Modeling and Control of Multivariable Process Using Intelligent Techniques
Sensors & Transducers. 2010;121(10):68-76
Journal Title: Sensors & Transducers
ISSN: 2306-8515 (Print); 1726-5479 (Online)
Publisher: IFSA Publishing, S.L.
Society/Institution: International Frequency Sensor Association (IFSA)
LCC Subject Category: Technology: Technology (General)
Country of publisher: Spain
Language of fulltext: English
Full-text formats available: PDF
Subathra Balasubramanian (Department of Chemical Engineering, National Institute of Technology, Tiruchirappalli-620015, Tamil Nadu, India)
Radhakrishnan T. K. (Department of Chemical Engineering, National Institute of Technology, Tiruchirappalli-620015, Tamil Nadu, India)
Abstract | Full Text
For nonlinear dynamic systems, the first principles based modeling and control is difficult to implement. In this study, a fuzzy controller and recurrent fuzzy controller are developed for MIMO process. Fuzzy logic controller is a model free controller designed based on the knowledge about the process. In fuzzy controller there are two types of rule-based fuzzy models are available: one the linguistic (Mamdani) model and the other is Takagi–Sugeno model. Of these two, Takagi-Sugeno model (TS) has attracted most attention. The fuzzy controller application is limited to static processes due to their feedforward structure. But, most of the real-time processes are dynamic and they require the history of input/output data. In order to store the past values a memory unit is needed, which is introduced by the recurrent structure. The proposed recurrent fuzzy structure is used to develop a controller for the two tank heating process. Both controllers are designed and implemented in a real time environment and their performance is compared.