In order to make full use of the information of the intelligent substation, help the operation and maintenance personnel to locate the faults of the secondary system of the intelligent substation more quickly and accurately, so as to ensure the safe and stable operation of the power system, a fault diagnosis method for secondary system of intelligent substation based on multi classification support vector machine(SVM) is proposed, and particle swarm optimization(PSO) algorithm is used to optimize the parameters automatically. The expert database is built according to the historical state data of the long-term operation of the intelligent substation and the processing results of the maintenance personnel. The original signal of the secondary system of the intelligent substation is a large number of {0, 1} state variables. The data group is defined in the form of switching value coding, and the data dimension of the signal is reduced by using the principal component analysis method. Finally, the fault diagnosis model of the secondary system of the intelligent substation based on the information of the terminal is constructed. The results of the comparison between the diagnosis model and the expert manual judgment show that the method proposed has excellent accuracy and applicability.