Neuro–Fuzzy Controller and Its Real Time Application

Abstract

Essence of the paper is to combine fuzzy logic and neural networks together, obtaining a robust, hardware friendly neuro-fuzzy controller suitable for real time application. Here, a simple rule base has been implemented on the Xilinx SPARTAN 3AN FPGA development board to bolster the feasibility and superiority of the neuro-fuzzy system (NFS) over fuzzy system (FS). Finally, the neuro-fuzzy controller has been realized to govern a classic control system problem of reference tracking like speed control of a separately excited dc motor using armature voltage control topology. Simulation results along with emulation testing have together concreted the effectiveness and superiority of such hybrid controller in real time applications.