ADAPTIVE TREE-STRUCTURED SELF-GENERATING RADIAL BASIS FUNCTION NETWORK AND ITS PERFORMANCE EVALUATION

Several algorithms have been proposed to identify a large scale system , such as the neuro-fuzzy GMDH, and the fuzzy modeling using a fuzzy n eural network. As another approach, Sanger proposed a tree-structured adaptive network But in Sanger's network, it is not clear how to deter mine the initial disposition of bases and the number of bases in each subtree. We propose a nonlinear modeling method called the adaptive tr ee-structured self-generating radial basis function network (A Tree-RB FN). In A Tree-RBFN, we take the maximum absolute error (MAE) selectio n method in order to improve Sanger's model. We combine Sanger's tree- structured adaptive network for an overall model structure with the MA E selection method for a subtree identification problem. In ATree-RBFN , the tuning parameters are not only the coefficients but also the cen ters and widths of bases, and a subtree can be generated under all lea f nodes. Then, the input-output data can be divided into the training data set and the checking data set, and an element of inputs in each s ubtree is selected according to the corresponding aror value from the checking data set. We also demonstrate the effectiveness of the propos ed method by solving several numerical examples.