An Adaptive Parameters Learning Algorithm for Fuzzy Logic Systems
Guanzhong Dai · Acta Simulata Systematica Sinica · 2004
According to the theories of adaptive fuzzy logic system and general parameter optimization, the characteristics of normal parameter-learning algorithms on fuzzy logic systems are firstly analyzed, and its some defects are discussed briefly. Aimed at improving parameter-training pace, a new adaptive parameter-learning algorithm is then advanced, which is deduced and explained in detail. Lastly, the proposed algorithm has been evaluated by a nonlinear function to approximate its ideal values; system抯 simulation precision and its rate of convergence are enhanced remarkably. Those results demonstrate the efficiency and feasibility of the proposed adaptive optimization algorithm, which can overcome some deficiencies of normal parameter-learning algorithms.