Intelligent control of a DC motor using a self-constructing wavelet neural network

This paper proposes an intelligent method to control the speed of a DC motor. This controller is a self-constructing wavelet neural network (SCWNN) in which the self-constructing and training algorithms are simultaneously performed. At first, there are no wavelets in the wavelet layer; they are automatically generated in the online control process. In order to increase the convergence speed of the proposed controller, adaptive learning rates (ALRs) updated at each sampling time are used. In the online control process, no identifier is used to approximate the dynamic of the controlled plant, because of the learning ability of the proposed controller. Several simulations are used to demonstrate the effectiveness and adaptiveness of SCWNN.