Improved Stabilization of Nonlinear DC Microgrids: Cubature Kalman Filter Approach

Abstract

This paper investigates the injecting power stabilization of nonlinear dc microgrids (MGs) with constant power loads (CPLs). By considering a centralized controller scheme, the limitations of the communication utilities are considered. Therefore, limited information is transferred through the nonideal noisy communication network. Consequently, a cubature Kalman filter (CKF) with a third degree is proposed to mitigate the effect of the noisy measurement and the noisy network on the system's information. Moreover, an estimation-based robust feedback controller is developed to design an optimal value for the injecting power. The considered CKF algorithm is robust against the system uncertainty and noisy environments and has a low computational time for high-order dc MGs with a high number of sources and CPLs. In addition, a systematic procedure to compute the feedback gain of the controller is presented, which can be numerically solved by linear matrix inequality techniques. Hardware-in-the-loop real-time simulation results verify the simplicity of the controller implementation, enhanced performance for the case of limited information, and better robustness against the noisy measurements compared to the state-of-the-art methods.