This paper proposes a stochastic optimization model considering the volatility of wind power and photovoltaic power in microgrid. The model optimizes the economic operation of a microgrid as well as minimizes the flow deviation at the point of common coupling (PCC) from scheduled values. First, the Latin hypercube sampling (LHS) and simultaneous backward reduction (SBR) technique are introduced to describe the stochastic nature of wind power and photovoltaic power. Then the random characteristic is introduced into the objective function of the stochastic model which is solved based on the genetic algorithm (GA). Simulation results demonstrate its rationality and effectiveness for day-ahead scheduling of a microgrid.