The current distribution grids incorporate generators and loads dependent on stochastic variables that generate uncertainty, an example of this is wind speed or solar irradiation. (For wind or photovoltaic generators respectively). Uncertainty represents a challenge in scheduling the power to be generated or demanded, since there is no certainty in estimating the specific cost. In effect, uncertainty has been associated with penalty cost [1], which occurs when the demand aggregator assumes a different power value to be dispatched. Uncertainty cost have been defined as mathematical tool that model stochastic behaviors through probability density functions, that allow to calculate the expected value of penalty cost to be calculated effectively. Smart Distribution Grids must manage bi-directional power flows that depends of stochastics behaviors through information and communication technologies. Its optimal operation consists of minimizing cost and maximizing profits in the exercise of buying and selling energy in 24-hour periods. For this propose, metaheuristic optimization algorithms have been used, because it is necessary to consider discrete, continuous, binary variables and nonlinear equations. The VNS-DEEPSO algorithm was sown to obtain the best performance in the IEEE-PES WCCI 2018 [2], obtaining of 7% in total average grid operating costs compared to the second algorithm. In this document the optimal power scheduling of a smart distribution grid is carried out in eight different scenarios, considering the uncertainty cost for photovoltaic, wind generator and connection nodes of electric vehicles. It was obtained that by replacing a conventional distributed generator with a sola generator with the maximum power capacity of 20% of the grid, a reduction in the total cost of optimal scheduling can be achieved by 40.16%. However, under this same scheme, but evaluating the impact of wind generation, an increase of 9.16% was reached over total costs. Therefore, the taking in count of eh uncertainty cost associated with each type of charging or generation technology has a considerable impact on the optimal scheduling of an smart distribution grid, which requires a specific analysis for each case depending on its generators, loads and environmental factors, as is exposed in the present study.