With the change of servo system parameters, the performance of the speed controller in the servo motor drive decreases. Aiming at this problem, a speed adaptive control method based on parameter identification is proposed. For the change of inertia, the model reference adaptive system (MRAS) is established in the continuous domain, and a proportional integral (PI) adaptive law is designed based on Popov hyperstability theory, which improves the convergence speed and steady-state accuracy of inertia identification. For load torque, a disturbance observer based on intermediate variable design is proposed. The identified values are fed back to the speed controller to realize on-line self-tuning of controller parameters. The experimental results show that the proposed method can accurately identify the moment of inertia and load torque for parameter adjustment of the controller and improve the controller′s robustness to parameter change.