Efficient Sensorless Speed Control of Induction Motors Using Hybrid Grey Wolf Optimizer and Neural Network

Abstract

The article addresses an efficient technique to drive a sensorless three phase induction motor with any desired speed and at different loading conditions by controlling both applied voltage and frequency. An artificial neural network (ANN) is developed to instruct the inverter frequency at the required reference speed to achieve maximum efficiency regardless of load torque. The applied voltage is controlled by a Proportional-Integral (PI) controller to drive the motor at the required reference speed and load torque. Speed of motor is estimated by using a model reference adaptive system (MRAS) without mechanical speed sensor to decrease the cost and increase system reliability. Grey wolf optimizer (GWO) is applied to generate the optimal parameters of a MRAS PI controller. Error between reference and actual motor speeds is minimized by optimizing the parameters of a voltage PI controller using GWO. To endorse the significance of the proposed methodology, it is compared with the classical V/f speed control method. The realized results by the proposed method support its applicability to attain maximum efficiency at varies operating conditions. The system is simulated in MATLAB/Simulink environment.
Copyright © 2016 Praise Worthy Prize - All rights reserved.