Sensor Less Vector Control of Induction Motor Based on MRAS Control

Abstract

The implementation of vector control technique needs speed information of motor from the tachogenerators, incremental encoders or resolvers to measure the speed. The speed and flux sensors require careful mounting, alignment and it needs special attention to reduce the noise due to electrical phenomenon. The speed sensors require maintenance and extra space for mounting which increases size and cost of drive system in practical implementation. These problems can be eliminated by speed estimation techniques without using any external sensors in machine. It can be achieved by using Model Reference Adaptive System (MRAS) speed estimation technique. MRAS is popularly used for speed estimation to avoid the sensors like flux and speed. MRAS technique consists of two models named as adaptive and reference models. Moreover, it has one adaptive mechanism PI controller to tune the flux error zero. The error generated from these two models will estimate the speed of the induction machine. The conventional MRAS speed estimation method along with Voltage source inverter normally provides poor steady state response of torque waveform and more distortion/ripples in speed, flux and armature current of induction motor in steady state region. To overcome these drawbacks, anew method of Space Vector Pulse Width Modulation (SVPWM) is introduced here. In the present method, stator flux components are the control variables, and these flux components are resolved into reference voltage vectors based on SVPW Mtechnique. Due to this, the performance of MRAS speed estimator can be improved and speed disturbances are minimized, and also reduce the ripples present in torque and stator currents of the induction motor drive. Also, it has good tracking of rotor speed and reaches to steady state quickly as shown in simulation results carried out by MATLAB/SIMULINK.