Planning and design for microgrid is one of the key technology. Comprehensive optimization (including combination optimization and capacity optimization) and dispatch optimization for distributed generations (DG) are studied. Based on the characteristics of DGs, a two-stage optimal planning and design model for microgrid is presented, which can apply to both stand-alone microgrid and grid-connected microgrid. For the first stage, multi-objective genetic algorithm based on non-dominated sorting genetic algorithm II (NSGA-II) is used to calculate the optimal system configuration, while the mixed integer linear programming (MILP) is used to deal with the dispatch optimization in the second stage. Using the presented model, a case study is made for stand-alone microgrid and grid-connected microgrid respectively, and the validity of the method is verified.