Simultaneous output feedback robust model predictive control and state estimation for DC microgrids

DC microgrids (DC MGs) are recently more visible in power networks. One related issue is the existence of Constant Power Loads (CPLs), which degrade the performance of a DC MG due to their negative impedances. In practice, the DC MGs also feed time-varying and/or uncertain CPLs (i.e., non-ideal CPLs). Besides, the uncertainties in the system model can influence the stability of the system. This paper introduces an output feedback control approach for the stabilization of DC MGs with parametric uncertainties, non-ideal CPLs, unmeasured disturbance, and measurement noise. The control scheme combines Robust Model Predictive Control (MPC) and state estimation using the Moving Horizon Estimation (MHE) method into a single min–max optimization. The state estimation is based on measured capacity voltages contaminated with measurement noise. In this method, physical constraints are also considered. Furthermore, a novel δ-active set primal–dual interior point method (IPM) as a fast solver to compute the objective function is employed and it is shown to be faster than the primal–dual IPM. The comparison and simulation results are promising and show the robustness and reliability of the proposed approach.