Modeling the MIMO Nonlinear Temperature System in Ceramic Roller Kiln Using Neural Networks and Neuro-Fuzzy System

摘要

Artificial Neural Networks (ANNs) and Neuro-Fuzzy Systems (NFSs) have been widely used in modeling and control of many practical non-linear process, however, most of them have focused on single output systems. In this paper, we would like to demonstrate a comparitive study between applications of ANNs and Coactive Neuro-Fuzzy Inference System (CANFIS) in modeling a Multi-Input - Multi-Output (MIMO) nonlinear temperature system of roller kiln in ceramic tile manufactering industry. From that, CANFIS was selected as a better approach in this case.