Smart Grid will be a new way to the development of the power network. In the development of Smart Grid, how to integrate the Renewable Energy and the Distributed generation (DG) safely and reliably is a problem urgent to be solved. Siting and sizing of DG planning in Smart Grid construction are researched in this paper. A multi-objective optimal model is established and normalized in the DG number, location and individual capacity of uncertainty. The improved adaptive genetic algorithm is used to solve this optimization problem. Finally, the validity of the method proposed in the paper is proved by an example.