基于时变Copula函数的风电出力相关性分析

Wind Power Correlation Analysis Based on Time-variant Copula Function

相邻风电场风电出力存在较强的相关性,但一段时间内其相关强度并非一成不变,所选取时间窗口不同,可能会得到不同的相关性结论。针对这一问题,基于Copula相关性分析理论,提出了采用时变Copula函数来分析风电出力时变相关性。并研究了3种时变Copula函数的特性,给出了时变相关系数的时变方程,再采用IFM法和AIC法进行Copula函数参数估计和拟合优度检验。通过实例分析,对比了静态Copula函数和时变Copula函数拟合优度,证明了时变Copula函数在风电场风电出力相关性分析中的可行性和优越性,为风电场风电出力相关性分析提供了更有效的新方法。

There often is a strong correlation between the adjacent wind farm＇s wind power,but the correlation intensity of wind power is not immutable within a period of time. When the selected time window is different,it may obtain differ- ent correlation conclusions. Based on the Copula correlation analysis theory ,time-varying Copula function is proposed to analyze correlation of wind power, three kinds of time-variant Copula functions are studied, also the time-variant cor- relation coefficient of each Copula is given, then the IFM and AIC methods to estimate the parameter and test the good- ness of fitting is wtilized. In this paper, goodness of the static Copula function and the time-variant Copula function is compared by a practical case ,which has demonstrated the feasibility and superiority of the time-variant Copula function in analyzing the correlation of wind farm wind power output. The proposed method offers better effectiveness to analyze the correlation of wind farm power output.