With the increase of photovoltaic(PV)penetration

the power grid needs a large capacity of regulation reserve to cope with the large random fluctuation of PV power. The energy storage system(ESS)can easily deal with the large fluctuation of PV power due to its fast charging and discharging capability

which can reduce the reserve capacity needed for frequency regulation. However

ESS cannot be used to cope with the PV power fluctuation independently owing to its high price. Therefore

this paper proposes a new strategy which uses limited ESS to smooth PV ramping power

to effectively suppress large fluctuations of PV output. Considering the restriction relation between regulating reserve and energy storage capacity

a mathematical model is developed for the calculation of frequency regulation costs after applying ESS. And the energy capacity and power capacity of ESS are optimally configured with the goal of maximizing equivalent return. Finally

the genetic algorithm(GA)is used to solve the optimization model

and simulation results based on a power grid with high penetration of PV are presented to verify the effectiveness of the proposed method.