The Java Based Programmable Logic Controller. New Techniques in Control and Supervision of a Flexible Manufacturing Cell.

Abstract

In this chapter, we explain new techniques and technologies applied to the control of a flexible manufacturing cell. We have proposed a new control platform: a Java based Programmable Logic Controller (PLC). The Java PLC comprises several modules where the real time control, the communication with industrial fieldbuses and the supervision via web technologies have been developed. This control architecture (Piedrafita & Villarroel 2006) and development environment is based on Petri Nets (PNs), Sequential Function Charts (SFCs) and the Real Time Java programming language. Our objective is to explore the application of new control techniques of manufacturing systems, and to test the use of the Java programming language as a platform for those techniques. To demonstrate the practical utility of the techniques, we have applied them to the control of a flexible manufacturing cell. This research follows earlier studies at the University of Zaragoza on the software implementation of PN. In those studies, Ada95 was the implementation language (Garcia & Villarroel 1996). For the current study, Java was chosen for the following reasons: • The possibility of executing the same code on different platforms. • To compare the Ada95 and Java implementations using the same concurrent and realtime characteristics. • Java is a language that is beginning to be used in the development of control and embedded systems. • Java has a real time extension that allows the necessary time predictability required in these types of applications. From the perspective of the software implementation of Discrete Event Control Systems, we have translated into the Java language some classic PN implementation techniques such as Enabled Transitions or Representing Places. We have also developed SFC implementation techniques such as the Deferred Transit Evolution Model and Immediate Transit Evolution Model. In the execution of a Petri net, a task, the coordinator, makes the firing of transitions and updates the marking. We propose a new PN implementation technique called concurrent coordinators, in which centralized and decentralized ideas are merged. The net is decomposed into several subnets following, for example, a functional criterion, and a