Multivariable Process Control Using Decentralized Neural Fuzzy Controllers
碩士 === 逢甲大學 === 化學工程學系 === 89 === This dissertation develops an intelligent control system for multivariable process systems using decentralized neural fuzzy controllers. With an incorporated static decoupler, the neural fuzzy controllers are able to learn to control the multivariable process adapti...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2001
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/89374421691890123205 |
| Summary: | 碩士 === 逢甲大學 === 化學工程學系 === 89 === This dissertation develops an intelligent control system for multivariable process systems using decentralized neural fuzzy controllers. With an incorporated static decoupler, the neural fuzzy controllers are able to learn to control the multivariable process adaptively by adjusting their membership functions as well as the fuzzy rules. The intelligent control scheme and the related parameter tuning rules derived based on steepest descent algorithm are presented systematically. To show the effectiveness of the intelligent control system, the simulation applications in this work include this control of Wood and Berry distillation column, solid-fuel system and a nonlinear distillation process. Extensive simulation results reveal that the proposed decentralized neural/fuzzy control system appears to be a promising approach to the intelligent control of multivariable process systems, which can provide satisfactory control performance and shorten the design time.
|
|---|