Fast fuzzy orthogonal least square algorithm. (Q2785171)

The authors give some improvements to the well-known orthogonal least squares algorithm for NARMAX models of a large class of nonlinear systems. By introducing fuzzy basis functions, which are defined in the paper for a specific fuzzy logic system, they develop the so-called fast fuzzy orthogonal least square algorithm.NEWLINENEWLINE This new algorithm can make use of language information and data information simultaneously.NEWLINENEWLINE A method is indicated to accelerate the calculation of the algorithm proposed. A simulation example on a glass furnace is also given to show the validity of the algorithm suggested herein.