On-line identification of nonlinear systems using Volterra polynomial basis function neural networks

An on-line identification scheme using Volterra polynomial basis function ( VPBF) neural networks is considered for nonlinear control systems. This com prises a structure selection procedure and a recursive weight learning algo rithm. The orthogonal least-squares algorithm is introduced for off-line st ructure selection and the growing network technique is used for on-line str ucture selection. An on-line recursive weight learning algorithm is develop ed to adjust the weights so that the identified model can adapt to variatio ns of the characteristics and operating points in nonlinear systems. The co nvergence of both the weights and the estimation errors is established usin g a Lyapunov technique. The identification procedure is illustrated using s imulated examples. (C) 1998 Elsevier Science Ltd. All rights reserved.