A NOVEL RECURRENT TYPE-2 FUZZY NEURAL NETWORK FOR STEPPER MOTOR CONTROL, 30-35.

Jafar Tavoosi

Keywords

Recurrent type-2 fuzzy, inverse control, stepper motor

Abstract

In this paper, a new fuzzy neural structure is proposed for controlling the experimental stepper motor. The structure of the new fuzzy neural network is very simple and has three layers. The first layer is the fuzzy operation, the second layer is the fuzzy rules’ layer and the feedback is formed around the neuron in this layer, and finally in the third layer, the type-2 fuzzy rules’ output with general feedback is calculated. This new structure is used as an adaptive inverse control to precisely control a stepper motor. The methodology is as follows: first, the inverse of the stepper motor is identified by the recurrent type-2 fuzzy neural network, and then the obtained recurrent type-2 fuzzy system is used as a controller for the stepper motor. The experimental results show that this method can provide an appropriate response to changes in load torque and motor parameters.

Important Links:

Go Back