An Application for the Control of Stochastic Petri nets via Fluidification Approach

Hanife Apaydın Özkan
1.718 513

Abstract


Petri nets are frequently used for modeling and analysis of discrete event systems. Similar to other modeling formalisms for discrete systems,  it suffers from state explosion. Fluidification can be used to overcome this difficulty yielding  fluid approximation of original Petri nets  in the sense of behaviours and  properties. This models are called  continuous Petri nets. In this work, stochastic Petri nets and their fluid approximation timed continuous Petri nets is considered. One of the main advantages of timed continuous Petri nets is to be able to design a controller by using more analytical techniques. But it is important to come back to a reasonable design or control in the original discrete setting.  In this work, a target state control strategy of timed continuous Petri nets will be interpreted for the control of underlying Stochastic Petri nets.  The efficiency of this interpretation  will be studied  on  a table factory system.

Petri nets are frequently used for modeling and analysis of discrete event systems. Similar to other modeling formalisms for discrete systems,  it suffers from state explosion. Fluidification can be used to overcome this difficulty yielding  fluid approximation of original Petri nets  in the sense of behaviours and  properties. This models are called  continuous Petri nets. In this work, stochastic Petri nets and their fluid approximation timed continuous Petri nets is considered. One of the main advantages of timed continuous Petri nets is to be able to design a controller by using more analytical techniques. But it is important to come back to a reasonable design or control in the original discrete setting.  In this work, a target state control strategy of timed continuous Petri nets will be interpreted for the control of underlying Stochastic Petri nets.  The efficiency of this interpretation  will be studied  on  a table factory system.


Full Text:

PDF