基于Pair Copula的随机潮流三点估计法

Pair Copula Based Three-point Estimate Method for Probabilistic Load Flow Calculation

现有Copula等方法难以对多维风电功率准确建模,且传统的点估计法无法直接应用于风电功率具有相关性的场合。针对这一问题,提出基于Pair Copula和概率积分变换的随机潮流点估计法。首先采用Pair Copula对多维风电功率进行建模,然后使用点估计法在独立正态概率空间中采样,最后,依据概率积分变换,把采样点变换到实际风电功率的概率空间中进行潮流计算,从而使点估计法能够处理具有任意概率分布的多维风电功率。对澳大利亚多个风电场出力样本的建模和分析验证了Pair Copula模型的优越性,基于IEEE 118节点系统的算例验证了所提方法的有效性。

Current methods,such as Copula theory,are inadequate to model multiple dependent wind power outputs accurately. Moreover, the point estimate method cannot handle the correlation among wind power outputs. Thus,an improved point estimate method based on Pair Copula and probability integral transformation is proposed for probabilistic load flow studies. The probabilistic model of multiple correlated wind power outputs is firstly constructed by Pair Copula. The point estimate method is then used to generate samples in the independent normal domain. Finally,based on the probability integral method,the samples are transformed into the actual probabilistic domain in order to find the characteristics of the power system operation. In this way, the point estimate method can handle multiple dependent wind generations with arbitrary distributions. The modeling and analysis for the power outputs of adjacent wind farms in Australia verify the goodness-of-fit of Pair Copula. The probabilistic load flow of the IEEE 118-bus system is solved to demonstrate the effectiveness of the proposed method.