DEXTER FUZZY LOGIC CONTROL ARCHITECTURE FOR Â MULTIVARIABLE SYSTEM (DMFLC)
This thesis discusses the extension in the Dexter Controller for Single Input Single Output (SISO) Systems. Dexter controller is based on the philosophy of Model Free Fuzzy Adaptive controller. The main aim of the thesis will be to work on the extension of Dexter SISO controller for the multivariable system. In the Dexter SISO Controller little a priori knowledge is assumed about the plant. An inverse model is developed online using an adaptive scheme. To the best of our knowledge Dexter controller has been successfully tested on a number of SISO plants. In the proposed work the MIMO Dexter will be designed based on Relative Gain Array (RGA) technique. This approach provides an effective methodology for the development of multivariable Dexter fuzzy logic controller. The controller will have minimal prior knowledge of the plant which implies that the coupling amongst the multiple inputs and outputs is not known. The designed controller will be able to offer good control performance in multivariable systems without disturbing the coupling of the MIMO system.