Flexible Manufacturing System Simulation Using Petri Nets

6
Flexible Manufacturing System Simulation Using Petri Nets
Carlos Mireles, Alfonso Noriega and Gerardo Leyva
1. Introduction to Petri Nets.

In 1962 German mathematician Kart Adam on his Ph.D. work Kommunikation mit automaten, he proposed a new way for modelling & representating
systems where events and transitions are present. This new representation is
now known as Petri Nets (PN).
Since then, several researchers have used this tool. In the USA and in Europe
several developments have been done. In 1980, in Europe, a work table was
controlled using PN and the related work was presented at one International
Conference. Researchers in France have done great contributions to the development and the applications of PN. Some of them used the PN to describe
Programmable Logic Controllers (PLC) and this application had a big influence in the development of the Grafcet.
Open Access Database www.i-techonline.com
2. Petri Nets Theory
2.1 Basic Concepts

2.1.1 Petri Net
Graphic and executable technique to specify and analize dynamic & concurrent discrete event systems.
2.1.2 Formal
Petri Nets analysis is a mathematical technique well defined. Many static and
dynamic properties of a PN (and therefore for a system represented by a PN)
can be mathematically proved.
Source: Manufacturing the Future, Concepts - Technologies - Visions , ISBN 3-86611-198-3, pp. 908, ARS/plV, Germany, July 2006, Edited by: Kordic, V.; Lazinica, A. & Merdan, M.
153
Manufacturing the Future: Concepts, Technologies & Visions
154
2.1.3 Graphics
This technique belongs to the area of mathematics known as Graph Theory,

which makes a PN being able to represented both by a graphic and by mathematic expressions. This graphic property provides a better understanding of
the system which is represented by the PN.
2.1.4 Executable
A PN can be executed and then the dynamics of the system can be observed.
2.1.5 Concurrency
Multiple independent entities can be represented and supported by PN.

2.1.6 Discrete event dynamic systems
A system that can change its current state based in both its current state and
the transition conditions between states.
2.2 Structure of a Petri Net
2.2.1 Formal Definition
A Petri Net has a set of Places, a set of Transitions, an Input Function and an
Output Function.
The structure of a Petri Net is the array (L, T, E, S, m0) where:
-
L is the set of places in the graph.

T is the set of transitions in the graph. L & T satisfy the following conditions L T = L T = .
E : LxT {0,1} is the input function which specify the connectivity between Places & Transitions.
S : LxT {0,1} is the output function which specify the connectivity between Transitions & Places.
2.2.2 Graphic Representation
A Petri Net is an oriented graph that contains two types of nodes: Places and
Transitions which are connected by oriented arcs that connect Places with
Transitions connectivity between nodes of the same kind is not allowed. In
the graphic representation, Places are shown as circles, Transitions are shown
155
as bars and the arcs are shown as arrows. Places represent conditions and
Transitions represent events. A Transition has certain number of places either
as inputs or outputs which are pre and post conditions. See figure 1.
2.2.3 Marking a Petri Net
The marking of a Petri Net is a positive integer i for every place Li. A mark is
represented as a dot within the circle for a given place. These marks move between the places which provide the dynamic feature of the Petri Net.
A Petri Net is considered marked when at least one place has a mark.One
place can have N marks, where N is a positive integer. If N = 0 then the place
has no marks. A marker M is a function M : L N that can be expressed by:
m1
m
M= 2
M

mN
(1)
mi is the number of dots for place L i .
Figure 1. Graphic representation of a PN showing the Input & Outputs places as well
as the Transition.
2.2.4 Interpretation of a Petri Net
Marks are the resources. Resources can be physical entites or no physical such
as messages or information.
Places are the locations where the resources are stored.
Transitions are actions that transfer the resources to generate new resources.
156
The weight of each arrow (a number on each arrow) is the minimum number
of resources needed in a place in order to get the action indicated by the transition.
2.2.5 Triggering a Transition in a Petri Net
To trigger a Transition, it has to be validated which means that every Place

connected to this Transition has to have at least one mark. The execution of a
Petri Net is controlled by the number and distribution of the marks on their
Places.
When a Transition is triggered, a change in the marking of the Petri Net is performed and every input place lose the number of marks indicate by the weight
in the arc connecting the Transition with that input place. On the other hand,
every output place gain the number of places inidicated by the weight in the
arc connecting the Transition with that output place. In figure 2 are shown different cases for the triggering of a transition.
L1
L1
t1
=>
L2
L1
L2
L1
t1
t1
L2
L3
L2
L1
t1
L3
L3
L3
L4
(b)
L2
=>
L4
t1
=>
L4
(a)
L1
L2
L1
L1
t1
L4
t1
L2
=>
L3
(c)
Figure 2. Different triggering cases for some Petri Nets.
t1
L2
(d)
L3
157
3. Petri Nets showing relationships amog sub-processes.
3.1 Task Sequencing

Figure 3 shows how different task can be done in sequential way.
Li-1
Li
Li+1
Figure 3. Task Sequencing
3.2 Task Selector

Figure 4 shows a task selector. This graph is useful when we need in a system
an EXOR function between transitions t1 & t2.
Lk
t1
Lj
Figure 4. Task Selector
t2
Lm
158
3.3 Task Synchronization

This is a very useful concept to synchronize two or more ub-processes assuming that they may have different running times or different evolution in time.
Any of the processes can be out of phase respect to the others, however this
model provides a wait state for any of the processes until the triggering of the
transition. Synchronization is achieved when, being validated by the proper
marking on the input places, the event occurs triggering the transition.
Figure 5. Different models to synchronize processes
3.4 Tasks Concurrency

Figure 6 shows how concurrent processes are represented.
t1
Figure 6. Tasks Concurrency
t1
t2
159
3.5 Sharing Resources

Sharing resources is done when two or more sub-processes are using a unique
resource in the system such a robot to load/unload two conveyors. The objective is to avoid the malfunction of one of the processes because the resource is
not available.
L4
L1
L2
L3
L5
Figure 7. Sharing Resources
4. Equation of a Petri Net.

Figure 8 shows a Petri Net which will be used to explain the Petri Net equation.
Figure 8. Sharing Resources
160
4.1 Input Matrix

This matrix provides the connectivity information from Places to Transitions.
In this case Place l1 goes to Transtion t2, Place l2 goes to Transition t2 and so on.
See figure 8.
l1
1
E =
0
l2
0
1
0
l3
t
0 1
t
0 2
t
1 3
(2)
4.2 Output Matrix

This matrix provides the connectivity information from Transitions to Places.
In this case Transtion t2 goes to Place l1, Transtion t2 goes to Place l2 and so on.
See figure 8.
t1 t 2
0 0
S =
1 0
0 1
t3
l
1 1
l
0 2
l
0 3
(3)
4.3 General Matrix of a Petri Net

The two last matrixes are used to define the Net Matrix given by:
A=SE
(4)
4.4 Evolution of a Petri Net

The evolution of the Petri Net is related with the movement of the marks in
the graph from one place to another. The recursive equation is given by:
k +1 = k + AV k
(5)
161
where:
represents the marking vector before the evolution of the net.
k
k +1
Vk
represents the General Matrix of the net.

represents the Transitions that are been triggered in the current
represents the marking vector after the evolution of the net.
evolution step
of the net.
4.5 A working Petri Net

Figure 9 shows a Petri Net before and after the evolution step (trigger of t1).
t5
t5
L1
L2
L1
t4
L2
t4
t1
L3
t1
L3
L4
t3
t2
a)
L4
t3
t2
b)
Figure 9. A Petri Net before and after the trigger of transition t1
5. Application Examples.
5.1 Wagon Control

Figure 10 shows a wagon that moves along a rail by the action of signals R
(right) and L (left). Extreme positions are detected by sensors A & B. Button M
starts the sequence when wagon is in initial position.
162
Figure 10. Wagon Control
There are three places (l1, l2 & l3): Idle, Moving from A to B & Moving from B to
A. There are three transitions (t1, t2 & t3): Switch M, wagon detected by sensor
A & wagon detetcted by sensor B. When M is press, the wagon moves from A
to B, once reaches this point goes back to A. Sequence can only starts when
wagon is in Idle again.
Figure 11 shows the Petri Net for this system.
Figure 11. Petri Net for the Wagon Control system
163
Following the matrixes for this system:

t1 t 2
1 0
E =
0 1
0 0
t3
l
0 1
l
0 2
l
1 3
t1 t 2
0 0
S =
1 0
0 1
t3
l
1 1
l
0 2
l
0 3
A=SE
1
A=1
Where
0
1
1
1
V0 = 0

0
1
0 = 0
0
1
0
(6)
(7)
k +1 = k + AV k then:
1 1 1 1 0
1 1 0
1 = 0 + AV 0 : 1 = 0 + 1 1 0 0 = 0 + 0 = 1
1 1 0 0 0 0
0 0
1 0 0 0 0
0 1 0
1 = 0 + AV 0 : 1 = 1 + 1 1 0 1 = 1 + 1 = 0
0 0
1 1 0 0 0 1
1 0 0 1 1
0 1 0
1 = 0 + AV 0 : 1 = 0 + 1 1 0 0 = 0 + 0 = 0
1 0
1 1 1 1 1 0
(8)
(9)
(10)
6. Simulation Software
In order to simulate the Petri Nets, a software kit was developed. Although
there are some simulators available, we took the decision to develop our own
as the main goal and to have a software system able to be interfaced with the
real world. The software was developed in Lab Windows CVI, having a
Graphic User Interface which allows the user to create easily the net that
represents the system that is needed to simulate. Figure 12 shows the tipycal
interface of the system.
164
Figure 12. Main view of the simulator
The operator inputs the data related with the PN structure, the information is
then interpreted by the simulator which shows the development of the PN. On
figure 13, the Petri Net shown on figure 11 was introduced to the simulator.
Figure 13. Petri Net for the Wagon Control system
Figures 14, 15 & 16 shows the status of the Petri Net when transitions t1, t2 & t3 are
triggered.
165
Figure 14. Petri Net for the Wagon Control system after t1 is triggered
Inside the simulator each circle represents a Place where the wagon is moving
to the right, to the left, or simply is on the idle state waiting for the button M to
be pushed (transition t1) to start a sequence.
166
It is worth to say that this simulator can easily be used as a Control System just
by adding an Acquisition Board for sampling the actual signals which in turn
can activate the transitions. The purpose with this was to develop a non expensive control system which can be used by very small companies that do not
have the capital to buy very expensive equipment. In comparison with a PLC
controller the cost involved when implementing this approach is very low and
also offers an intuitive way of developing the control system.
7. Benefits of a Petri Net.
Petri Nets have the following benefits:
a) They have a graphic representation which allows a simply and secure
way to observe how the system is working.
b) They have semantics well defined that specify the net.
c) They can be used to represent several types of systems.
d) They have an explicit description for states and actions.
e) They provide interactive simulations where the results are shown directly on the graphic diagram of the net.
In future research this system can be used on the control of production lines
where series of events (discrete events) need to be controlled, such as a robot
feeding a conveyor, a conveyor moving cans or packages into storage areas,
etc.
8. References
Mireles, C. et al. (2004), Flexible Manufacturing System Simulation using Petri
Nets
http://www.lurpa.ens-cachan.fr/grafcet/generalites/presentation_uk.html
Manufacturing the Future
Edited by Vedran Kordic, Aleksandar Lazinica and Munir Merdan
ISBN 3-86611-198-3
Hard cover, 908 pages
Publisher Pro Literatur Verlag, Germany / ARS, Austria

Published online 01, July, 2006
Published in print edition July, 2006

The primary goal of this book is to cover the state-of-the-art development and future directions in modern
manufacturing systems. This interdisciplinary and comprehensive volume, consisting of 30 chapters, covers a
survey of trends in distributed manufacturing, modern manufacturing equipment, product design process,
rapid prototyping, quality assurance, from technological and organisational point of view and aspects of supply
chain management.
How to reference
In order to correctly reference this scholarly work, feel free to copy and paste the following:
Carlos Mireles, Alfonso Noriega and Gerardo Leyva (2006). Flexible Manufacturing System Simulation Using
Petri Nets, Manufacturing the Future, Vedran Kordic, Aleksandar Lazinica and Munir Merdan (Ed.), ISBN: 386611-198-3, InTech, Available from:
http://www.intechopen.com/books/manufacturing_the_future/flexible_manufacturing_system_simulation_using
_petri_nets
InTech Europe
University Campus STeP Ri

Slavka Krautzeka 83/A
51000 Rijeka, Croatia
Phone: +385 (51) 770 447
Fax: +385 (51) 686 166
www.intechopen.com
InTech China
Unit 405, Office Block, Hotel Equatorial Shanghai

No.65, Yan An Road (West), Shanghai, 200040, China
Phone: +86-21-62489820
Fax: +86-21-62489821

Flexible Manufacturing System Simulation Using Petri Nets

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Flexible Manufacturing System Simulation Using Petri Nets

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Flexible Manufacturing System Simulation Using Petri Nets

Uploaded by

Copyright:

Available Formats

6