Further Study on the Parameter Convergence of Fuzzy Models in Nonlinear System Identifications
非线性系统辨识中模糊模型参数收敛问题的进一步研究

This paper investigates the persistent excitation conditions under which the parameters in the fuzzy system model converge to their true values when the standard Mamdani type fuzzy system is constructed and the orthogonal projection parameter-tuning algorithm is used for nonlinear system identi-fication. Algorithms are proposed accordingly for generating the input signals with persistent excitation property for the identi-fications of nonlinear moving average (N-MA) and second-order nonlinear auto-regressive moving average (N-ARMA) systems.