Robust Cooperative Predictive Control of DC Microgrids with Constant Power Loads

The proportion of constant power loads (CPLs) in DC microgrids is gradually increasing. However, because the point of load (POL) converter is regulated strictly, the negative input resistance character of CPLs leads to the reduction of system damping which can cause system instability. In this work, an effective robust cooperative predictive control method for DC microgrids is newly proposed in the case of large-scale constant power loads (CPLs) access, which can realize simultaneously strong-robust operation, accurate voltage regulation and power sharing of DC microgrids. First, the mathematical model of a bidirectional converter with CPLs is established and the instability trajectory of the phase plane is analyzed. Then, the instability factor is defined innovatively to quantity the degree of system instability caused by CPLs and its predictive model is derived. At last, the cost function including voltage regulation, power sharing and instability factor is constructed. The optimal control variable is obtained by rolling optimization solution which will regulate the voltage set point for droop control. Simulation results prove that the proposed method can maintain the stability of the system under the power of CPLs reaches 500KW.