The smart distribution network communication system carries many tentacles of business data of the power grid. With the extensive access of new energy power, after the distribution network connects renewable power supplies, the data necessary for services such as dispatch, control, and management of the micro-grid system will inevitably need to be transmitted through the distribution network communication system. How to dynamically, intelligently, and adaptively adjust the allocation of data stream service transmissions, so that the orderly division of labor and cooperation of various nodes, collaborative operation, is a new requirement of intelligent distributed data processing. This paper adopts a parallel data stream processing method to effectively integrate data sources of different sources through stream density, and uses time-division multiplexing technology to transmit different levels of data streams in time slots, using container-based control methods and series of data preprocessing. Technology to ensure the reliability of data. The research results of this paper provide effective theoretical and experimental basis for the distribution network automation control system.