Programmable Controller Design Based on a Synchronized Colored Petri Net Model and Integrating Fuzzy Reasoning.

Abstract

The main purpose of this paper is to present the use of a synchronized colored Petri net model as the formalism to specify control for discrete event real time low cost programmable controllers (PLCs), integrating an extended fuzzy controller with rule based approximate reasoning c apabilities. The goal of the proposed model is the integration of different formalisms in a common specification. Supported formalisms include the ``classical'' programmable controller languages, and new paradigms, like fuzzy controllers and approximate reasoning, as well. A description of the in- and output signals to be handled by the controller is made, namely possible types and internal representations. The analysis of the Petri net model in order to support automatic code generation is pursued through the construction of the reachability tree for the non autonomous net model. Support for distributed control implementation is also discussed. Examples of modelling a FIFO production system and intelligent buildings monitoring are presented to illustrate the modelling capabilities and applicability.