---

K.Naga Sujatha, Dr.K.Vaisakh and Anand.G

Abstract
This paper presents a control scheme combined with neural network, fuzzy controller and PI- controller for the brushless DC motor. The neural network control learned continuously and gradually becomes the main effective control. Performances of the proposed neural network are compared with the corresponding fuzzy PI controller and conventional PI controller. Matlab/simulink software was used to simulate the proposed scheme. Neural network improves speed response and also reduces torque ripples. The simulation results are verified with new control strategy.


INTRODUCTION
ermanent magnet Brushless DC (BLDC) motors are becoming very popular rapidly in industries such as

automotive, aerospace, consumer, medical, industrial automation equipment and instrumentation because of their high efficiency, high power factor, silent operation, compact form, reliability, and low maintenance.

The speed regulators are conventional PI controllers in order to achieve high performance drive. The unexpected change in load conditions or environmental factors would produce overshoot, oscillation of the motor speed, oscillation of the torque, long settling time and causes deterioration of drive performance. In view of improvement in speed response during start up, an abrupt change in command torque and reduction in torque ripple, a kind of neural network which adjusts control rule according to inputs of neural network is presented. Sometimes NN are proved to be more efficient and their performance is less sensitive to parametric variations than conventional controllers. With the learning ability of neural network, neural networks have widely been recognized as a powerful tool in industrial control, commercial prediction, and image processing applications etc. Many authors havehinted the neural networks as powerful building blocks for a wide class of complex nonlinear system control strategies.


For more

---