A self-learning fuzzy control method based on RBF neural networks

This paper proposes an self-learning fuzzy control method based on an improved radial basis function neural networks (RBFNN). The architecture of the proposed approach is comprised of a fuzzy controller and an RBFNN. For such an architecture, firstly, an analytical formula is employed to design fuzzy controller. Then, RBFNN based on an efficient locally regularized forward recursive (LRFR) algorithm is described and employed to learn the model of the plant. Finally, the parameters of fuzzy controller are tuned online by self-learning algorithm based on RBFNN. The simulation studies for a heating, ventilation and air-conditioning (HVAC) system demonstrates the validity and performance of the proposed learning algorithm.