The invention discloses an intelligent power grid slice distribution method based on reinforcement learning, and the method is characterized by comprising: classifying the power businesses of an intelligent power grid according to the business types; enabling the classification to correspond to different slices; and constructing a reinforcement learning model of the smart power grid slices according to the service indexes of the smart power grid, and completing distribution of the smart power grid slices through the reinforcement learning model, thereby realizing resource scheduling managementof the smart power grid. The businesses types of the smart power grids are classified, the classifications correspond to different slices, and distribution of the smart power grid slices is completedthrough the constructed reinforced learning model of the smart power grid slices. Therefore, the integration problem of the 5G network slicing technology based on reinforcement learning and the intelligent power grid is solved.