Abstract: Complex temporal and spatial dependencies exist among wind speeds of various wind turbines in wind farms,leading to difficulties in improving the accuracy of wind-speed prediction. Analyzing the spatio-temporal dependency of wind speed,understanding the mutual influence among wind turbines remain problems to be solved. Accordingly,this paper first selects the appropriate Copula function for combination by analyzing the goodness of fit of single-Copula function;Then,by constructing the hybrid Copula function model to analyze the correlation of the wind speeds of multiple wind turbines in the wind farm;Finally,apply the expectation maximum（EM）algorithm to solve the correlation coefficient of the model and complete the wind-speed prediction. Combining optimization algorithms to improve the Copula function overcomes the difficulty in finding the spatio-temporal dependency of wind speed and lay a foundation for obtaining accurate wind-speed forecasts. The validity of the method is verified by using the measured wind-speed data of wind turbines in a certain area of China. Experimental results show that the model improves the accuracy of wind-speed prediction based on the accurate analysis of the spatio-temporal dependency of wind speed.