Robust Feedback-Linearization Technique for Grid-Tied LCL Filter Systems Using Disturbance Estimation

In this paper, feedback linearization (FBL) technique together with disturbance observer approach is proposed to mitigate the effect of the resonant frequency of grid-tied \boldsymbol{LCL} filter systems. The state-feedback control law is employed to achieve stabilization of the \boldsymbol{LCL} filter system under a wide range of resonant frequency variation. The disturbance observer (DO) is designed to counteract the effect of model uncertainty and unknown disturbance aiming to achieve asymptotic stability under feedback linearization control. Specifically, the observer is designed to estimate an additional input, representing uncertainty and unknown disturbance, from measurable variables and then the state-feedback control utilizes the disturbance estimate to compensate for its effect. An interesting feature of the composite controller is its ability to achieve good tracking performances even in the presence of model uncertainty and external disturbance. The composite controller was implemented for experimental evaluation and performance testing. High performance with respect to disturbance rejection and parameter variation has been demonstrated.