The integration of renewable distributed generation (DG) in power systems has been increasing day by day. One of the most promising DG technologies is wind turbine among the renewable sources. Therefore, the optimization of DG whose the output power is varying with time is very crucial for the future power systems. However, it is difficult to establish a suitable objective function by taking into account of time varying characteristics. In this paper, a methodology based on weighting factors is proposed in order to minimize energy loss by finding the optimal sizes of wind turbines. The optimization is carried out by using the genetic algorithm with utilizing power flow analysis. The contribution of this paper is to allow considering the time varying characteristics of both load and wind-generation profile in a pairwise manner without violating the harmony of correspondence between load and generation profile. In addition, the proposed methodology is merged with the fuzzy-c means clustering to reduce execution time and allow long term planning due to the fact that the computational burden of the genetic algorithm is substantially high. The proposed methodology is applied to the IEEE-30 bus test system for 4 days and annual energy loss minimization scenarios. The results show that energy loss can be reduced significantly by using the proposed methodology