Modelling and Implementation of Supervisory Control Systems Using State Machines with Outputs
Moacyr Carlos Possan, Andr Bittencourt · InTech eBooks · 2012
IntroductionThe growth of the complexity of automated systems in industry has occurred extensively in recent years due to the production demand increase, quality improvements and flexibility to restructure the manufacturing systems in order to satisfy new procedures.Nevertheless, the evolution of control devices and their functionalities, such as processor speed, memory and network communication has advanced in parallel with the factories' requirements.Although the evolution of automation in industrial processes is a fact, there is still a scarcity of formal methods for analysis, project and implementation of control systems for Discrete Event Systems (DES) in order to reduce the development time, reduce human resources investment, and satisfy the operational requirements for certain systems in an effective way.Furthermore, the occurrence of programming bugs resulting in errors due to interruption in the process and losses due to poorly designed software is obviously unacceptable nowadays in this market that has a just-in-time mindset and is strongly focused on profits.Usually, the projects for supervisory control systems are based on the knowledge of the system's practitioner, according to his experience in programming.The usage of formal methods is sparse, such that the reuse of documentation and source code as well as the dissemination of the knowledge generated are both impaired.Moreover, the automation of manufacturing systems has brought an increase in the complexity of control systems, so that to elaborate and implement robust and reliable control logic is not a trivial task.In order to minimize the risks due to programming errors and to permit a formal method for modelling DES, Ramadge & Wonham (1989) introduced the Supervisory Control Theory (SCT), which guarantees optimal control logic (nonblocking and minimally restrictive) for these systems.A Discrete Event System (DES) consists of a system with discrete states that are driven by events.In other words, its state evolution depends on the occurrence of discrete asynchronous events over time (Cassandras & Lafortune, 2008).Discrete Event Systems are quite common in the industry nowadays and the events may be classified as uncontrollable and controllable.Examples of controllable events are the start and end of an operation and examples of uncontrollable events are the activation and deactivation of a presence sensor.www.intechopen.com