Virtual synchronous machine provides inertia and damping support for power systems, but at the same time, it will bring the similar problems of frequency and power oscillation as traditional synchronous generators. The oscillation of frequency is accompanied by the oscillation of active power. Most of existing virtual synchronous machine control methods focus on improving the system frequency stability, ignoring the power oscillation. Therefore, an optimal control strategy of inertia damping parameters of virtual synchronous machine is proposed. Taking the minimum sum of squares of frequency deviation and active power deviation as the objective function, the genetic algorithm is used to explore the minimum value of the newly constructed objective function. This optimization method is then verified by a specific test model. Based on this time?domain simulation model in PSCAD, the proposed method can therefore be validated. The capability to balance the frequency oscillation and output active power oscillationwill make the system frequency oscillation and power oscillation enhanced.