The modern large-scale power system is confronted by many kinds of uncertainty factors

the disturbance brought by which has become one of the major threats to safe operation of the system. In view of this problem

a new method of large-scale power system component vulnerability assessment is proposed. Based on the improved entropy and by referring to a sensitivity analysis of the wind power system and branch capacity margin

the distribution information on disturbance power safe transfer is investigated to establish the node vulnerability assessment indicator for measuring the ability of node to resist disturbance. On this basis

the branch outage is translated into virtual power injection disturbance on the two-terminal nodes by the compensation method. The branch vulnerability is then assessed with the entropy weight method by combining with node vulnerability assessment. Results of an IEEE 300-bus system case study have validated the effectiveness and quickness of the presented method. The factors that influence the key element vulnerability are analyzed

providing reference for error control and operation management in power systems. This work is supported by State Grid Corporation of China (No. dz71 13 036).