【摘要】 风电功率预测是提高风电场功率上报准确率和风电场收益、提高大规模风电并网后电力系统的安全和稳定性的有效方法。目前对于风电功率超短期和短期预测的研究已取得了阶段性研究成果,但是对于风电功率中长期预测方面的研究尚未取得实质进展,预测效果难以满足实际工程应用的需要。针对现有风电功率中长期预测方法存在的缺陷,提出了一种基于改进GM-ARMA组合模型的风电功率中长期预测方法,采用遗传算法优化组合模型的参数,以得到最优模型,进而提高了风电功率预测精度。仿真结果表明该方法对年度和月度风电功率预测均有较好的应用效果,比普通GM-ARMA组合模型具有更高的预测精度。更多还原

【Abstract】 Wind power prediction is an effective method to improve the power reporting accuracy and revenue of wind farms, and to improve the safety and stability of power systems after large-scale wind power grid connection. At present, the research on ultra-short term and short term prediction of wind power has achieved stage research results. However, the research on medium and long term prediction of wind power has not made substantial progress and the prediction effect cannot meet the needs of practical engineering application. In view of the defects of the existing medium and long term wind power prediction methods, a medium and long term wind power prediction method based on the improved GM-ARMA combination model was proposed. Genetic algorithm was used to optimize the parameters of the combination model to obtain the optimal model, thus improving the wind power prediction accuracy. Simulation results show that this method is effective in both monthly and annual wind power forecasting, and has higher prediction accuracy than the general GM-ARMA combined model.更多还原