Author :
Almeida, E. Emanuel ; Luntz, Jonathan E. ; Tilbury, Dawn M.
Abstract :
The contribution of this paper is the introduction of the event-condition-action (ECA) paradigm for the design of modular logic controllers that are reconfigurable. ECA rules have been used extensively to specify the behavior of active database and expert systems and are recognized as a highly reconfigurable tool to design reactive behavior. This paper develops a method to design modular logic controllers whose dynamics are governed by ECA rules, with the ultimate goal of producing reconfigurable control. Modularity, integrability, and diagnosability measures that have in the past been used to measure the reconfigurability of manufacturing systems are used to assess the reconfigurability of the developed controllers. For the modularity measure, criteria found in computer science to evaluate the modularity of object-oriented programs are adapted to evaluate the modularity of modular logic controllers. The results of this paper are that reconfigurability is highly dependent on the level of modularity of the logic control system, and that not all "modular" structures are reconfigurable. There are approaches, such as the one shown in this paper using ECA rules, that can greatly increase the modularity, integrability, and diagnosability of the logic control system, thus increasing its reconfigurability. Note to Practitioners-This paper has been motivated by the problem of designing reconfigurable modular logic controllers. Reconfiguration is important in manufacturing, but it has also been an issue in the software design domain. There are software systems that currently exist, such as active data bases or expert systems with very powerful reconfiguration capabilities enabled by event-condition-action (ECA) rules. This paper applies the ECA concept to the design of modular logic controllers. This paper begins by describing what an ECA logic system is and then focuses on how ECA logic systems can be implemented with modular control approaches. To this end, two designs are - - considered. First, modular finite state machines are used to construct ECA logic systems, and a theoretical framework is built using this approach. Three qualitative measures for reconfigurability (modularity, integrability, and diagnosability) are presented and the controllers are evaluated using these measures. Second, an implementation using the IEC 61499 function block standard is presented as it is a widely understood and accepted standard for modular control applications. Future work entails theoretical analysis using modular verification techniques that exploit a controller structure
Keywords :
IEC standards; active databases; control system CAD; discrete event systems; expert systems; finite state machines; flexible manufacturing systems; object-oriented programming; process control; production engineering computing; software engineering; IEC 61499 function block standard; active database; computer science; diagnosability measures; discrete event systems; event-condition-action system; expert systems; integrability measures; modular finite state machines; modular logic controllers design; modular verification techniques; object-oriented programs; reconfigurable logic controllers; reconfigurable manufacturing systems; rule-based systems; software design domain; Computer science; Control systems; Databases; Design methodology; Diagnostic expert systems; IEC standards; Logic design; Manufacturing systems; Reconfigurable logic; Software design; Data bases; discrete event systems (DES); finite state machines; logical control; rule-based systems;