1989 Fiscal Year Final Research Report Summary
A Study on FMS Dynamic Modeling by Petri Net
| Project/Area Number |
62550084
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
機械工作
|
|---|
| Research Institution | Hokkaido University |
|---|
Principal Investigator |
KAKAZU Yukinori Faculty of Engineering Hokkaido University, Professor, 工学部, 教授 (60042090)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
OKINO Norio Faculty of Engineering Kyoto University, Professor, 工学部, 教授 (30001093)
|
|---|
| Project Period (FY) |
1987 – 1988
|
| Keywords | Petri Net / FMS / Design Automation / Parallel Distributed Processing / CAD / CAM / CAM / Neural Network / Connectionist Model |
|---|
| Research Abstract |
In this project, we aimed at constructing an FMS control model which is capable of representing and controlling dynamic behavior of FMS. In realizing this model, we found some interesting new problems as well as previously expected ones. Through our approach to these problems, we obtained some new views and theoretical results to the proposed study area. We summarize our study results as follows.
(1) To realize the dynamic modeling, various types of FMS has to be considered. (2) To cover such variety and to discuss FMS theoretically, generalization ( or formalization of FMS has to be introduced, and design process control problem gives useful analogy to the proposed problems. (3) To model process from raw materials to finished products, unified 3 dimensional geometric model has to be developed. (4) To realize FMS having expected flexibility, the model has to have capability of expressing dynamic state space. (5) To describe ( or express ) FMS having dynamic state space, we need extended Petri Net and , as further extended one, Connectionist Model. (6) we give new definitions and theorems to the extended Petri Net and, based on these, new algorithms are developed and applied. (7) As a extended type of Connectionist Model, we develop multi-layer connectionist model. (8) To construct a dynamic scheduling model for time-varying FMS, an optimizing method using Hopfield type neural network was developed. (9) We attempted to find some similar mechanism between natural system and production system.
Finally, we are expecting that these study results may give a basis for development of IMS( Intelligent Manufacturing System).
|
