1996 Fiscal Year Final Research Report Summary
Development and Application of the Polymerization System with Fuzzy Temperature Control for Highly Viscous and Eexothermal Polymerization
| Project/Area Number |
07555550
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (A)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 試験 |
|---|
| Research Field |
反応・分離工学
|
|---|
| Research Institution | Yokohama National University |
|---|
Principal Investigator |
KAMIWANO Mitsuo Yokohama Nat.Unv., Faculty of Engineering, Professor, 工学部, 教授 (70017896)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
TARUMOTO Jun Soken Chem.& Eng.Co.Ltd., Research Center, Investigator, 研究開発センター, 副主任研究員
YAMAMOTO Kazumi Soken Chem.& Eng.Co.Ltd., Research Center, Head Investigator, 研究開発センター, 主任研究員
NISHI Kazuhiko Yokohama Nat.Unv., Faculty of Engineering, Assistant, 工学部, 助手 (20262412)
KAMINOYAMA Meguru Yokohama Nat.Unv., Faculty of Engineering, Assistant Professor, 工学部, 助教授 (50233945)
|
|---|
| Project Period (FY) |
1995 – 1996
|
| Keywords | polymerization system / fuzzy control / model prediction control / viscous fluids / high exothermal polymerization / Stirred tank reactor / numerical simulation |
|---|
| Research Abstract |
In this study, We developed a batch polymerization reactor which have five cooling coils in the vertical direction. There are sixteen thermocouples to measure temperature distribution in the reactor. The temperature value from each thermocouples are sent microcomputer. The microcomputer calculated operation values for each cooling coils based on a suitable control theory, which is 'fuzzy' or 'model prediction' etc..
We investigated the relationship temperature distribution in the polymerization reactor and the properties of produced polymers at first.
As a result, We found that on-off control reactor and PID control reactor could not temperature control when reaction fluid was high viscosity. But the fuzzy control reactor and the model prediction reacter had control capacity through the reaction time. And produced polymers from the fuzzy control reactor and the model prediction reactor had a narrower molecular weight distributions then the other control reactor. In the case of compared the fuzzy control reactor with the model prediction control, the latter one was better. Because model prediction control was not necessary adjustment of fuzzy constants which is indispensable operation in fuzzy control.
In this study, we developed the numerical simulator for batch polymerization reactor with several control system. This simulator was calculated three dimensional velocity, temperature distribution, monomer conversion distribution in reactor and molecular weight distribution of produced polymer at any time.
|
