Online Power Estimation of non-Ideal CPLs in Shipboard DC MGs using Cubature Kalman Filter

DC microgrids (MG) used in DC shipboard power systems (SPS) involve the integration of power electronic loads, which might behave as constant power loads (CPL). The incremental negative impedance of CPLs might degrade system stability. To ensure an effective control of DC SPSs, eliminating the undesired behavior of CPLs is a necessity. This aim requires the spontaneous power value of the time-varying uncertain CPLs. Since integration of current sensors in DC MGs is costly and the resistance of these sensors degrades ripple filtering, estimation methods should be used to obtain the instantaneous power of CPLs. In this paper, a 3rd degree cubature Kalman filter (CKF) is developed to estimate the power of the CPLs alongside estimation of CPLs' and source's currents in a DC MG. By considering the CPLs powers as artificial states and augmenting them into the system states, not only the DC MG states but also the unknown values of the CPLs powers may be estimated. The proposed estimator is tested on a DC MG that feeds one CPL. The experimental results show that the proposed CKF is indeed able to estimate instantaneous power consumption of the CPL as well as source and load currents.