Stability Constrained Economic Operation of Islanded Droop-Controlled DC Microgrids

This paper proposes a method to determine optimal droop values of dispatchable distributed generation (DG) units in an islanded droop-controlled dc microgrid. Objectives are to minimize fuel cost and maximize small signal stability of the system. In this paper, a detailed state-space model of an islanded droop-controlled dc microgrid including the converter, network parameters, and network interconnection is developed. The developed model is then used to obtain the dominant eigen value of the system. Particle swarm optimization is used to determine the optimal droop values of the dispatchable DG units in order to simultaneously reduce the fuel cost, increase system damping, and push the most dominant eigen value away from the imaginary axis to the left of the s-plane. The bi-objective optimization is solved using a fuzzy max-min based approach. The proposed method is validated on a sample three-bus droop-controlled DCMG test system. Comparison of the result obtained by the fuzzy max-min approach with that of controlled elitist genetic algorithm is also presented in this paper.