Abstract: There is strong complementarity between wind and water resources in seasonal distribution. Using wind power good peak regulation and frequency modulation characteristics, the intermittency and instability of wind power supplied can be overcomed, and the wind power and hydropower can remain a certain range and result in a relatively stable output. According to the basic principle and research status of the coordination between wind power and hydropower, control method and its limitations are studied. In order to improve the coordination effect, the combination of the fruit fly algorithm and the BP neural network method are proposed to control the wind power and hydropower system.