In order to improve the efficiency of power system restoration after a power outage,it has become an inevitable trend that energy storage station which has advantages of fast response and frequency control is employed in system restoration. Currently,energy storage is mainly utilized as an auxiliary power source for renewable energy to smooth out fluctuations of renewable energy. Large-scale energy storage can also be utilized as black start power source to provide starting power for units to be restored. However,the above-mentioned studies have focused on the use of the output power of energy storage,while ignoring the role of energy storage and frequency modulation capabilities in improving system safety,thus failing to improve the recovery efficiency of the system to a greater extent. To this end a power system load restoration strategy that considers the power storage and frequency modulation capability is proposed. With the goal of restoring as many important loads as possible,a fuzzy chance constraint model for power system load restoration is constructed considering the power characteristics,frequency response characteristics and other safety constraints of energy storage. It is further transformed into a deterministic 0-1 programming problem through clear equivalence classes,and the artificial bee colony algorithm is employed to solve it. Taking the IEEE 39 system as an example for simulation analysis,the simulation results show that the strategy proposed in this paper improves the efficiency of load restoration.