Professional Documents
Culture Documents
Continuous Timed Petri Net Model of Brayton Cycle For A Gas Turbine
Continuous Timed Petri Net Model of Brayton Cycle For A Gas Turbine
AbstractThis work developed a continuous timed is developed in Section 3. In Section 4 the continuous
Petri Net model of a gas turbine, using a rst linear Petri Net model based on the equations of section 3 is
order approximation for every state of the Brayton presented. In Section 5 the simulation of the Brayton cycle
Cycle. The Brayton cycle rules the functioning of a
gas turbine, and it is composed by four states: Com- as an hybrid system is shown. Finally, the conclusions and
pression, Combustion, Expansion and Cooling. In every further work are presented in Section 6.
state (excepting the cooling phase) temperature and
pressure are modeled as rst order linear systems, II. Fundamentals
therefore, every system is translated into a continuous A general description of the gas turbine process and an
timed Petri Net. The complete model can be used in
further work in order to design supervisory controllers introduction to continuous Petri Nets are introduced in
to ensure an ecient performance of a gas turbine. this section.
Although the model presented is extremely simplied,
it will be used as a starting point to developed more A. Gas turbine
complex models. For further reading, review (1) and (2). A generic gas
turbine is composed by three elements, each one is asso-
I. Introduction
ciated to a state of the Brayton cycle. The elements are
Gas turbines are used in transport systems as aircraft shown in Fig. 1. The general process is explained below.
and in power generating systems. The principle of the
gas turbine is developed by the Brayton cycle, a ther-
modynamic process which intervenes in the gas turbine
components. The steady-state behavior of the gas turbine
has been widely investigated in engineering area. More-
over, The dynamic behavior has been studied using non
linear models of its components, leading to complicated
mathematical representations. The gas turbine thermody-
namic and mechanic nature is considered dynamic but
the control algorithms are referred to specic punctual
measurements, making the control algorithm work as a
supervisor instead of a dynamic regulator. In this scenario,
the need of modeling the gas turbine system as a Discrete
Event System (DES) has become greater in the latest
years. However, it is impossible to ignore the dynamic
behavior of the system during the Brayton cycle, so a
standard modeling of the gas turbine as discrete state
system is not enough to capture its behavior. Therefore,
in this work the Brayton cycle is considered as an hybrid
system with four discrete states, compression, combustion, Figure 1. Gas turbine
expansion and freezing. Each state has an inherent pres-
sure and temperature dynamics, which has been reduced A contained quantity of air is in the compressor at at-
to rst order systems and modeled as continuous Petri mospheric pressure, then the compression starts increasing
nets. In Section 2, the basic concepts of gas turbines and the pressure of the air and leading it into the combustion
continuous Petri Nets are introduced. The mathematical chamber. Hence, the fuel is injected and the combustion
reduction of the gas turbine dynamics used in this work begins, increasing the temperature of the air-fuel mix and
directing it in a very fast speed to the turbine. The turbine The state equation of a ContPN is shown in Eq. 1
stars rotating, decompressing the air in the output and
generating a constant angular speed. Finally, the air in the ) = C f ( )
m( (1)
output is freeze by the environment. This closed process
is named the Brayton cycle and its introduced in the where is the time variable. C is the net incidence
following subsection. matrix, C = P ost P re.
A ContPN is called innite server semantic if the ow
B. Brayton cicle of a transition ti is
The Brayton cycle is used to dene the behavior of a
gas turbine. It is composed by four states: fi = i minppre(ti ) {m(p)/P re[p, ti ]} (2)
1) Compression. Ideally is an adiabatic process (no Eq. 1 can be expressed as a piecewise linear system given
heat transfer with the environment). However, the by
temperature does increase, as has been shown in
several experiments. m
= C(m)m (3)
2) Heat addition (combustion). An isobaric process (no
change on the uid volume). The ring rate matrix is denoted by = diag(1 , ..., |T | ).
3) Expansion. Again, an ideally adiabatic process. A conguration of a ContPN at m is a set of (p, t) arcs
4) Heat rejection. An isobaric process. describing the eective ow of all transitions:
In Fig. 2 the Brayton cycle in terms of the temperature 1
P re[i,j] if pi is constraining tj
and the pressure is shown. For the propose of this work, (m)[i, j] = (4)
the variables considered in the Brayton Cycle are pressure 0 otherwise
and temperature. Usually in Brayton process variables III. Mathematical description
such as entropy are analyzed, but for simplicity of the
measurements, entropy is not included in the present In this section the model of the Brayton cycle as ah
analysis. hybrid system is developed. A high-level discrete event
system is constructed using the four states of the Brayton
cycle, explained in section II-B. In each state, the dynamic
equations of temperature and pressure are obtained, based
on the thermodynamical relationships introduced in (6).
A. High-level discrete state model
Brayton cycle is composed by 4 states
1) Compression
2) Combustion
3) Expansion
4) Cooling
In add to this states, an initial state representing no
operation of the gas turbine is included in the model. The
corresponding Finite State Machine of the Brayton cycle
is presented in Fig. 3
B. Mathematical representation of temperature and pres-
sure
Figure 2. Brayton cycle In each state of the system, the temperature and pres-
sure evolve according to dynamical non linear equations.
In this description is remarkable that each state occurs For this work, the relationships have been reduced into
in a component of the gas turbine. rst order linear dynamical equations in the form of
C. Continuos timed Petri Nets equation 5.
The following subsection has been taken directly form
= m + u
m (5)
(3). The reader is refered to (4) and (5) for related subjects
in Continuos timed Petri Nets. > 0 and > 0.
A continuous timed Petri net is a 3-tuple ContP N = With u an input variable which increases the value
(N, , m0 ), where (N, m0 ) is a continous Petri Net and of f . The coecients and are determined based
|T |
: T {R+ } is a function associating a ring rate in experimental results of the temperature and pressure
with each transition. behavior in each state.
CoDIT16 - April 6-8, 2016, Malta 390
P2 = P2 P2 + P2 (6) P4 = P4 P4 + P4 (15)
P2
T2 = T1 (10)
P1
T1
T2 = (P2 P2 + P2 ) (11)
P1
T1
T2 = P2 T2 + P2 (12) Figure 4. Compressor Petri net model
P1
CoDIT16 - April 6-8, 2016, Malta 391
M1 P2
B. Combustion chamber Petri net model M2 T2
M3
M4 P3
Using Eqs. 13 and 14 the Petri net model shown in Fig. M5 Air
5 is developed. M6 T3
M7 T0
M8 P4
M9 T4
Table I
Place numeration
T1 (17)
P2 = P2 P2 + P2 M3
P1
M 6 = T3 [M6 M7 ] + T3 M5
T3
M 9 = P4 M9 + P4 M3
P3
Merging the models shown above, a complete gas tur-
bine Petri net model is obtained with the following inci-
dence matrix C and ring rate matrix , using the place
index as shown in Tab. I.
V. Simulation results
Figure 6. Turbine Petri net model Using SIMHPN matlab toolkit (7), the dynamic models
of each state in the Brayton cycle are simulated using the
values presented in Eq. 17 . The results are shown in Figs.
Using the table I, Eqs 6-16 are translate in terms of 7-14.
CoDIT16 - April 6-8, 2016, Malta 392
1 1 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 1 1 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 1 1 0 0 0 0 0 0 0
0 0 1 0 1 1 0 0 0 0 0 0 0 0 0
C=
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 (17)
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 1 1 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 1 1
80 1200
70 1000
Temperature (C)
Temperature (s)
60 800
50 600
40 400
30 200
20 0
0 1 2 3 4 5 6 0 1 2 3 4 5 6
time (s) time (s)
Figure 7. T2 (temperature in the output of compressor) Figure 9. T3 (temperature in the output of combustor)
5
x 10
5 5
x 10
4
3.5 4
Pressure (pa abs)
3
Preesure (pa abs)
3
2.5
2
2
1.5
1 1
0.5
0
0 1 2 3 4 5 6
0 time (s)
0 1 2 3 4 5 6
time (s)
500 4000
3500
400
3000
300 2500
2000
200 1500
1000
100
500
0 0
0 1 2 3 4 5 6 0 1 2 3 4 5 6
time (s) time (s)
Figure 11. T4 (temperature in the output of turbine) Figure 13. (angular speed of the rotor)
5
x 10
2
1.5
Pressure (pa abs)
0.5
0
0 1 2 3 4 5 6
time (s)
Figure 12. P4 (pressure in the output of turbine) Figure 14. Brayton cycle simulates using continous timed Petri net
models