An adaptive identification and control scheme using radial basis function networks

In this paper adapt ive identification and control of non linear dynamical systems are investigated using radial basis function networ ks (RBF). Firstly, a novel approach to train the RBF is introduced. which employs an adaptive fuzzy' generalized learning vector quantization (AFGLVQ) technique and recursive least squares aigorit hm with variable forgetting factor (VRLS). The AFGLVQ adjusts the centers of the RBF while VRLS updates the couneccion weights of the network. The identification algor ithrn has the properties of rapid convergence and pF.-rsistent adaptability that make it suitable for real-tune control. Secondly, on the basis of the one-step ahead REF predictor. the control law is optimized iteratively through a numerical stable Davidon's least squares-based (SDLS) minimization approach. Four nonlinear examples are simulated to demonstrate the effectiveness of the identification and control algorithms.