Experimental study of an interval type-2 fuzzy neural network using sliding-mode online learning algorithm

In industry, fuzzy controllers and fuzzy sliding mode controllers have a variety of applications to handle uncertainties in control systems. Even though type-1 fuzzy sets are generally used to construct the membership functions of these controllers, through their capability to better model uncertainties type-2 fuzzy logic systems may provide more accurate results. The use of a sliding mode control-based algorithm for the training of type-2 fuzzy neural networks makes this paper the first study in the concerned field to work in a real-time control application. Instead of trying to minimize a cost function, the learning parameters are tuned so that the error is enforced to satisfy a stable equation. The parameter update rules for a type-2 fuzzy neural network with two inputs and one output are derived, and the stability of the proposed learning algorithm is proven via a Lyapunov function. This algorithm is applied to a real-time laboratory setup with time-varying and nonlinear load conditions. The real-time results demonstrate the effectiveness of the proposed method in controlling the servo system under the disturbances and uncertainties.