A combination method for wind power predication based on cross entropy theory

Wind power predication is significant for power system dispatching and safe-stable operation. For solving the issue that wind power prediction depends on reference samples deeply, a new combined method of wind power prediction based on cross entropy theory was proposed according to information fusion technique. In the method, wind power prediction was seen as information fusion, and the cross entropy theory was applied to judge the cross degrees among different prediction methods. The weights of different wind power prediction methods was dynamically set according to their cross levels. The basic idea and implementation details of the proposed method were presented. Application results in an actual wind farm show that the proposed method can effectively identify the cross degrees among different prediction methods and improve prediction precision. © 2012 Chinese Society for Electrical Engineering.