Design of Fuzzy logic based level controller for surge tank system

This paper presents the design of fuzzy logic controllers (FLC) for a highly nonlinear dynamic system of Surge Tank. The controllers are designed from two approaches: model free (Mamdani) and model based (Takagi-Sugeno). For model free approach, Mamdani inference engine is formed from the expert knowledge of the system using which a fuzzy PD controller is synthesized. For model based approach, first the Takagi-Sugeno fuzzy modeling of the real system is performed; with the proposed fuzzy system having fixed structure that is linear in parameters. Fuzzy clustering with optimal output predefuzzification method has been employed for the formation of rule base as well as for the parameter identification of the proposed fuzzy system. Then a state-feedback fuzzy controller is presented for the TS model. The performance of both the FLCs is compared with each other and also with that of a linear PID controller. Finally the merits and demerits of these controllers are discussed.