MATEC Web of Conferences 132, 04020 (2017)

DOI: 10.1051/matecconf/201713204020
DTS-2017

The calculation problem of thermodynamic processes

in a steam turbine
Alexander Chistyakov1,*, Alla Nikitina2 and Elena Protsenko3
1Don State Technical University, Rostov, Russia
2Southern Federal University, Taganrog, Russia
3Taganrog Institute of A.P. Chekhov (branch) of the Rostov State University of Economics, Taganrog, Russia

Abstract. This paper covers the calculation of aerodynamic processes in a steam turbine using the modern
information technologies and computational methods, that are contributed to the increasing the accuracy of
the calculations. The practical significance is the development and application of the model of aerodynamic
processes in a steam turbine, and determination of limits and prospects with using of proposed model. The
aerodynamic processes in a turbine are characterized by the air and heat flow non-uniformity, significantly
affect the reliability and efficiency of the turbine. The calculation was performed taking into account the
complex geometry of the turbine, and can be applied to any turbine of similar design with minor changes.

1 Introduction 2 Problem statement

At present, the question about the optimization problem The main equations of air dynamics are:
of installation and exploitation of steam turbines is the – Navier-Stokes motion equations
actual. With the development of modern technology and vx v v v
 vx x  v y x  v z x 
industry needs to the exploitation of turbines, t x y z
requirements are the more stringent associated with the
1 P  2v  2v  2v 
reliability and efficiency of their operation. A large     2x  2x  2x   fx ,
number of existing turbines is practically close to the  x  x y z 
elaboration its resource. Therefore, the introduction of v y v y v y v y
more modern units is required. The fundamentals of the  vx  vy  vz 
t x y z
theory of heat transfer and analysis results of transfer
processes are required to assess the reliability and 1 P   2vy  2vy  2vy 

  2  2  2   f y ,
efficiency of the facility. Therefore, these data should be  y  x y z
 
taken into account in designing of steam turbines.
vz v v v
Thermal systems modelling include the problems of  vx z  v y z  vz z 
optimal control of thermal modes, due to which we can t x y z
(1)
choose the best from different implementations. The 1 P   2v  2v  2v 
optimization of thermal modes is reduced to the solution      2z  2z  2z   f z ,
 z  x y z 
of heat conduction problem. Mathematical modelling of
– continuity equation
thermal processes in technogenic systems is relevant at
the present. Due to it, we can check the correctness of t    u x    v  y    w  z  0 . (2)
engineering ideas and correct errors at the stage of – state equation
designing by the simple, low-cost means. The

mathematical model is represented by the scheme P RT , (3)
«model – algorithm – program», and must contain the М
structure, characteristic features of the process and where V  vx , v y , vz  are velocity vector components;
should be described by the equation system or functional p is a pressure; ρ is a density; М is the molar mass;
relations. After the designing, it is necessary to R is the universal gas constant;  is the turbulent
determine the real values of temperature at significant exchange coefficient; T is the temperature,  is the
points of the steam facility and analyze the compliance сoefficient describing the deviation of pressure
with required values. superheated steam from the ideal gas.

*
Corresponding author: cheese_05@mail.ru
© The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution
License 4.0 (http://creativecommons.org/licenses/by/4.0/).
MATEC Web of Conferences 132, 04020 (2017) DOI: 10.1051/matecconf/201713204020
DTS-2017

We assume that the air environment is initially in  1    rvr  1    v     vz 

quiescence. Therefore, the initial conditions have the    0,
t r r r  z
form:
The equation (5) in the case of axial symmetry has
u  0, w  0, P  Pa , (3)
the form:
where Pa is the initial pressure.  1    rvr     vz 
The system of equations (1) and (2) are considered   0. (6)
t r r z
with the following boundary conditions:
– on an impervious boundary
v v y v 3 The state equation
w x   x,b (t ), w   y ,b (t ), w z   z ,b (t ),
n n n The main equation for the superheated steam is the
Vn  0, Pn  0 ; equation for the specific Gibbs energy, consisting of two
– on the lateral permeable borders parts – relating to ideallocation as  0 and describing real
vx v y v P
 0,  0, z  0, 0; component  r :
n n n n
g (p, T) / RT   ( , )   0 ( , )   r ( , ) (7)
– on the source
9
P  0  ln    ni0 J i0
(8)
v x  U , v y  V , vz  W ,  0;
n i 1
43
where U , V , W are components of velocity vector on the
 n    0,5 Ii Ji
r i (9)
source;  x ,  z are components of tangential shear stress. i 1

The system of Navier-Stokes’ equations in the where   p / p * and   T * /T , p*  1MPa ,

cylindrical coordinate system has the form: T *  540 К .
vr v v v v 2 v 1 P From the relation for the specific volume
 vr r   r    vz r   
t r r  r z  r   
vP / RT     ,     
   1   rvr   1  2 vr 2 v  2 vr    
    2  2    fr , follows the next:
 2
r  z 2 
 r  r r  r 
 43
Ji 
1   ni I i  p / p * T * /T  0,5    RT .
Ii
v v v v v v v 1 P PM 
 vr       r  vz      i 1 
t r r  r z r  
   1   rv   1  2 v 2 vr  2 v 
    
 r r r  r 2  2 r 2  z 2  
 f , 4 Turbulence model
   
vz v v v v 1 P We used the model of Abramovich-Sekundow, which
 vr z   z  vz z    takes into account very important factors such as the
t r r  z  z
(4) presence of hard walls, the prehistory of the flow,
 1   vz  1  2 vz  2 vz  convective and diffusion turbulent pulsation transfer:
  r  2  2   fz .
 r r  r  r 
2
z  vтурб 3 vтурб 3
  vтурб 
  vi    vмол  kvтурб  
The system of equations (4) in the case of axial t i 1 xi i 1 xi  xi 
symmetry has the form:
 vтурб  vтурб  vмол   vтурб 
vr v v 2 v 1 P vтурб f   D   S, S  2 ,
 vr r    vz r   
t r r z  r  8vмол  Lmin

   1   rvr    2 vr  3
vi  vi v j 
   
 r r r  z 2  r
f , D  x   ,
xi 
(10)
j  x j
    i 1 j

v v v v v z 2  1, 47 z  0, 2
 vr    r  vz   f ( z )  0, 2 ,
t r r z z 2  1, 47 z  1
   1   rv    v 
2
where k  2,   50,   0, 06 , Lmin is the minimum
    f ,
 r r r  z 2   distance to a solid wall; vмол is the molecular
   
vz v v 1 P viscosity, vтурб is the turbulent viscosity.
 vr z  vz z   
t r z  z Turbulence model (10) is considered with the
(5)
 1   vz   2 vz  boundary condition  v   t , x, у, z  0.
  r  2   fz .
турб n  x , у , z Г
 r r  r  z  Complete model (10) with initial condition:
The continuity equation (2)
coordinate system has the form:
in the cylindrical  v   t , x, у , z   v
турб 0.

2
MATEC Web of Conferences 132, 04020 (2017) DOI: 10.1051/matecconf/201713204020
DTS-2017

The next equation is used to calculate values of The function 1 ( ) is defined:

dynamic viscosity vмол :
1 ( )  b0  b1  b2 exp B1 (  B2 ) 2  ,
vмол v0    v1  ,    v* , (11)
where b0 = –0,39707; b1 = 0,4; b2 = 1,06; B1 = –0,171;
where v* = 55,071·10-6 Pa·s;   T / T * ; T is the
B2 = 2,392, and the function 2 ( ,  ) is defined:
temperature, [К]; T * = 647,226 K;    /  * ;  is a
density, [kg/m3];  * =317,763 kg/m3; v0  v0 / v* ; v0 is  d1  9/5
 10  d2   exp C1 (1   ) 
14/5
2 ( ,  ) 
the dynamic viscosity of water vapor in the limit of zero  
density  
 Q  1 Q  3/2 C3 
 3 
1 d 3 S Q exp   (1   )  d4 exp  C2  5  ,
v0     0,5   H i  i  , (12)  1  Q     
 i 0 
where Q and S are functions of the magnitude
where H 0  1 , H1  0,978197 , H 2  0,579829 ,
    1  С4 :
H 3  0, 202354 .
 5 6 j  1/  for   1;
v1  ,   exp    H i , j 1/   1   1  . (13) 2 C5 /  0,6 ; S  
Q 
i
 0,6
 i 0 j 0  С6 /  for   1.

The coefficients di and Ci have the next values:

5 The heat conductivity problem d 1 = 0,0701309; d 2 = 0,0118520; d 3 = 0,00169937;
Thermal processes in turbine G are described by the d 4 = –1,0200; C1 = 0,642857;
heat conduction equation: C2 = –4,11717; C3 = –6,17937; C4 = 0,00308976;
T T T T   T  C5 = 0,0822994; C6 = 10,0932.
c  vx  vy  vz   
t x y z x  x  The equation (14) is applicable with the following
(14) values of the temperature and pressure: р  100 МPа
  T    T 
       qv  P  , for 0  T  500 ºС; р  70 МPа for 500  T  650 ºС;
y  y  z  z 
which can be written in the case of axial symmetry р  40 МPа for 650  T  800 ºС.
The error values in the liquid field at the
as: crTt rvrTr  rvzTz  r  Tzz   rTrr  rqv  P . (15) temperatures 25-200 ºС and pressures up to 5 МPа is
In the system (14), (15): T is the temperature, [ K ] ; equaled to 1,5 % at higher temperatures up to 300ºС –
 is the coefficient of conductivity of water;  is the 2 %. For the water steam at temperatures up to 550 ºС
under pressure 0,1 МPа error is equal to 1,5 %, under
density of the metal; c is the coefficient of the metal
pressures up to 40 МPа – 3 %. The equation (18), in
conductivity; r is the polar radius; qv  P  is the source contrast to the theoretical conclusions, determines not
function. We will consider equation (15) with boundary infinite, but finite conductivity value at the critical point,
conditions of third order: which does not allow to estimate the error values near its
Tn( x, r , t )   nT   n , (16) critical point.

where n is the normal vector to the surface G .

We will consider equation (15) with the initial 6 Splitting schemes into physical
condition: processes
T ( x, r , 0) T0 ( x, r ), ( x, r )  G.
 (17) The initial aerodynamic model are divided into three
The next equation is used to calculate the coefficient subtasks according to the of the pressure amendment
of conductivity of the water and steam in the method [1–4]. The first subtask is representative by the
international practice: diffusion-convection problem on the basis of which the
velocity field are calculated at the intermediate time step:
  0 ( )  1 ( )  2 ( ,  ) (18)
vr  vr v v2 v    1   rvr   2vr 
where λ is the conductivity, W/(m·К); τ = Т/Т*; Т is the   vz r   
 vr r   f ,
 r r z  r r r  z2  r
absolute temperature, К; Т* = 647,256 К; δ = ρ/ρ*; ρ is a    
density, kg/m3; ρ* = 317,7 kg/m3. The conductivity of
vˆ v v v v v    1  rv   2v 
water steam in the state of ideal gas 0 ( ) is determined  vr    r  vz       f ,
 r r z  r r r  z2  
by the equation:    
3
0 ( )   0,5  ак к , а0 = 0,01028; а1 = 0,02996; vz  vz v v  1   vz 2
  vz 

 vr z  vz z   
r z r  z 2   f z .

к 0
 r r  r  
а2 = 0,01561; а3 = –0,004224.

3
MATEC Web of Conferences 132, 04020 (2017) DOI: 10.1051/matecconf/201713204020
DTS-2017

Calculation of pressure distribution (the second vˆr  vr 1 P vˆz  vz 1 P

subtask):   ,   .
  r   z
 1    rvr     vz     P  2 P where  is a time step; v is a value of velocity field on
   r   2 the previous time layer; v is a value of velocity field on
t r r z r r  r  z
the intermediate time layer; v̂ is on the current time
or with consideration the state equation layer.
  PМ /  RT     rvr     vz 
r  r 
t r z 7 Software description
  P  2 P
 r   r 2 . The input model parameters are the next: rotation
r  r  z frequency of the turbine, the range of output speeds at
The simplified hydrostatic model of water motion is the turbine blades, number of blades, the number of
used as the initial approximation for this problem. It is nozzle channels, the width of the nozzle channel to the
significantly decreased the calculation time. impeller, the vapor pressure of the working wheels. The
The third subtask is determined the distribution of the calculations were performed in three functional areas
steam velocity at the upper time layer by the explicit 1 – 3. The location of areas which were determined by
formulas: the velocity field is given in Fig. 1.

Fig. 1. The geometry of turbine housing and the nozzle grid.

4
MATEC Web of Conferences 132, 04020 (2017) DOI: 10.1051/matecconf/201713204020
DTS-2017
Fig. 2. The velocity of steam movement in working chamber.
The flow pattern inside a pair of working chambers
(shown in a longitudinal and radial component of the
velocity vector) is given in Fig. 2.
Fig. 3. Distribution fields of the main calculated physical quantities inside for the first working chamber
8 Conclusion
The mathematical model of aerodynamic processes in
the steam turbine was developed. The main equations are
the next: the Navier-Stokes equations, continuity
equations and state equations. The initial system of
equations was determined in cylindrical coordinate
system taking into account the axial symmetry of steam
flow field in the turbine. The splitting schemes into
physical processes (the pressure amendment method)
was used for approximation of problem at the temporary
variable. According to this method, the solution problem
is calculated in three steps. The solution of diffusion-
convection problem with high Peclet numbers is required
in calculation of velocity field components without
pressure. New difference schemes were obtained on the
basis of the modification of the "cabaret" scheme and
used for this problem solution. Owning to these
difference schemes, the accuracy of diffusion-convection
problem solution is increased in 2,5 – 3 times, compared
to the traditional "cabaret" schemes, which are effective
*
Corresponding author: cheese_05@mail.ru
5
Color reflects the speed of movement of the steam.
The units of measurement are meters per second. The
developed software allows to calculate the movement of
air environment within areas with cylindrical symmetry
and allows to identify patterns of velocity fields and
pressure within axial steam turbines. The permeability of
the medium is used in the mathematical model to
describe the nozzle geometry of the lattice. The
developed software can calculate the steam movement in
multiple stages at the same time.
Distribution fields of the main physical values
calculated for inside the first working chamber (1 is the
longitudinal and radial component of the velocity vector,
m/s; 2 is the rotation speed, rev/min; 3 is the pressure, atm.;
4 is the turbulent viscosity, Pа·s; 5 is the temperature, °C;
6 is the density, kg/m3) are given in Fig. 3.
for solution such problems. The software
implementation was performed on the basis of the
developed algorithms. Results of numerical calculations
of aerodynamics processes in the steam turbine were
obtained.
This work was financially supported by the FTPRD «Research
and Development for the Development of the Scientific and
Technological Complex of Russia for 2014-2020», Project No.
14.577.21.0260.
References
1. A.A. Samarsky, E.S. Nikolaev, Methods of solving
grid equations (M.: Science, 1978)
2. A.I. Sukhinov, A.E. Chistyakov, J. Mathematical
modeling, 24, no. 1, 3-20 (2012)
3. A.N. Konovalov, J. Differential equations, 40, no. 7,
(2004)
MATEC Web of Conferences 132, 04020 (2017) DOI: 10.1051/matecconf/201713204020
DTS-2017

4. A.N. Konovalov. Siberian mathematical journal, 43, International Conference on Informatics,

no. 3, 552 (2002) Electronics and Vision (ICIEV), 1128 – 1133 (2016)
5. A.I. Sukhinov, A.E. Chistyakov, A.V. Shishenya J. 23. A.I. Sukhinov, A.E. Chistyakov, A.V. Nikitina A.A.
Mathematical modeling, 25, no. 11, 53-64 (2013) Semenyakina, I. Korovin, G. Schaefer Conference
6. A.I. Sukhinov, A.E. Chistyakov, N.A. Fomenko, J. on Informatics, Electronics and Vision (ICIEV).
Izvestiya SFedU engineering sciences, no. 4, 87-98 1134 – 1139, (2016)
(2013) 24. A.I. Sukhinov, A.E. Chistyakov, A.A. Semenyakina,
7. A.A. Samarskiy, Theory of difference schemes. A.V. Nikitina Computer researches and modeling. 8,
(Nauka, Moscow, 1989) no. 1., 151 – 168 (2016)
8. A.A. Samarskiy, A.V. Gulin, Numerical methods 25. A.I. Sukhinov, D.S. Khachunts, A.E. Chistyakov,
(Nauka, Moscow, 1989) Computational Mathematics and Mathematical
Physics, 55, No. 7., 1216-1231 (2015)
9. K.A. Beklemishev, I.B. Petrov, A.V. Favorsky, J.
Mathematical modeling, 25, no. 11, 3-16 (2013) 26. A.I. Sukhinov, A.E. Chistyakov, A.V. Shishenya
Mathematical Models and Computer Simulations. 6,
10. I.B. Petrov, A.V. Favorsky, A.V. Sannikov, I.E.
No. 3.,324-331 (2014)
Kvasov, J. Mathematical modeling, 25, no. 2, 42-52
(2013) 27. N. Buzalo, P. Ermachenko, T. Bock, A. Bulgakov
A.Chistyakov, A. Sukhinov, E. Zhmenya, N.
11. A.I. Sukhinov, A.E. Chistyakov, A.A. Semenyakina
Zakharchenko Procedia Engineering, 85, No. C.,
A.V. Nikitina, J. Computational methods and
84-93 (2014)
programming: new computing technology, 16, no. 2,
256-267 (2015) 28. A.I. Sukhinov, A.E. Chistyakov, E.F. Timofeeva,
A.V. Shishenya, Mathematical Models and
12. A.V. Nikitina, A.A. Semenyakina, A.E.Chistyakov,
Computer Simulations, 5, No. 2, 122-129, 2013
E.A. Protsenko, I.V. Yakovenko, J. Fundamental
research, no. 12-3, 500-504 (2015) 29. A.I. Sukhinov, A.E. Chistyakov, Mathematical
Models and Computer Simulations, 4, Iss. 4, 398-
13. A.E. Chistyakov, D.S. Hachunts, A.V. Nikitina,
409, (2012)
E.A. Protsenko, I.Yu. Kuznetsova, J. Modern
problems of science and education, no. 1-1, 1786
(2015)
14. A.I. Sukhinov, A.V. Nikitina, A.A. Semenyakina,
E.A. Protsenko, J. Engineering journal of Don, 38,
no. 4 (38), 52 (2015)
15. A.V. Nikitina, Y.A. Abramenko, A.E. Chistyakov,
J. Informatics, computer science and engineering
education, no. 3 (23), 49-55 (2015)
16. A.E. Chistyakov, A.V. Nikitina, G.A. Ougolnitsky
V.M. Puchkin, I.S. Semenov, A.I. Sukhinov, A.B.
Usov, J. Game Theory and Applications, 17, 37-48
(2015)
17. A.E. Chistyakov, A.V.Nikitina, V.V. Sumbaev, J.
Herald of computer and information technologies,
no. 8 (146), 3-7 (2016)
18. A.V. Nikitina, A.A. Semenyakina, A.E. Chistyakov,
J. Herald of computer and information technologies,
no. 7 (145), 3-8 (2016)
19. A.I. Sukhinov, A.E. Chistyakov, A.A. Semenyakina
A.V. Nikitina, J. Computer researches and
modeling, 8, no. 1, 151 – 168 (2016)
20. A.I. Sukhinov, A.V. Nikitinа, A.A. Semenyakina,
A.E. Chistyakov, CEUR Workshop Proceedings,
10th Annual International Scientific Conference on
Parallel Computing Technologies, PCT 2016; 1576,
308-319, (2016)
21. A.V. Nikitina, A.I. Sukhinov, G.A. Ugolnitsky, A.B.
Usov, A.E. Chistyakov, M.V. Puchkin, I.S.
Mathematical Models and Computer Simulations. –
2017. 9 (1), 101-107.
22. A.I. Sukhinov, A.E. Chistyakov, I.I. Levin, I.S.
Semenov, A.V. Nikitina, A.A. Semenyakina 5th