Chattering-Free Adaptive Wavelet Neural Network Controller Design for a Class of Nonlinear Systems

In order to ensure the conventional adaptive wavelet neural network control systemsʹ stability, a compensator could be designed to dispel the approximation error. The most frequently used of compensator is like a sliding-mode control which maybe cause chattering phenomena to wear the bearing mechanism. To tackle this problem, this paper proposes a chattering-free adaptive wavelet neural network control (CAWNNC) system. The proposed CAWNNC system is composed of a neural controller and a smooth compensator. The neural control uses a fuzzy wavelet neural network to online approximate an ideal controller and the smooth compensator is used to remove the chattering phenomena of conventional sliding-mode control completely. Then, the proposed CAWNNC approach is applied to two chaotic nonlinear systems to investigate its effectiveness. Through the simulation results, the proposed CAWNNC scheme can achieve favorable tracking performance and the convergence of the tracking error and control parameters can be accelerated by the developed PI adaptation learning algorithm.