A hybrid analytical/intelligent approach to the modeling and control of DEDS

Abstract

Many manufacturing processes, such as those arising in electronic circuit assembly and packaging, are characterized by both continuous variables and discrete event dynamics. The design of innovative end effective monitoring and control strategies for these processes requires the development of new models that can capture the hybrid continuous/discrete dynamic nature of the processes and can address the performance objectives in a systematic way. Although then, exists an extensive theory of modeling and control for continuous variable dynamic systems (CVDS), as of yet there is no comparable counterpart for the study of discrete event dynamic systems (DEDS), and there is no general systematic approach for combining CVDS and DEDS methodologies for the study of hybrid processes such as those arising in manufacturing systems. This paper introduces a new methodology for modelling, simulation and control of hybrid CVDS/DEDS processes with application to manufacturing systems. A key feature of the proposed approach is the use of a hierarchical modelling philosophy that capitalizes upon control theoretic concepts and tools from the area of fuzzy logic and neural networks. >