A Fast Learning Algorithm for Wavelet Network and its Application in Control

A novel control approaches based on the wavelet network is proposed to control the nonlinear system. To accelerate the convergence of the wavelet network, the hybrid learning algorithms is presented, which consists of the BFGS learning algorithms plus the least squares algorithm. The hybrid learning algorithm is compared with the gradient algorithm and with variable learning rate back-propagation, the results indicate that proposed algorithm is much more efficient than either of the other technique. The proposed strategy was applied to identify and control a nonlinear system, the effectiveness of the proposed control scheme is verified by simulated results.