Fuzzy clustering based models for supervision of industrial processes

The application of fuzzy clustering techniques has recently become in a very useful alternative in the area of modeling and identification of complex industrial process. In particular, fuzzy clustering techniques such as fuzzy c-means and the Gustafson-Kessel (GK) algorithms will be analyzed and applied in details in this paper. These algorithms will be implemented in the construction of Takagi-Sugeno fuzzy models for the gas-liquid separation process and the water-oil separation process, which are important processes in the oil industry. This sort of modeling will be the base for the design of a supervisor control system. Validations of the obtained fuzzy models will be performed and some conclusions will be established.