STRUCTURE SELECTIVE UPDATING FOR NONLINEAR MODELS AND RADIAL BASIS FUNCTION NEURAL NETWORKS

Selective model structure and parameter updating algorithms are introd uced for both the online estimation of NARMAX models and training of r adial basis function neural networks. Techniques for on-line model mod ification, which depend on the vector-shift properties of regression v ariables in linear models, cannot be applied when the model is non-lin ear. In the present paper new methods for on-line model modification a re developed. These methods are based on selectively updating the non- linear model structure and therefore lead to a reduction in computatio nal cost. A real data set is used to demonstrate the performance of th e new algorithms. (C) 1998 John Wiley & Sons, Ltd.