A stochastic programming model for monthly power dispatch associated with the risk of energy conservation is proposed for dealing with an issue that the determined method of energy conservation can’t be used for the risk management of energy conservation scheduling under uncertain environments. First of all, a benefit index of electricity conservation per unit for the power dispatch units with energy conservation is established on the specific math functional relationship between unit’s power-dispatching and its coal consumption rate; the uncertain factors of random variation of load and wind power have significant effects on the benefit evaluation of electricity-conservation, and a risk index of electricity-conservation is established based on the concept of semi-absolute deviation from economy. Furthermore, a monthly power dispatch associated with the risk of energy conservation is developed in the framework of stochastic programming. In order to reach a balance of contradiction between model-solving’s efficiency and accuracy, the unit commitment variables and unit power dispatch are respectively defined in a peak, a medium and a valley load. The proposed model is hiring a GA embedding with object relative dominant and Monte-Carlo simulation method. Finally, a provincial power corporation’s power system is used for simulation, and it is verified that the proposed method is effective.