With the expansion of the scale of smart grid and the improvement of service real-time requirements, lowlatency communication network becomes the key infrastructure to guarantee the automation and control of power system. This paper proposes an intelligent route optimization scheme based on multicast routing algorithm. The layered architecture and performance requirements of smart grid wide-area communication network are analyzed, and the shortest path tree algorithm is proposed. Combined with the multi-constraint QoS model, the improved multi-tree multi-constraint routing algorithm based on cost correction (MTMCR-CC) is designed. By dynamically adjusting the link cost function and constructing multicast tree candidate sets, network congestion prevention, load balancing and fault isolation are realized. Simulation results show that compared with the traditional SPRS and HCARS algorithms, MTMCR-CC always has a lower packet loss rate, the average bandwidth of the link reaches 69.72% of the maximum value, and the success rate of the node 25-position communication reaches 99%, and also performs significant optimization in the key indexes such as node load balancing. In the simulation scenario, the network traffic in all paths is lower than 2.5×107 bytes. It is proved that the proposed algorithm can meet the demand of millisecond response and high reliability of power system.
