Performance analysis of Adaptive Neuro Fuzzy Inference Systems (ANFIS) for speed control of brushless DC motor

This paper presents a performance analysis of Adaptive Neuro Fuzzy Inference Systems (ANFIS) which is applied to control the speed of Brushless Direct Current Motor (BLDCM). The proposed strategy of speed BLDCM controller is operating which is similar to the conventional direct current motor. This strategy is implemented base on quadratic axis control and zero direct axis current. The aim of the speed controller is to obtain the speed motor operating similar to a speed setting. The input of the controller is speed measured by speed sensor encoder. The output is level voltage supplying stator windings BLDCM through pulse wide modulation controller of inverter. The ANFIS controller is used to decide the pulse wide to of the inverter. It has two input variables speed error, speed error rate and the output variable is the voltage to represent stator current. The variables are represented into fuzzy sets called fuzzy membership functions that their function form and number will be varied until acquired the best characteristic of dynamic response systems. Using MATLAB Simulink, the performance analysis at this strategy is emphasized on the transient parameter of dynamic response. The simulation results show that the best response of speed control is the bell function with five membership functions.