As the exclusive communication network for power system, the electric power communication network has obvious characteristics of the electric power industry. In order to identify the crucial nodes of network accurately, the index of power network impact factor is proposed based on the special relationship between power system and power communication network. The nodes weights are evaluated on the basis of the category factor and the load factor, namely the position and function in the station where the nodes belong. Meanwhile, combining with a new definition for coefficient polymerization indexes, it analyzes the distribution state of the critical nodes. The simulation results show that the proposed algorithms can distinguish the importance degree of the nodes with similar topological positions by combining the application characteristics of power communication network, and has great reference value in power communication network operational risk analysis.