Under the power environment of Internet of things used in autonomous multi-robot systems, due to the influence of various actual environmental factors, the security protection method based on security partition and physical isolation of nodes is difficult to fully meet the information security protection requirements of the power system, and the network threats such as impersonation, eavesdropping, and dexterity on communication links are still difficult to avoid. In order to improve the security of node data transmission in autonomous multi-robot power systems, reducing network energy consumption has become the bottleneck of network application and development. In this paper, we propose an Improved Effective Clustering Algorithm (IECA) and transmission algorithm of power grid data based on autonomous multi-robot systems. The distance between nodes sending data to the next-hop node is equivalent to the energy model. The remaining energy in the cluster head node sending data in the next hop is recorded, and the remaining energy is subtracted from the equivalent energy to find a maximum value. The corresponding node of the maximum value is taken as the number of nodes According to the received node. Compared with the LEACH algorithm and GAF algorithm, this algorithm can not only extend the lifetime of nodes but also has more efficient performance.