Mahdi Hadian - Department of Industrial Engineering Amirkabir University of Technology (Tehran Poly technique)، Azarab Industries Co.
Hosein Fazel Zarandi - Department of Industrial Engineering Amirkabir University of Technology (Tehran Poly technique)
Mahdi Rezaei - Azarab Industries Co.

An indirect method or fuzzy clustering algorithm is selected to build a linguistic fuzzy model of a steam generator controller. Based on output data clustering and projection onto the input spaces, number of clusters is determined and rules are generated automatically and a linguistic fuzzy MISO model is constructed. Then, by implementing an adapted neural Network for selected type of membership functions, all membership function parameters will be optimized. Compared to other algorithms, the proposed new algorithm and methods are computationally and conceptually simple. The obtained model will be compared with a conventional (PID) controller for a specified boiler.

Steam generator, Fuzzy control, Fuzzy clustering, Parameter identification, Neural networks