Abstract: By establishing the electromechanical co-simulation model of disconnector and analyzing the fault simulation results,this paper expounds the mapping relationship between the power and stress of the whole stroke of the disconnector,and proposes the segment extraction method of typical fault power characteristics according to the corresponding relationship between the fault characteristics and stroke interval. A total of 10 original characteristic quantities are extracted such as motion time,the mean value and variance of output power in different motion stages. The algorithm is verified in the diagnosis of the typical faults such as jamming,three-phase un-synchronization and inadequate action in the simulation. The SVM algorithm is used to realize the intelligent monitoring and fault diagnosis of the mechanical state of the disconnector. In the end,through the fault simulation experiment of GW4 disconnector,the effectiveness and accuracy of the proposed intelligent algorithm in industrial practice is verified.