The accurate estimation of inertia is the premise of analyzing the security and stability of the system frequency, but the existing methods can not quantitatively evaluate the effect of virtual inertia on the equivalent inertia of the power system. In view of this, an inertia estimation method of power system with virtual inertia response of wind power is proposed. Firstly, the theoretical expression of equivalent inertia of power system including virtual inertia response of wind power is derived. Secondly, the phasor measurement unit (PMU) is used to obtain the active power and frequency data at the node of the power generation device. Thirdly, the active power output of the power generation device is taken as the input of the identification model, and the frequency disturbance is taken as the output. Using controlled autoregressive (CAR) model and time-varying forgetting factor recursive least squares (TFF-RLS) algorithm evaluated the equivalent inertia of the whole network under different wind power proportions. Finally, the feasibility and applicability of the proposed method are verified by an improved IEEE-10-machine 39-node system. Compared with the traditional identification method, the accuracy of the inertia estimation of the proposed method is improved, and it is suitable for the time-varying inertia estimation of the new energy power system.