In order to improve the accuracy and reliability of ultra-short term and short term wind speed prediction, the correlations among the wind speeds, air pressures, wind directions and air temperatures of seven typical locations on the main path of China winter monsoon are analyzed. First, based on the characteristics of correlation/distance indexes, the Pearson correlation coefficient is sorted out to calculate the correlation between the wind speed and its influence factors. The average optimal lag time between wind speeds are calculated by the Sample Cross Correlation Function (SCCF). Then, the correlation between the wind speed and the first principal component of influence factors and their optimal lag time are calculated on the time window. The results show that the Taiwan Strait is a wind energy resource-rich area that is particularly suitable for the use of spatial correlation for ultra-short-term prediction. If the Aozi wind speeds are predicted by the air pressures, wind speeds, wind directions and air temperatures of Beijing, the prediction errors are insensitive to the optimal lag time. The air pressure difference between Beijing and Aozi has the most important impact on prediction errors. This work is supported by National Natural Science Foundation of China (No. 51625702).