Some crucial nodes of the electric power communication network have important significance to the safe and reliable operation of the network. To identify the key nodes, a method based on Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) algorithm is proposed. First, it builds the evaluation system of the nodes, and considers each node as a solution and each evaluation index as a property. Then, it combines the subjective method of analytic hierarchy process with objective method of entropy weight method to calculate the comprehensive weight, and the comprehensive weight is added to each index. Finally, the multiple attribute decision making method is used to calculate the important degree of each node, and the key node is identified according to the important degree. The actual power grid of one province is used to test the theory. Simulation results show that this method can recognize the key node more accurately than other existing methods, proving the practicability and validity of this method. This work is supported by Science and Technology Support Project of Sichuan Province (No. 2017GZ0349) and Science and Technology Project of State Grid Xinjiang Electric Power Company (No. SGXJXT00TJS1600206).