考虑异方差效应的风电不确定性建模及其在调度中的应用

随着风电在电力系统中渗透率的不断提升,其不确定性为电网的安全经济运行带来了重大挑战。为获得精准的风电不确定性模型,帮助运行人员实现系统的安全经济运行,文中提出了考虑异方差效应的风电预测误差条件概率分布建模方法。首先,分析了风电预测误差与各类因素的相依性水平,并基于分析结果与动态Copula理论,建立了风电波动性与风电预测误差的动态相依性模型;之后,针对边缘分布所显示出的时域特征,结合差分整合移动平均自回归(ARIMA)模型与广义自回归条件异方差(GARCH)模型,考虑异方差效应,建立了时变边缘分布模型;最后,将两模型相结合,给出了不同波动水平下的风电条件预测误差分布情况,并在不确定性机组组合模型中进行验证,证明了模型的有效性。

Modelling of Wind Power Uncertainty Considering Heteroskedasticity Effect and Its Application in Power System Dispatching

As the penetration rate of wind power in the power system continues to increase, its uncertainty poses a great challenge to the safe and economic operation of the power system. In order to obtain accurate wind power uncertainty model and help operators to achieve safe and economic operation of the system, this paper proposes a conditional probability distribution modelling method of wind power forecasting error considering the heteroskedasticity effect. Firstly, the dependence of wind power forecasting error and various factors is analyzed. Based on the results and the dynamic Copula theory, a dynamic dependence model of wind power forecast error is established. Then, based on the time-domain features displayed by the edge distribution, combined with the autoregressive integrated moving average (ARIMA) model and the generalized autoregressive conditional heteroskedasticity (GARCH) model, this paper develops a time-varying edge distribution model with the consideration of heteroskedasticity effect. Finally, the two models are combined to give the forecasting error distribution of wind power conditions at different fluctuation levels, and the verification is performed in the uncertain unit combination model, which proves the validity of the model.