Despite all achievements of the microgrids, planning a cost-effective structure is a complicated process due to the different parameters that should be taken into account at any decision level. This chapter will focus on recent progresses in the application of computational intelligences and heuristic techniques in Microgrids that can cover different challenges of this area including the sizing and management optimization, predictive maintenance, estimation of exploitable energy, real-time self tuning system, siting, operation scheduling, and variety of other applications. Using these methods leads to a reliable network partitioning with less CPU effort and save the operating costs of the distributed nodes and interruption costs. Also, they provide more mobility to add extra restrictions to the aforementioned issues such as sizing and scheduling of power generation sources. To demonstrate the efficiency of the evolutionary algorithms (EAs), a comparison is carried out between the EA methods and conventional energy management system such as interval linear programming and mixed-integer.