The probability distribution model of wind speed is widely used in risk assessment of wind power integrated system. Unfortunately, the current risk assessment mainly focuses on annual risk indices and cannot reflect the time-varying characteristics of wind speed and system risk. In this paper, the time-periodic characteristics of wind speed are proposed and it is described as two parts superposition. The first part is the long-term and gentle monthly trend which can be represented as a time-periodic fitting function, while the other one is the short-term and rapid daily fluctuation characteristics which can be denoted as a random variable that obeys a certain probability distribution. The wind speed model with time-periodic characteristics is built through curve fitting based on wind speed data for several years. Accordingly, wind power output of wind farms can be determined. Furthermore, the mid- and long-term risk indices of wind power integrated system can be calculated through Monte Carlo simulation, and can reflect time varying characteristic of system risk. The validity of proposed method is verified through a case analysis adopting IEEE-RTS79 system and actual wind speed of a wind farm. The result of risk assessment can provide reference for power system planning, mid- and long-term dispatching and monthly generation scheduling making. This work is supported by National Natural Science Foundation of China (No. 51707018) and State Grid Gansu Electric Power Company (No. 52272815001A).