Abstract: It is difficult to merge wind power into a grid, owing to wind power’s uncertainty and prediction inaccuracy. Wind speed is an important factor affecting wind power, so the accuracy of wind speed prediction has a major impact on the wind power prediction. An optimized prediction model based on core vector regression（CVR） is proposed in short-term wind speed forecasting. The wind speed data from a wind farm are collected hourly as the inputs. The particle swarm optimization （PSO） method is used to optimize the CVR model parameters. Experimental results show that the method has higher prediction accuracy than the CVR and support vector regression （SVR） method.

Key words: wind speed, wind power, short-term forecasting, particle swarm optimization（PSO）, core vector regression（CVR）