Short-term wind power prediction based on DE-XGBoost
Wind power prediction can provide important decision reference for the safe and stable operation of power system,thus it is of great significance to study how to improve the accuracy of wind power predic-tion.To solve the problems of poor prediction accuracy of short-term wind power,a combined prediction model of optimized eXtreme Gradient Boosting(DE-XGBoost)based on Differential Evolution algorithm(DE)is proposed to optimize extreme gradient lift trees.The differential evolution algorithm with fast con-vergence speed,good optimization effect and low complexity is used to optimize the model parameters of XGBoost to achieve accurate prediction of wind power.Simulation results show that,compared with other prediction models,DE-XGBoost model has higher prediction accuracy and stronger generalization ability,which can provide detailed data for wind power scheduling operation.
wind power predictionprediction accuracyDifferential EvolutionDE-XGBoostcombined model
万方数据
维普
