See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/262424007

Control valves - Modeling and simulation

Conference Paper · July 2009

CITATIONS READS

9 11,115

3 authors:

Cristian Pătrăşcioiu Casen Panaitescu

Petroleum - Gas University of Ploiesti Petroleum - Gas University of Ploiesti
66 PUBLICATIONS   145 CITATIONS    56 PUBLICATIONS   254 CITATIONS   

SEE PROFILE SEE PROFILE

Nicolae Paraschiv
Petroleum - Gas University of Ploiesti
68 PUBLICATIONS   181 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Production Potential Resumption of Abandoned Oil Reservoirs View project

All content following this page was uploaded by Cristian Pătrăşcioiu on 17 May 2017.

The user has requested enhancement of the downloaded file.

 Proceedings of the 5th WSEAS Int. Conf. on DYNAMICAL SYSTEMS and CONTROL

Control Valves - Modeling and Simulation

CRISTIAN PATRASCIOIU1, CASEN PANAITESCU2, NICOLAE PARASCHIV1
1
Computers & Control Department
2
Chemical Engineering Department
Petroleum-Gas University of Ploiesti
Bd. Bucuresti 39, Ploiesti, 100680
ROMANIA
cpatrascioiu@upg-ploiesti.ro
c.panaitescu@gmail.ro
nparaschiv@upg-ploiesti.ro

Abstract: - The paper presents the researches of the authors in the modeling and the simulation of the control
valves. The control systems with the control valves are complex structures and non-linear characteristics
because the construction and the hydraulic phenomena associated of control valves. The authors have
elaborated a mathematical model for the control valve. The model may be utilized to verify the work
characteristic of the control valve for all operating points. The paper has four parts. First part describes the
structure of the control valve. The second part contains the control valve model elaborated by the authors. The
elements of the model are: the centrifugal pump model, the pipe model, the intrinsic and the work model of the
control element. The third part is destined to elaboration of the simulation program. In the last part, the authors
have presented and have analyzed the numerical results of the simulation.

Key-Words: - model, control valve, hydraulic system, pump, pipe, numerical simulation

1 Introduction have continued the development and numerically

The control valve represents an important element of solved of the control valve model.
the control systems. The control valve is a non-linear
element, characterized by the complexity of the 2 Control valve – element of the
mechanical construction and the hydraulic control system
phenomena. The control valve is an element of the control
One important problem of the control engineers is system and it is the most widespread control element
the work characteristic of the control valve. This in the field of chemical and petrochemical industry.
problem forces the control engineers to model and The industrial control valve contains an electronic-
simulate the control valves. The fundament of the pneumatic converter, a pneumatic actuator with
control valve’s mathematical modeling is membrane and a control element with a seat, to see
represented by [1, 2]. the structure presented in figure 1.
The model of the control valve is used into
mathematically model of the control system. If the
control system is equipped by centrifugal pump, the
numerical modeling of the control valves is an actual
problem [3, 4]. Usually, the authors present the
solutions associated to the operational characteristic
of the control valve determined by using the
simplifying hypothesis, which consider that the pipe Fig. 1. The structure of the control valve
drop pressure is not modified.
The authors of this paper have focused the The significance of the variables and the
researches in the domain of the numerical modeling subsystems: I/P –electro-pneumatic converter; SM –
and simulation of the control valves into control actuator; OR – control element; u – electrical signal
system equipped by centrifugal pumps. command; pc – pneumatic signal command; h – the
stroke of the actuator; pSM – disturbances associated
The first part of the researches has been presented in
the papers [5, 6]. In the present article, the authors

ISSN: 1790-2769 63 ISBN: 978-960-474-094-9

 Proceedings of the 5th WSEAS Int. Conf. on DYNAMICAL SYSTEMS and CONTROL

to the actuator; pOR – disturbances associated to the of the pipe Pout.

control element.
For the control system, the control valve is
considered a mono-variable system, the input is the
u command of the controller and the output is the
manipulated variable m, associated to the process.

Many factories made the control valves

characterized by many constructive tips, by
standardized flow module and geometric dimension.
An example of the industrial control valve is
presented in figure 2 [7].

Fig. 3. The model of the control valve in hydraulic

systems

3.1 The model of the centrifugal pump

For the chemical and oil refining industry, the most
usually hydraulic power source is the centrifugal
pump. In figure 4 is presented a usual centrifugal
pump and the various pressure-flow rate
characteristics.

Fig.2. The control valve made of the Pre-Vent [7].

3 The model of the control valve in

hydraulic systems
The authors have elaborated a mathematical model
of control valve. It is based on the following
elements: the model of the centrifugal pump; the
model of the pipe; the work characteristic model of
the control element; the intrinsic characteristic of the
Fig. 4. The centrifugal pumps and the pressure-flow
control element and the energetic balance of the
rate characteristics
hydraulic system [1, 3, 4, 5 6]. The model is defined
by the diagram presented in figure 3. The input of The mathematical model of the centrifugal pump
the control valve’s mathematical model is the h may be approximated using the following equation:
stroke of the actuator and the output is the Q flow
which goes through the valve. P0 = a 0 + a1Q + a 2 Q 2 + a 3 Q 3 , (1)
where P0 represents the pump output pressure and Q
The intern variables of the model are: the flow module of – the flow rate. Using the polynomial regression
the control valve Kv, the pump output pressure P0,, the method, the authors have determinate the numerical
drop pressure of the control valve ΔPv , the drop coefficients of the relation (1), associated to static
pressure of the pipe ΔPp and the pressure of the exit characteristic pumps presented in figure 4, table 1.

ISSN: 1790-2769 64 ISBN: 978-960-474-094-9

 Proceedings of the 5th WSEAS Int. Conf. on DYNAMICAL SYSTEMS and CONTROL

Table 1. The numerical coefficients of the mathematical model (1)

Numerical coefficient Standard
Pump tip deviation
a0 a1 a2 [m H2O]
32-13 6.614213E+00 4.727929E-02 -3.938231E-02 6.621437E-02
32-16 8.851658E+00 7.501579E-01 -1.081358E-01 3.631092E-01
50-20* 1.418321E+01 -2.746576E-02 -1.080953E-03 2.887783E-01
150-26 2.037071E+01 -7.850898E-03 -3.007388E-05 4.614905E-01
32-12 9.351883E+00 -1.262669E-01 3.939355E-04 1.890947E-01
50-13 8.227606E+00 -4.412075E-02 -1.679803E-05 1.578524E-01
40-16 1.258403E+01 -9.795947E-02 7.796262E-05 5.434285E-01
32-20 1.781198E+01 -2.755295E-01 1.361482E-03 3.952989E-01
40-20 1.742569E+01 -1.339487E-01 2.504382E-04 2.826216E-01
*
selected pump for the simulation

3.2 The model of the pipe Re > 3000 . (6)

The model of the pipe is represented by the drop
The relationships (5) and (6) represent two non-
pressure for a straight pipe, with circular section,
linear equations. For the intermediary flow regime,
where the fluid flows in a turbulent regime [11]
2300 < Re < 3000 , will be utilized the equation
8L ⎡ N ⎤
ΔPp = λ Q2 ⎢ m2 ⎥ . (2) ⎛ 2ε 18.7 ⎞ 1
π 2 D5 ⎣ ⎦ f (λ ) = 1.74 − 2 lg ⎜⎜ + ⎟⎟ − =0 . (7)
⎝ D Re λ ⎠ λ
where λ is the friction coefficient; L – pipe length, in
m; D – pipe diameter, in m; Q – volume flow of the For the turbulent flow regime, Re > 3000 , will be
fluid, in m3/s. utilized the equation
The value of the λ friction coefficient depends of the ⎛ ε 2.51 ⎞ 1
flow regime, characterized by the Reynolds number f (λ ) = −2 lg ⎜⎜ + ⎟⎟ − = 0. (8)
⎝ 3.7 D Re λ ⎠ λ
Re, by the rugozity coefficient ε D and the
diameter D of the pipe. The number Reynolds is
defined by the relationship 3.3 The intrinsic and the work characteristic
of the control element
Dw The intrinsic characteristic represents a
Re = , (3)
ν mathematical model of the control element in
standard working and this represents the dependence
where w represents the linear velocity of the fluid, in
between the standard flow module K v and the
m/s; ν - the cinematic viscosity of the fluid, in m2/s.
stroke h
The friction coefficient is calculated using the K v = f (h ) . (9)
Moody’s formula based on curve-fit on the
The standard flow module depends only on the
experimental data [7, 8, 9, 10]:
mechanical and hydraulic characteristics of the
64 control element. The geometrical construction of the
λ= , Re < 2300 ; (4) control element may be: valve shut – off type with
Re
one seat, valve shut – off type with two seats, three-
1 ⎛ ε 18.7 ⎞ way valve shut – off type, corner valve shut – off
= 1.74 − 2 lg ⎜⎜ 2 + ⎟⎟ ,
λ ⎝ D Re λ ⎠ type, flap valve shut – off type, spherical valve shut
– off type. The most important mathematical models
2300 < Re < 3000 ; (5) of the intrinsic characteristic associated to the valve
shut – off type with one seat are the linear
1 ⎛ D 2.51 ⎞ characteristic and the logarithmic characteristic.
= −2 lg ⎜⎜ + ⎟⎟ ,
λ ⎝ 3.7ε Re λ ⎠

ISSN: 1790-2769 65 ISBN: 978-960-474-094-9

 Proceedings of the 5th WSEAS Int. Conf. on DYNAMICAL SYSTEMS and CONTROL

The mathematic model has the forms [2]: variables of the simulation program are presented in
table 2. The hydraulic system has the structure
a) the linear characteristic
presented in figure 5. The program calculates the Q
Kv Kv0 ⎛ Kv0 ⎞ h flow rate corresponding to the relative value of the
= + ⎜1 − ⎟ ; (10) stroke of the control valve, h h100 ∈ [1K100] % . The
Kvs Kvs ⎝ Kvs ⎠ h100
b) the logarithmic characteristic model of the control valve presented in figure 3 is
represented by the non-linear equation
Kv Kv 0 ⎛ h Kvs ⎞
= exp ⎜⎜ ln ⎟, (11) ΔPv
Kvs Kvs ⎝ h100 Kv 0 ⎟⎠ f (Q ) = Q − K v =0. (14)
ρ
where h is the valve shut – off stroke versus the seat;
H100 – the nominal value (maximum) of the valve The link between the stroke of the control valve
shut – off stroke; Kv0 – the value of Kv for h = 0 ; Kvs h h100 and the flow rate Q are the relation (10) or
– the value of Kv at H100 nominal stroke. (11), dependent by the intrinsic characteristic (linear
or logarithmic). The pressure drop of the control
The work characteristic of the control element valve ΔPv is calculated using the energetic balance
represents the dependency between the Q flow rate of the hydraulic system (13).
and the flow module, the control valve drop pressure
and the density of the fluid Table 2. The input variables of the simulation
program
ΔPv Variable Measure unit Value
Q = KV . (12)
ρ Pipe
Diameter m 0.05
3.5 The energetic balance of the hydraulic Length m 20
system Rugozity - 0.03
The hydraulic system may be associated to: hydraulic
Fluid
system without ramifications, hydraulic systems with
Viscosity m2 s-1 0.92e-6
ramifications, hydraulic systems with three-way control
Density Kg m-3 476
valves. The authors have focused their researches on the
Control valve
hydraulic system without ramifications, figure 5. This
hydraulic system is characterized by the pump output Intrinsic Linear
pressure P0,, the pressure drop of the control valve characteristic
ΔPv , the pressure drop of the pipe ΔPp , the pressure Kvs m3 h-1 25
Kv0 m3 h-1 1
of the exit of the pipe Pout and the flow rate Q. Pump
Type 50-20 The coefficient of equation
(1) are presented in table 1

The elements of the energetic balance are defined by

the output pressure pump (1) and the pipe drop
pressure (2).

Fig. 5. The structure of the hydraulic system with 5 Numerically results

control valve The numerically results are represented by the next
graphically correlations: the dimensional flow rate –
The energetic balance of the hydraulic system is the valve stroke, the flow module – the valve stroke,
defined by the relation the pipe drop pressure – the valve stroke, the valve
Po = Pout + ΔPv + ΔPp . (13) drop pressure – the valve stroke and the friction
coefficient – the valve stroke.

4 The Simulation Program The flow module versus the valve stroke correlation
The authors have elaborated a special program for describes the intrinsic linear characteristic of the
the calculus and the graphical representation of the control valve, figure 6.
work characteristic of the control valve. The input

ISSN: 1790-2769 66 ISBN: 978-960-474-094-9

 Proceedings of the 5th WSEAS Int. Conf. on DYNAMICAL SYSTEMS and CONTROL

100 3,50E-03
90
3,00E-03
80

P ip e d r o p p r e s su r e [b a r ]
70 2,50E-03

60 2,00E-03
K v [% ]

50
1,50E-03
40
30 1,00E-03
20
5,00E-04
10
0 0,00E+00
1 11 21 31 41 51 61 71 81 91 1 11 21 31 41 51 61 71 81 91
Stroke [%]
Stroke [%]

Fig. 6. The intrinsic characteristic of the control Fig. 8. The pipe drop pressure versus the valve
valve stroke

Because has used the linear model of the flow The same cause determines the variation of the
module, the intrinsic characteristic has resulted valve drop pressure. Because the pipe drop pressure
linear also. is very small, the valve drop pressure is very big,
figure 9.
The work flow rate characteristic is quasi linear, 1,4
figure 7. This control valve characteristic determines 1,2
the control system characteristic and in this case, the
V alve d rop p ressu re [b ar]

characteristic of the control system is linear if the 1,0

transducer characteristic is linear too. 0,8
100
0,6
90
80 0,4

70 0,2
F lo w r a te [% ]

60
0,0
50 1 11 21 31 41 51 61 71 81 91
40 Stroke [%]
30
Fig. 9. The valve drop pressure versus the valve
20
stroke
10
0 The characteristic of the valve drop pressure is
1 11 21 31 41 51 61 71 81 91 similarly as the pump characteristic (theses
Stroke [%] characteristics are calculated versus the valve
stroke). The conclusion of this result is that 99% of
Fig. 7. The work characteristic of the control valve the hydraulic energy of pump is lost into control
valve. Though the control valve work characteristic
In the example presented in table 2, the authors have is linear, that is a wrong solution for the control
considered a relative small pipe, without the local system because is an expensive solution due to the
hydraulic resistors. In this case, the pipe drop lost energy of the control valve.
pressure has small values, figure 8.
The results of the simulation program have validated
the model of the friction factor. In the figure 10 is
presented the correlation between the friction factor
value and the control valve stroke.

ISSN: 1790-2769 67 ISBN: 978-960-474-094-9

 Proceedings of the 5th WSEAS Int. Conf. on DYNAMICAL SYSTEMS and CONTROL

References:
0,059
[1] Marinoiu, V., Control elements. Control Valves,
Editura Tehnică, Bucureşti, 1999 (Romanian).
0,058 [2] * * * Control Valves Handbook,
F r ic tio n fa c to r

http://www.documentation.emersonprocess.co
0,058 m/groups/public/documents/book/cvh99.pdf
[3] Hayes R. W., Match valves with pumps to
improve system performance, Hydrocarbon
0,058
Processing, May 1995.
[4] Stedham D., Are your valves operating at top
0,057 effiency ?, Hydrocarbon Processing, January
1 11 21 31 41 51 61 71 81 91 1995.
Stroke [%] [5] Patrascioiu C., Mihai L., The Modeling of
Control Valves in Hydraulic Systems,
Proceedings of the 9th International
6 Conclusion Carpathian Control Conference, Sinaia,
The paper has presented the following: Romania, 2008.
− The structure analysis of the control valve; [6] Patrascioiu C., Mihai L., The Modeling and
− The mathematical model of the control Simulation of the Control Valves in Hydraulic
valve; Systems, Petroleum-Gas University of Ploiesti,
− The simulation program and an example of Bulletin, Technical Series, LX, No. 3B, 2008.
the input data of the model; [7] www.pre-vent.com
− The results of the simulation program. [8] Jinescu, G., Vasilescu, P.,Jinescu, C., The Real
Fluide Dynamic into process plant, Editura
The structure of the control valve mathematical SemnE, Bucureşti. 2001 (Romanian).
model is properly of the authors. The elements of [9] Chin, W., C., Computational Rheology for
the model are presented in literature. The authors Pipeline and Annular Flow, Boston, MA, 2000.
have solving the model and they have elaborated a [10] Fester V.G., Slatter P.T., Dynamic similarity for
special simulation program. The numerical results non-Newtonian fluids in globe valves,
are: the dimensional flow rate, the flow module, the Chemical Engineering Research and Design
pipe drop pressure, the valve drop pressure and the (87), 2009, 291–297.
friction coefficient, all results are depended of the [11] Romeo E., Carlos R., Monzón A., Improved
valve stroke. explicit equations for estimation of the friction
The model and the simulation program will be used factor in rough and smooth pipes, Chemical
to study the work characteristic of the control valve Engineering Journal 86 (2002), p. 369–374.
and to study the performances of the control
systems.

ISSN: 1790-2769 68 ISBN: 978-960-474-094-9

View publication stats