IR-GAMs)的气温负荷分解方法。首先,通过MDTW构建关联概率模型,动态量化气温与总负荷的非线性耦合强度;其次,将关联概率作为动态权重融入IR-GAMs损失函数,经迭代重加权优化抑制非气温干扰并消除负值负荷;最后,基于我国某地区实际数据,采用Granger因果检验与小波相干分析验证方法有效性。结果表明:所提方法估计的夏季最高气温负荷约占同期总负荷的40%,与基准测算最为接近,优于局部加权回归的季节趋势分解(seasonal-trend decomposition using LOESS
Ingrid Daubechies,Ronald DeVore,Massimo Fornasier,C. Si̇nan Güntürk.Iteratively reweighted least squares minimization for sparse recovery[J].Communications on Pure and Applied Mathematics,2010.
Tao Hong,Pu Wang,Laura White.Weather station selection for electric load forecasting[J].International Journal of Forecasting,2015.
Florian Ziel.Quantile regression for the qualifying match of GEFCom2017 probabilistic load forecasting[J].International Journal of Forecasting,2019.