In this study, the integral linear quadratic regulator (LQR) with servomechanism for machine-side power converter in PMSG-based variable-speed wind energy conversion systems (WECSs) has been proposed. The solution of the algebraic Riccati equation (ARE) has been found for the extended dimension of the state space equation of the system. The state vector has been extended with the integral of the angular shaft speed of the permanent magnet synchronous generator (PMSG) to penalize the errors. The maximum power tracking point (MPPT) algorithm is achieved by minimizing tracking errors between the angular shaft speed reference based on wind speed estimation and its actual values in the variable speed WECS. Also, the estimated aerodynamic torque is used to define the reference electromagnetic torque. This is possible when WECS is partially loaded and pitches angles are fixed at the position to generate maximum power. The mean absolute percentage error of the angular shaft speed of the PMSG-based WECS has been reduced by more than 71% under model uncertainty and noise presented case than in the traditional disturbance observers-based compensation scheme. While the disturbance observers for estimation model uncertainty are eliminated, the use of the high order disturbance observer for aerodynamic torque estimation proved to be necessary to enhance the reliability of wind speed sensors and hence the whole WECS.