Abstract: There are many interference factors in the gain process of generator excitation system, resulting in poor gain robustness of generator excitation system. The current research on external interference factors is insufficient, and the gain control effect is poor. In order to improve the robustness of gain control of generator excitation system, a simulation method of gain robustness of generator excitation system under external multiple source disturbances is designed. The uncertain nonlinear analysis process is introduced, and the APSO algorithm is innovatively used to code the interference factors to be identified and correct the abnormal data. The correction function of nonlinear system with uncertain parameter vector is projected synchronously, and the multi-source interference factor is calculated. The data model of generator excitation system including rotor motion equation and rotor motion stability analysis model of synchronous generator is established. Combining schur compensation theorem and matrix decoupling, the stability matrix is solved to limit the excitation voltage control quantity; LPV control system is used for system gain scheduling, and output feedback controller is designed. According to the relationship between active power and voltage, the optimal control law of the system is set to realize the gain robust control of the generator excitation system under external multi-source disturbances. The experimental results show that the power angle change, speed response and transient potential response can be restored to normal within 3 s after the application of the proposed method under external multi-source interference, which proves that the proposed method has high robustness.