Optimal MRAS speed estimation for induction generator in wind turbine application

Abstract

This paper propose improved sensorless vector control of squiral cage induction generator for variable speed wind energy conversion system. The rotational speed of the induction generator is estimated with the Model Reference Adaptive System - MRAS observer. The estimated rotational speed is used as the feedback of the control loop in the converter control system. Proportional integral controller in the MRAS observer is optimized via Genetic Algorithm, Partical Swarm Optimization and Simulated Annealing. Comparative analise of the optimal speed estimation of induction generator is also presented. The performance of the sensorless controled variable speed wind turbine drive is evaluated through simulation in Matlab/Simulink. Experimental results are gained via laboratory model based on dSpace DS1104 digital control card. © 2010 IEEE.