一种基于聚和经验模态分解方法的风速预测模型

A novel wind speed forecasting model based on ensemble empirical model decomposition

首先将原始数据进行季节调整,剔除季节因素,然后进行聚和经验模态分解(EEMD),并将预处理后的数据分别交由改进的相关向量机(RVM)模型分别进行预测.最后整合非线性预测模型的输出结果得到最终的预测值.通过对西北地区2个观测站的真实数据进行验证,实验结果表明,该方法可以得到精度更高的预测结果,且模型具有更好的稳定性及泛化性能.

Firstly, the seasonal factors in the data are removed by using seasonal adjustment methods. Then, the processed datais decomposed by EEMD method and the pre - processed data are respectively predicted by the improved relevance vector machine（ RVM）model. Finally, the final prediction value is obtained by integrating the output results of nonlinear prediction models. Theproposed method is validated by real data from two observatories in Northwest China. The experimental results show that the proposed model can obtain higher prediction precise and the model has better stability and generalization performance.