Energy management of residential microgrids using random drift particle swarm optimization

The energy management problem of residential microgrids is studied in this paper. Both the grid-connected and islanded modes of operation are considered. The problem has a dynamic nature due to the existence of the energy storage devices in the microgrid components. In addition, the inclusion of deferrable loads introduces integer decision variables to the problem formulation. Consequently, the problem is formulated as a mixed integer dynamic optimization problem in the case of the grid-connected mode of operation. In order to solve the problem, a powerful variant of the particle swarm optimization algorithm known as the random drift particle swarm optimization is adapted in this paper. The results of applying this variant have been compared with those of two state of the art optimization techniques, namely the particle swarm optimization and the differential evolution algorithm. The results of the comparison show that a remarkable saving in the operational cost of residential microgrids can be achieved through utilizing the random drift particle swarm optimization technique for solving the problem.