Stantion, 1981)

It is assumed that a single-stage irreversible reaction occurs between the oxidizer and

the fuel entering the flame front. In contrast to the known models of diffusion

combustion, the assumption of an infinitely high rate of chemical reactions is not used.

In this regard, along with the equation for the average flow velocity and the

concentration of the passive impurity, the equation for the average fuel concentration

is also considered. The rate of a chemical reaction  assumed to be proportional to the

product of the fuel concentration C f , oxidant concentration C0 and average flow

velocity across the flame u  u  which makes it possible to substantially
, C f  C0   ,
z  z 

simplify the integration of the equations. The system of equations is solved as follows:

self-similar solutions of equations are determined that describe the field of average

velocity and passive impurity concentration, each of which is closed. When solving

these equations, the coefficient of turbulent viscosity is assumed to be constant. Within

the framework of the developed model, the position of the flame front is determined,

as well as the distribution of temperature and the concentration of reactants in the

diffusion flame.

The paper published by P. Libby and P. Williams provides an overview of the

latest achievements in the field of research on chemical reactions in turbulent flow.

(Libby P., Williams P., 1981) Two limiting cases are considered - combustion of a

homogeneous mixture and diffusion combustion. It is shown that the assumption of an

infinitely high rate of a chemical reaction makes it possible to substantially simplify

the description of the combustion process. Calculation methods are analyzed that make

it possible to take into account small deviations of the medium from the

thermodynamically equilibrium state. It is emphasized that such calculations are

especially important when predicting the concentration of toxic agents (nitrogen oxide

and soot) in a flame. The influence of heat release on the flow field, in particular, on

the characteristics of turbulence, is also considered. Particular attention is paid to the

effects associated with density fluctuations. Based on the analysis performed, it was

concluded that the models developed to describe flows with a constant density may be
unacceptable for calculating turbulent combustion.

When considering the efficient combustion of gaseous fuel, one should take into
account the results of studies of a swirling jet, which are successfully used in the
national economy as one of the rational ways to intensify the processes of mixing and
combustion.

In the article of H. Haji the spatial integration method was used to solve the
equations describing the axisymmetric incompressible turbulent swirling jet flow. The
k  model with isotropic turbulent viscosity is applied. In addition to the distribution
of average velocities, such turbulence characteristics as Reynolds stress, turbulent
kinetic energy, and dissipation rate are determined. Good agreement with experimental
data is obtained. It is shown that the assumption of isotropic turbulent viscosity is quite
justified. (Haji H., 1986)

The article written by V. Timoshenko provides an overview of the work of the
Institute of Technical Mechanics of the National Academician of Ukraine and the State
Space Agency of Ukraine on gas-phase and heterogeneous reactions and phase
transitions and aspects of their application. (Тимошенко В., 2013)

In the article titled “Моделирование турбулентного реагирующего течения
методом крупного виксрея, совмещенным с методом Монте-Карло для расчета
подсеточных напряжений”, an algorithm and a computer program were developed
for calculating compressible turbulent reacting flows by the combined method of large
eddies with the closure of Subgrid turbulent flows using the filtered joint velocity
probability density function and scalar variables (Large Eddy Simulation / Velocity-
Scalar-Mass-Filtered Density Function, LES/VSMFDF). The complex has been tested
for the problems of an isothermal non-stationary mixing layer, a mixing layer with
chemical reactions, and an air flow around a square cylinder. The results obtained are
in good agreement with data from the DNS, LES/Smagorinski Subgrid model, and
LES/SSM. (Тухватуллина Р., Иванов В., Фролов С., Басара Б., 2018)

An analysis of the scientific literature titled “Горение в сверхзвуковом потоке”
shows that the processes of homogeneous combustion of gases, the main component
of which is methane, have not yet been studied enough. Due to the imperfection of the
fundamental foundations of chemical exothermic reactions under conditions of a
turbulent jet flow, scientific work in this area continues. (Баев В., Головичев В.,
Третьякова П., 1984: 12)

Formulation of the problem
A turbulent jet of combustible gas is considered, which flows out of a round
nozzle with a diameter of 2a and propagates in a cocurrent flow (in a flooded space) of
an oxidizer at a finite chemical velocity.

The velocity distribution in the outlet section of the nozzle and in the co-flow, as
well as the initial (at x=0) distributions of temperature and concentration of fuel and
oxidizer, will be considered given, uniform, and homogeneous. There is also a
tangential discontinuity of indicators at the nozzle boundary.

Mathematical model and solution method
In the approximation of the theory of a turbulent boundary layer, the system of
equations for the turbulent motion of a multicomponent gas in the presence of a
chemical reaction, i.e. combustion, between the components and when the Lewis
number for the components is equal to ( Lei 1, i.e. Pr  Sci ), can be written in cylindrical
coordinates, taking into account the Sekundov t  92 model: (Nora Boukhalfa, 2016:
16)

 u u 1   u 
u      (  t )r ,
x r r r  r 

  ur     r   0,
 x r

 H H 1   H  (1)
u      (  t )r ,
 x r Pr r r  r 

 cn cn 1    1  cn 
u      t  r   n n  1..N ,
x r Sc r r   Re  r 

  t  t 1  t 
u    Pv  Dv   r   c0 t   
x r r r  r 

 1  r r  r   c1  c0 t  t r .

This system is closed by the following dependencies:

N R0  N 1

H  cpT  c2h2*, cp  cpncn, p   T  const, m   mn / cn  ,
n1 m  n1 

 4 2 c2F2 A2 N1  11
t 2
Pv  c1F2  tG1  A 3G 3  G2 

 1t 2  1

  2 t 


c3 t  2  N2 .
 r 

c5 t2G12  u u   c6t  N1dw t,   c7F1t 
Dv  2  c4t     .
aSZ  r r  d 2

F1  N1dw  0.4c8  2 1.3  0.2
, F2  2 .
 t  c8  t,  1.3 1

  t , G1  u , 2u N1  t , N2  N1 .
7 r G2  2 , r r

r

t,  0, ks  0, A1  0.5, A2  4, c0  0.8, c1 1.6, c2  0.1, c3  4.0.

c4  0.35, c5  3.5, c6  2.9, c7  31.5, c8  0.1, cs  0.1, aSZ  0.2.

Here and further u, v – average longitudinal and radial components of the velocity

vector (m s1) ; x, r – longitudinal and radial coordinates m; , Т – density kg / m3 

and absolute temperature (K) of the gas mixture; p – static pressure (Pa); Pr, Scn –

turbulent analogues of the Prandtl and Schmidt numbers; cn – mass concentration n -

th gas component in the mixture (kg/kg); n – mass extinction formation rate n  th gas

component kg / m3s ; cp  cpncn N and cpn – heat capacity of the gas mixture and n - th
n1

component at constant pressure J / kg* K ; hn* – calorific value n - th component J / kg ;

t – kinematic turbulence coefficient m2 / s; H – total enthalpy of gas  j.

The gas mixture is assumed to be perfect, therefore its state satisfies the

Mendeleev-Clapeyron equation:

p  R0T / m , (2)

 N 1
where m  cn / mn  and mn – molar masses of the gas mixture and n - th gas
 n1 

component kg / mol; R0 – universal gas constant 8341.41J / mol K.

The boundary conditions of the problem for the input section x  0 are given as:

F2 at the cut of the pipe

Fo,r  

F1 on satellite treacle

Here F  u, T, cn, t . The value of the total enthalpy for the inlet section is
determined according to the above formula.

At x  0 , no flow conditions are specified on the jet axis:

F  x,0  0, v x,0  0 ,

y

and at the jet boundary, according to the F1 values of the cocurrent.
Methane is considered as a combustible gas. The one-stage methane combustion

kinetics in air is given by the stoichiometric equation

CH4  2O2 CO2  2H2O  h2* .

Indexes 1, 2, 3, 4 and 5 highlight the parameters of the components - oxygen,
methane, carbon dioxide, water vapor and nitrogen, respectively.

According to this entry, the molar masses of the components are

m1  32 g / mol, m2  16 g / mol, m3  44 g / mol, m4  18 g / mol, m5  28 g / mol, 1  5 mol, 2 1 mol,

3  3mol, 4  4 mol, 5  0 mol. Net calorific value of methane – 50.2 MJ/kg.

Then, according to M. Hamdamov, the reaction rate of methane with oxygen has
the form: (Хужаев И., Хамдамов М., 2021)

2   20 A 2 3  Ar2   кг 
2 с1 с2 exp     3 .
10  4.1868  T  м с

At the same time, for the remaining components, the rate of chemical
transformations in the dimensional form is represented as

1  1m1 2 , 3  3m3 1, 4   4m4 1, 5  0 5  0.
 2m2  2m2  2m2

Conservation equation n - th component ( n 1..5) is written as

L(cn )  n , where LCn   u cn   cn  1   cn 
x
 t r  .
x Scnr r  r 

The boundary conditions of the problem for the mass concentration of
components for the inlet section x=0 are given in the following form:

At 0  r  a : cn 0, r  cn 2;

At 0  r  r : cn 0, r  cn 1;

In particular, if methane is 50% by mass in the composition of the combustible
mixture, and oxygen is 50%, then you can take c2 2  0,500, c1 2  0,500 . The

composition of the air is taken as c1 1  0,232, c5 1  0,768 . In this case, the remaining

concentrations of components at the input of the computational domain have zero
values: (Hamdamov M., Mirzoyev A., Buriev E., Tashpulatov N., 2021)

c2 1 c3 1  c4 1  c3 2 c4 2  c5 2  0 .

At x  0 symmetry conditions are set on the jet axis: cn x,0  0, and at the jet

r

boundary - according to the air flow indicators: cn x,r   cn 1.

By virtue of ignoring thermal and barodiffusion, as well as the same values of the
turbulent analogue of the Schmidt number for the components, the convective and
diffusion transfers of the components have the same intensity. In this regard, and taking
into account the combustion kinetics, we write the conservation equations for five
components in the form

L(c2 )  2 , L(c1)  1m1 2 , L(c3 )   3m3 2 , L(c4 )   4m4 2 , L(c5 )  0.
 2m2  2m2  2m2

Using the first of these equations, we reduce the following three equations to a
homogeneous form:

L  c1  1m1 c2   0, L  c3  3m3 c2   0, L  c4   4m4 c2   0.
  2m2    2m2    2m2 

At n 1,3, 4 we introduce the notation cn  cn  nmn c2 (the top sign refers to n 1
 2m2

, and the bottom– n  3, 4 ), write equations in a single form L(cn )  0 and formulate the

boundary conditions for the air flow at the inlet in the form (cn )1  cn 1  nmn c2 1, for

 2m2

fuel flow – (cn )2  cn 2 nmn c2 2 , as well as on the boundary and axis of the jet.

 2m2

If we introduce relatively redundant functions in the form of dependence

Cn  cn  (cn )1 , then for the transformed three equations and the concentration

(cn )2  (cn )1

equation c5  c5 inert gas, the boundary conditions become the same: at the fuel inlet

– Cn   1 , at the air inlet and at the boundary layer boundary – Cn   0 , and on the jet21

axis – Cn x,0  0. Accordingly, these equations become mutually equivalent and can

r

be replaced by a single equation for C . In this regard, instead of five equations of the

diffusion problem, we solve the equation for C without source term and equation for

fuel concentration c2 considering the rate of a chemical reaction.

Reverse transition from relatively excessive concentration C to mass

concentrations of components cn at n 1,3,4 carried out according to the formula

cn  (cn )2  (cn )1 C  (cn )1   nmn c2 ,

m2 2

and at n  5 – according to the formula c5  (c5 )2  (c5 )1 C  (c5 )1 .

The process of combustion of gaseous fuel is carried out at relatively low speeds
relative to the speed of sound. Therefore, the kinetic energy of the gas can be neglected.
Then, a substitution similar to that introduced for the concentration can also be used
for the total enthalpy, which leads to the normalization of the total enthalpy:

H  H  H1  C .

H2  H1

Here H1  cp 1T1  c2 1 h2*, H2  cp 2 T2  c2 2 h2* .

In this case, the value of the temperature of the mixture is calculated by this
formula:

H c2h2*  N

T   / cp  H1  H2  H1 H  c h*  /  c pncn .

22
  n1

The value of the density of the gas mixture necessary for further calculations is

found by this formula:

 Np .

R0T cn / mn
n1

We passed to dimensionless variables and Mises variables. We passed to
dimensionless variables using the characteristic quantities a, u2, 2 :

x  x , r  r , u  u ,    , Re  u2a , G G1a G2a2 N1 N2a 1  , G2  , N1  , N2  .
a a u2 2  u2 u2 u2 u2

For example, the equation for the conservation and transfer of fuel takes the form:

c2 c2 1    1  c2  A2 2 3 3  Ar2 
2u2u  2u2v  2au2  t ar   20 2 с1 с2 exp  ,
ax ar Scar ar   Re  ar  10  4.1868  T

where dimension 2 translated into SI.

When passing to dimensionless variables, both sides of the equation are divided
by осuoc . Therefore, the source term takes the form:

a

a a Aос 2 3 3  Ar2  A aос 2 3 2  Ar2 
2   2 с1 с2 exp      20 с1 с2 exp   .
осuoc 20 2
осuoc 10  4.1868  T  10  4.1868 uoc  T

As a result, we get

c2 c2 1    1  c2  2 3  Ar2 
u  v    t  r   Ar1с1 с2 exp  .
x r Sc r r   Re  r   T

Here we introduced the notation:

Ar1  A aос2 , Ar2  3.00  4.1868104 К.

20 2 8341.41
10  4.1868 uoc

The turbulence equation is written in full form

 t  t 1 t  1   t
u    Pv  Dv   r   c0t    r   c1  c0 t  ;
x r r r  r  r r r

 t  t  t au2  t au2 2  t t 
u    2uu2  2 u2  2u2  u   ;
x r xa ra  x r 

2u2 2  4 2  1 1 1   2 t 
Pv   c1F2  tG1  A 3G 3  c2F2 A2 N1  t 2G 2  c3 t 2  N2  .
   
1t 2 1
   Re   r 

  1 
 c6t  N1dw t,   c7 F1 t  
2  c5 t 2 2  u u   Re 
Dv  2u2  2 G1  c4 t     .
 asz  r r  d 2

 
 

2   t  t  1   u2a   tu2a 
2u2  u     2u2  Pv  Dv   2  2 ra   c0u2a t 2 

 x r  ra r   Re  ra 

1   u2a   u2a  t
 2  ra2    c1  c0 u2a t   .
ra r   Re   a r

Otherwise, taking into account t  t , in each term a dimensionless factor is

au2

formed  2 u 2 :
2

2   t t  2 1  1  t  1    1   t 
2u2  u     2u2  Pv  Dv   r   c0 t     r     c1  c0  t   .
 x r   r r   Re  r  r r   Re   r 

Reducing by 2u22 , we get

 t  t 1  1  t  1    1  t 
u    Pv  Dv   r   c0t     r     c1  c0 t  ;
x r r r   Re  r  r r   Re  r 

Thus, the differential equations take the form:

 u u    1  u 
u      t  r  ,
 x r r r   Re  r 

  ur      r   0,
 x r

 H 1    1  H 
 H
u      t  r  ,
 x r Pr r r   Re  r 

(3)
 c2 c2 1    1  c2 
u       t  r   Ar1c1 c2 exp( Ar2 / T ),23

 x r Scr r   Re  r 

  t  t 1  1   t 
u    Pv  Dv    r   c0t   
 x r r r   Re  r 

 1   1    t
   r     c1  c0 t   .
 r r   Re   r

In closing relations, we pass to dimensionless variables as follows:

p   R0 T  2 R0 T2.
m m2

From here we find the expression for density:   mT2 .

m2 T

We transform the conditions for the required ones.

At x  0 and 0  r 1 we have : u  1, H  1, cn  cn 2 .

At x  0 , r  1 and at x  0, r  r : u  u1 , H  0, cn  cn 1 .

u2

On the jet axis at x  0 : u  0,   0, H  0, cn  0.
r r r

A feature of the problem under consideration is that the flow propagates along the

axis х from left to right and there are no boundary conditions in the right part of the

flow region, since they do not affect the propagation processes. Accordingly, it is

expedient to describe the process in von Mises coordinates and introduce the stream

function  according to the relations:

ur   , r   
r x

(hereinafter, we omit the sign of non-dimensionality in equations and relations). The
formally introduced second coordinate is   x.

The derivatives in the equations are replaced according to the dependencies:

               r  ,
x  x  x   x   

           ur  .
r  r  r  r  

In this case, the dimensionless equation of conservation of the total mass is
uniquely satisfied.

The left sides of the equations are transformed into

u x  v r  u    r v         v r u   u  .    

The components of the right sides of the equations are transformed as

1  1r    1  r  
r   u   t r u  

r r  r  r     Re    

1   2  1  2 u   u    2ur2 
 u   t r  l t  .
    Re           

The value of the transverse coordinate required for the calculation is found by the

formula r2   2 d , what we will return to in the process of discretization of

0 u
equations and conditions. The differential equations for the conservation of substances
take the form:

 u  1 u   2u r2  l  t  u  ,
      

    2u r2  l  t  H  ,
H  1
  Pr     
(4)
 c2 1    2u r 2  l  t  c2  c12c2 2
    Ar1 exp( Ar2 / T ),
  Sc      u

  t Pv Dv 1    2ur 2  1   t 
    c0 t   
  u u      Re   

1  r  1   ur  t 
     c1  c0  t  
 .
     Re     


T  H1   H2  H1  H  c2h2* ,   mT2 .
cp (m)2T

G1  ur u , G2   2u2r2 2u 2 , N1  ur t , N2  ur N1 .

   2    

 4 2  1 1 1   2u2r 2 2 t 
Pv  c1F2  tG1  A 3G 3  c2F2 A2 N1  t 2G 2  c3 t 2  N2 .
   
1t 2 1
   Re    2 

 1
 c6t  N1dw t,   c7F1 t 
c5t2 2 urc4t  u u   Re 
Dv  2 G1     .
asz      d 2

At the entrance   0 the following conditions are imposed on the computational
domain:

At 0    1: u  1, H  1, c2  c2 2 ,t  0.007.

At 1    : u  u1, H  0, c2  c2 1 , t  0.005.

At   0 we have at   0 : u  0, H  0, c2  0, t  0.
   

At    : u  u1, H  0, c2  c2 1 , t  0.005.

Transition to discrete variables
We write the equations in a single form:

U 1   U 
  Kh   fU .
 U    

Here K   2u r2 l t  ,

U  u : U  1, fU  0;

U  H : H  Pr  Sc, fH  0;

U  c2 : c2  Sc, fc2  Ar1 c12c2 2 exp(Ar2 / T ).

u

On the right side of the equations, r2 – the square of the dimensionless radial
coordinate. We calculate it with the second order of accuracy

    s 2 s2  j2  j1 2 s1 s1 .
ri, j  ri, j1  2 s1 s1
i, j1ui, j1  i, j ui, j

This is the formula for i - section; s – iteration sequence number for i - th section.
For i 1- th section. r2 calculated according to the data of the previous section. To reduce
the entries and the amount of calculations, we first calculate

   i, j 2 ui, j ri, j 2
Ki, j  j  l   t i, j ,

where ri, j  ri, j 2 / h2 .

At j  0 accepted Ki,0  0 and ri,02  0.

We approximate the equations for a fictitious node i  0.5, j with the second order

of accuracy in both coordinates:

        Ui, js1 Ui1, j  2 1  Ki, j1 s  Ki, js
h 4U jh Ui, j1 s1  Ui, js1  Ki, js  Ki, j1 s Ui, js1  Ui, j1 s1 

  Ki1, j1  Ki1, j  Ui1, j1  Ui1, j    Ki1, j  Ki1, j1  Ui1, j  U     fU i, j s   fU i1, j .

i 1, j 1 2

Multiply both sides of the equation by 4U jh2 and we give similar terms with the
selection of terms with elements of the previous time section:

      Ki, j1 s  Ki, js Ui, j1 s1  Ki, j1 s  2Ki, js  Ki, j1 s  4U j /  Ui, js1  Ki, js  Ki, j1 s Ui, j1 s1 
    2U jh2  fU i, j s  Ki1, j1  Ki1, j Ui1, j1  Ki1, j1  2Ki1, j  Ki1, j1  4U j /  Ui1, j 

   Ki1, j  Ki1, j1 Ui1, j1  2U jh2  fU i1, j ,

Where   h / h2 . Let us introduce the notation:

a jU   Ki, j1 s  Ki, js , c jU   Ki, js  Ki, j1 s ,

bjU   a jU   c jU   4U j /  , d jU   2U jh2  fU i, j s ,

    g jU    Ki1, j1  Ki1, j Ui1, j1  Ki1, j  Ki1, j1 Ui1, j1 
  Ki1, j1  2Ki1, j  Ki1, j1  4U j /  Ui1, j  2U jh2  fU i1, j .

As a result, the finite difference equation takes the form:

U  s1 U  s1 U  s1 U  U 
a j Ui, j1  bj Ui, j  c j Ui, j1  d j  g j .

If we assume that the values of the coefficients are known  j1 (U ) and  j1 (U) in

Ui,s j 1   (U ) s j   (U ) , then to calculate the values of the sweep coefficients we obtain the
j 1
j1Ui,

formulas:

 j(U )  a j(U ) (U ) c j(U )  j1 (U )  d jU   g jU 
(U ) (U ) (U ) ,  j  (U ) (U ) (U ) .
bj   j1 c j bj   j1 c j

Selection g jU requires additional computer memory, but at the same time, in the

process of successive approximation, it significantly reduces the amount (time) of
calculations and makes it possible to use a larger step in  .

The implementation of boundary conditions is carried out in the usual mode.

 t Pv Dv 1    2ur2  1  t  1   r  1   ur t 
    c0 t        c1  c0 t   ;
 u u      Re         Re     

In the second group, the approximation of equations and the sweep process are
similar to those used in the first group.

The calculation for a fixed section consisted of three stages. In the first stage, the
components of the finite difference equations were calculated with the exponents of
the previous section: ri1, j , Ki1, j , g j... and organized a zero approximation. The other two

stages represented iterative processes in terms of speed and other indicators. The stage

of the iterative process in terms of speed includes the calculations ri, j , Ki, j , g u , running

j

process and checking the fulfillment of the approximation condition
maxi0,1,...Ny ui, jm  ui, jm1  u iterative process by (m is the iteration number). If the condition

is not met, then the iterative process in speed continues. If the condition is met, then
the transition to the second stage is carried out, where the turbulence, fuel concentration
and total enthalpy indicators for the next approximation are determined by the sweep
method and the approximation in terms of total enthalpy and density is checked. If the
condition for the total enthalpy and density is not met, then the calculations for the
speed are resumed (the second stage) and the third stage is repeated.

RESULTS
Calculations were carried out in accordance with the above. The mass
composition of air was given as c11  0.232, c5 1  0.768. The main flow rate was 61 m/s,
and cocurrent – 18.3 m/s. Air temperature – T1  300K , and combustible mixture –
T2  300 K . Step of numerical integration over x  was equal to 0.001, on  – 0.02. When

passing to the dimensionless coordinate r data were presented with a constant step of
0.02.

On fig. 1 shows isotherms in the field of intense heat and mass transfer, which
were obtained with a  0.05 м .

Fig.1. Isotherms in the field of intense heat and mass transfer at a  0.05 м

Near the nozzle, the highest temperature is reached at a certain distance from the
axis. Nonmonotonic behavior of isotherms up to x / a  60 shows that the line of the
highest temperature in the cross section first moves away from the jet axis. Reaching
1690 К at r / a  5 , x / a  60 , it moves towards the flow axis and after x / a  84 the
maximum temperature in the section coincides with the axial temperature. Further, the
axial temperature increases and at x / a  698 reaches 2052 K . After that, the highest

temperature in the cross section, i.e. axial temperature decreases. At the end of the
conditional section Hось  0.01 on the jet axis the temperature is  761K . (fig.1.).

1

Pасчет
0.8

0.6

0.4

0.2

0

-4 1 6 11 x/a16

Fig.2. Comparison of the calculation of the axial excess temperature with the
experiment.

Calculations have shown that in the initial section the maximum temperature will
not be on the flow axis. At a  0.001 м the largest radial coordinate of the highest
temperature (690 K) in the section r / a  2.6 observed at x / a 14. The combination of
the highest and axial temperatures (at 867 K) occurs at x / a  30 . The highest
temperature in the calculation area is 1134 K, and the axial temperature at the end of
the conditional flame (at Hось  0.01 ) equal to 756 K.

2300 Tos, Tmax
1900

1500

1100

700 1 2 3 4 5 6

300 100 200 300 400 500 600 x/a 700
0

Fig. 3. Curves of the largest (dotted) and axial (solid) temperature values
along the dimensionless length of the torch at various radii of the fuel nozzle: 1, 2

– 0.05 m; 3, 4 – 0.01 m; 5, 6 – 0.001 m

Note that each calculation was repeated in the Ansys Fluent 6.3 program code and
the results of two calculations were compared. Ansys Fluent is a powerful tool for
streamlining the design and engineering process in the field of computational fluid and
gas dynamics. In Fluent, it is possible to simulate the combustion process in furnaces,
taking into account turbulence, heat and mass transfer, chemical reaction, and flow
swirl. A typical field of mass concentrations of components was obtained using two
programs. On fig. 4 they were compared with x / a  5 .

Fig. 4. Graphs of mass concentrations of components at x / a  5
If at x / a 1 the initial jet is expanded by about one and a half times, then at
x / a  5 (fig. 4) – up to three times. The highest concentrations of combustion products
are formed first at the jet boundary, and then they move away from the flow axis.
Molecular nitrogen, which makes up the bulk of the air, gradually penetrates towards
the axis of the jet.
The concentration of combustible gas decreases in the graphs, both due to the
expansion of the distribution area, and due to the reaction with oxygen.

Fig. 5. Entropy

Fig. 6. Turbulent kinetic
energy

CONCLUSIONS
Methods for introducing the Schwab-Zeldovich functions and relatively excess
enthalpy are shown for the case of turbulent combustion of methane in a cocurrent air
flow with a finite velocity flowing out of the nozzle. A numerical method based on
finite-difference schemes is proposed for solving the problem of the propagation of an
axisymmetric fuel jet in a cocurrent air flow in von Mises variables using the Sekundov
t  92 turbulence model.
For the numerical solution of the problem, an approximation scheme of the second
order of accuracy in both coordinates was used, which gives a more accurate picture of
the object and the physical essence of the combustion process. An algorithm and a
calculation program with internal and external iterations for the longitudinal velocity
and other indicators of heat and mass transfer processes have been developed. Based
on the results of the computational experiment and comparison with field experiments,
it was revealed that with an increase in the concentration of fuel in the composition of
the main flow mixture, expansion and elongation of the flame front are observed.

REFERENCES:
1. Bayev, V., Golovichev, V., Tretyakova, P. (1984) Goreniye v sverxzvukovom
potoke. – Novosibirsk: Nauka. (In Russian)
2. Kondratyev, L., Medvedev, S. (1983) Primeneniye dvuxparametricheskoy
modeli turbulentnosti dlya rascheta struynix yavleniy // Fiziko-ximicheskiye protsessi
v energeticheskix ustanovkax. – Minsk. – C. 16-19. (In Russian)
3. Ievlev, V.M. (1975) Turbulentniye dvijeniya visokotemperaturnix ploshadey
Sred. – Moskva: Nauka. (In Russian)

4. Shvab, A.B. (1948) Svyaz mejdu temperaturnimi i skorostnimi polyami
gazovogo fakela // V sb. Issledovaniye protsessov goreniya naturalnogo topliva. Pod
krasniy. G.F. Knorre. – Gosenergoizdat. (In Russian)

5. Vulis, L., Yarin, L. (1978) Aerodinamika fakela. – Leningrad: Energiya. (In
Russian)

6. Abramovich G.N. (1984) Teoriya turbulentnix struy. – Moskva. (In Russian)
7. Artyuks, L., Kashkarov, V., Tishkanbayeva, M. (1983) Ob effekte
laminarizatsii diffuzionnogo turbulentnogo fakela // Chislenniye metodi resheniya
zadach matematicheskoy fiziki. – Alma-Ata. – C. 28-35. (In Russian)
8. Bush William, Fink Stantion. (1981) On Diffusion Flames in Turbulent
Comeumption in a Circular Fuel Shear Flows: Modelling Reactant Comsumption in a
Circular Fuel Jets // AIAA Journal, 19, № 3. – P. 372–379.
9. Libby, P., Williams, P. (1981) Omplication of Recent heoretical Studies in
Turbulent Combustion // AIAA J., 19, № 3. – P. 261–274.
10. Haji H. (1986) The Prediction of Turbulent Swirling Jet Flow // Int. J. Heat
and Mass Transfer, 29, R 2. – P. 169–182.
11. Timoshenko, V.I. (2013) Matematicheskoye modelirovaniye turbulentnix
ximicheski-reagiruyushix techeniy gazoviks i dvuxfaznix smesey v struyaks i kanalov
// Texnicheskaya Meksika. – № 4. – S. 123-136. (In Russian)
12. Tuxvatullina, R., Ivanov, V., Frolov, S., Basara, B. (2018) Modelirovaniye
turbulentnogo reagiruyushyego techeniya metodom krupnogo viksreya,
sovmeshyennim s metodom Monte-Karlo dlya rascheta podsetochnix napryajeniy //
Goreniye i vzriv. T. 11. – №2. – S. 66-75. (In Russian)
13. Nora Boukhalfa. (2016) Chemical Kinetic Modeling of Methane Combustion
// Procedia Engineering. Vol. 148. – P. 1130-1136.
14. Kozubkova, M., Krutil, Ya., Nevrliy, V. (2014) Eksperimentalnoye
issledovaniye i chislennoye modelirovaniye goreniya metana v oblastix so slojnoy
geometriyey // Fizika goreniya i vzriva, t. 13. 50, № 4. – C. 8-14. (In Russian)
15. Ivanova, A., Karno, A. (2016) Predeli vosplameneniya dlya okisleniya metana
v paraks vodi // Institut problem ximicheskoy fiziki beg. VIP. II Chernogolovka. – S.
125-127. (In Russian)
16. Khujaev, I., Hamdamov, M., Abdusattorov, S. (2018) A numerical method
for solving the problem of an axisymmetric turbulent jet of propane-butane mixture
under diffusion combustion // Problems of Computational and Applied Mathematics.
– No. 3 (15). – P. 61–78.
17. Khujaev, I., Hamdamov, M. (2020) Numerical Method for Calculating
Axisymmetric Turbulent Jets of Reacting Gases During Diffusion Combustion // J.
Adv. Research in Dynamic & Control Systems, Vol. 12, 07-Special Issue.

18. Xujayev, I., Xamdamov, M. (2021) Rasprostraneniye osesimmetrichnoy
turbulentnoy strui metana v sputnikovom potoke vozduxa pri gorenii s konechnoy
skorostyu. ISSN 1812-3368. Vestnik MGTU im. N.E. Baumana. Ser. Yestestvenniye
nauki. № 5. DOI: 10.18698/1812-3368-2021-5-89-108. (In Russian)

19. Hamdamov, M., Mirzoyev, A., Buriev, E., Tashpulatov, N. (2021) Simulation
of non-isothermal free turbulent gas jets in the process of energy exchange // E3S Web
of Conferences 264, 01017.

https://doi.org/10.1051/e3sconf/202126401017. CONMECHYDRO.

FOYDALANILGAN ADABIYOTLAR RO‘YXATI:
1. Баев, В., Головичев, В., Третьякова, П. (1984) Горение в сверхзвуковом
потоке. – Новосибирск: Наука.
2. Кондратьев, Л., Медведев, С. (1983) Применение двухпараметрической
модели турбулентности для расчета струйных явлений // Физико-химические
процессы в энергетических установках. – Минск. – C. 16-19.
3. Иевлев, В.М. (1975) Турбулентные движения высокотемпературных
площадей Сред. – Москва: Наука.
4. Шваб, А.Б. (1948) Связь между температурными и скоростными полями
газового факела // В сб. Исследование процессов горения натурального топлива.
Под красный. Г.Ф.Кнорре. – Госэнергоиздат.
5. Вулис, Л., Ярин, Л. (1978) Аэродинамика факела. – Ленинград: Энергия.
6. Абрамович Г.Н. (1984) Теория турбулентных струй. – Москва.
7. Артюкс, Л., Кашкаров, В., Тишканбаева, М. (1983) Об эффекте
ламинаризации диффузионного турбулентного факела // Численные методы
решения задач математической физики. – Алма-Ата. – C. 28-35.
8. Bush William, Fink Stantion. (1981) On Diffusion Flames in Turbulent
Comeumption in a Circular Fuel Shear Flows: Modelling Reactant Comsumption in a
Circular Fuel Jets // AIAA Journal, 19, № 3. – P. 372–379.
9. Libby, P., Williams, P. (1981) Omplication of Recent heoretical Studies in
Turbulent Combustion // AIAA J., 19, № 3. – P. 261–274.
10. Haji H. (1986) The Prediction of Turbulent Swirling Jet Flow // Int. J. Heat
and Mass Transfer, 29, R 2. – P. 169–182.
11. Тимошенко, В.И. (2013) Математическое моделирование турбулентных
химически-реагирующих течений газовикс и двухфазных смесей в струякс и
каналов // Техническая Мексика. – № 4. – С. 123-136.
12. Тухватуллина, Р., Иванов, В., Фролов, С., Басара, Б. (2018)
Моделирование турбулентного реагирующего течения методом крупного
виксрея, совмещенным с методом Монте-Карло для расчета подсеточных
напряжений // Горение и взрыв. Т. 11. – №2. – С. 66-75.

13. Nora Boukhalfa. (2016) Chemical Kinetic Modeling of Methane Combustion
// Procedia Engineering. Vol. 148. – P. 1130-1136.

14. Козубкова, М., Крутил, Я., Неврлий, В. (2014) Экспериментальное
исследование и численное моделирование горения метана в областиx со сложной
геометрией // Физика горения и взрыва, т. 13. 50, № 4. – C. 8-14.

15. Иванова, А., Карно, А. (2016) Предели воспламенения для окисления
метана в паракс воды // Институт проблем химической физики бег. VIP. II
Черноголовка. – С. 125-127.

16. Khujaev, I., Hamdamov, M., Abdusattorov, S. (2018) A numerical method for
solving the problem of an axisymmetric turbulent jet of propane-butane mixture under
diffusion combustion // Problems of Computational and Applied Mathematics. – No.
3 (15). – P. 61–78.

17. Khujaev, I., Hamdamov, M. (2020) Numerical Method for Calculating
Axisymmetric Turbulent Jets of Reacting Gases During Diffusion Combustion // J.
Adv. Research in Dynamic & Control Systems, Vol. 12, 07-Special Issue.

18. Хужаев, И., Хамдамов, М. (2021) Распространение осесимметричной
турбулентной струи метана в спутниковом потоке воздуха при горении с
конечной скоростью. ISSN 1812-3368. Вестник МГТУ им. Н.Э. Баумана. Сер.
Естественные науки. № 5. DOI: 10.18698/1812-3368-2021-5-89-108.

19. Hamdamov, M., Mirzoyev, A., Buriev, E., Tashpulatov, N. (2021) Simulation
of non-isothermal free turbulent gas jets in the process of energy exchange // E3S Web
of Conferences 264, 01017.
https://doi.org/10.1051/e3sconf/202126401017 CONMECHYDRO.