Adaptive sliding-mode type-2 neuro-fuzzy control of an induction motor

Abstract

An innovative adaptive control method for speed control of induction motor based on field oriented control is presented in this paper. The fusion of sliding-mode and type-2 neuro fuzzy systems is used to control this system. An online learning algorithm based on sliding-mode training algorithm, and type-2 fuzzy systems is employed to deal with parametric uncertainties and disturbances, by adjusting the control parameters. The sliding-mode adaptive mechanism tune the parameters of type-2 membership functions (antecedent part) and the consequent part parameters, according to the inputs: speed error and its derivative, in structure of type-2 neuro fuzzy system. Since the parameters of the induction motor may vary, and the information that is used to construct the membership functions and the rules of fuzzy logic system is uncertain, type-2 neuro fuzzy structure is selected as the controller. The results obtained by using this approach are compared with those of type-1 counterpart. The proposed adaptive sliding-mode type-2 neuro-fuzzy controller can control the induction motor with higher performance as it is compared with type-1 neuro-fuzzy systems while it shows more robustness to variations in the parameters and measurement noise.