基于DCC-GARCH的海上风电场出力空间相关性分析及预测

多座海上风电场出力之间存在一定的空间相关性，构建合适的风电出力相关性模型有助于提高风电出力的预测精度。针对空间相关性具有时变特性以及难以描述和衡量，提出基于动态条件相关广义自回归条件异方差（DCC-GARCH）模型的海上风电场出力相关性模型。利用多维正态分布和DCC-GARCH模型拟合多风电场的皮尔森相关系数，求解随时间变化的风电场出力空间相关系数，在准确表征空间相关性大小的同时体现空间相关性的时序变化特征。基于DCC-GARCH模型建立多座风电场出力动态空间相关性短期预测模型。基于江苏省盐城市海上风电场数据进行算例分析，结果验证了所提方法的合理性和有效性。

Spatial correlation analysis and prediction of offshore wind farm output based on DCC-GARCH

There exists certain spatial correlation between the outputs of multiple offshore wind farms, it is helpful for improving the prediction accuracy of wind power output to construct suitable wind power output correlation model. Aiming at that the spatial correlation has time-varying characteristic and is difficult to describe and measure, an output correlation model of offshore wind farm is proposed based on dynamic conditional correlation generalized auto regressive conditional heteroskedasticity(DCC-GARCH) model. The multi-dimensional normal distribution and DCC-GARCH model are used to fit Pearson correlation coefficient of multiple wind farms, the spatial correlation coefficient of wind farm output which varies with the time is solved, which accurately represents the size of spatial correlation while reflects the time-varying characteristic of spatial correlation. A short-term prediction model of output dynamic spatial correlation for multiple wind farms is built based on DCC-GARCH model. Case analysis is carried out based on the data of offshore wind farms in Yancheng City, Jiangsu Province, and results verify the rationality and effectiveness of the proposed method.