Research On Modeling, Scheduling And Controller Of Reconfigurable Manufacturing System Using Petri Nets

A reconfigurable manufacturing system (RMS) is designed at the outset of rapid change in structure in response to multi-type, varied-loting, multi-function, quick delivery, sudden changes in market. As an advanced manufacturing system, RMS has been increasingly paid attention to, and has been carried out deep research in the area of theory and applications. Because of the new characteristics such as modularity, integrability, customization, convertibility and diagnosability, the development of the suited key enabling technologies is considered as one of focuses in the field. The modeling, scheduling and design of controller are the key theories and methods for improving the management, optimization and automation level of manufacturing system. Therefore, their research and application results mark the level of manufacturing system. In this dissertation, the methodology of the modeling, scheduling and controller design of RMS are presented based on its characteristics using Petri nets which is a powerful formalized tool for DEDS. Moreover, the mass numerical simulations and comparison study are investigated. Our results promote the research of RMS, and help to solve the related problems as wells.The main contributions in this dissertation are summed up as follows.(1)The framework and key enabling technologies are presented. By analyzing the demands of multi-type and varied-loting production mode, the definition, characteristics and components of RMS are described. Moreover, its structure and design principle are presented. Then all the issues related to key enabling technologies to construct an integrated and feasible RMS are proposed and detailed discussion including configurable planned layout and optimizing, system modeling technology, scheduling technology, reconfigurable machine tool design, modularized controller design, stitching material flow technology, interface configuration principle and system quick diagnosis technology. The implementation scheme is also illustrated.(2)A modular modeling approach based Extended Stochastic Petri nets (ESPN)...