The Experimental Implementation of an Energy Management System for a Grid Connected Microgrid by using a Multi-period Imperialist Competition Algorithm

The improvement of flexibility, optimum using of power generation units with less environment affects are increased over and over noting the development of the Microgrid (MG) in power systems. An optimum energy management system (EMS) has special importance issues in order to generate power with the least possible production cost. The main duty of the proposed EMS can be attributed to the optimum management of the generation units, demand side management and power exchange with the national grid. Because the problem considered for the MG application has a nonlinear and discrete nature, so a heuristic algorithm called multi-period imperialist competition algorithm (MICA) is used to implement an optimum EMS. The fulfillment of the load requirement, the technical specifications related to each of the resources and the national grid constraints are included in the discussed problem. Furthermore, several scenarios are applied to evaluate the capability of the proposed algorithm then the obtained results are compared to those found with an EMS based on mixed integer non-linear programming (MINLP), multi-period artificial bee colony (MABC) (EMS-MABC) and particle swarm optimization (PSO) (EMS-PSO) approach. The obtained results demonstrate the efficiency of the proposed algorithm for complete supplying load demand, total production cost reduction (about 17%, 10% and 22% respectively) also the reduction of MCP relative to EMS-MINLP, EMS-MABC and EMS-PSO algorithms.