中文摘要：

光伏发电相关性以及波动性会对系统的运行产生影响,因此需要通过计及光伏出力相关性的概率潮流（probabilistic load flow,PLF）计算来获取系统运行特征量的统计信息。文中提出采用改进Nataf变换处理光伏相关性。在传统Nataf变换基础上,结合三阶多项式正态变换简化其计算。同时,为解决PLF的输入和输出变量之间非线性关系带来的计算复杂性,提出采用多重积分法（multiple integral method,MIM）和Gram-Charlier级数,仅需在少数输入点处进行潮流计算即可得到输出量统计特征的高精度结果。对IEEE 39节点系统进行仿真计算,结果验证了所提算法的有效性、准确性、计算高效性。与半不变量法的对比则验证了半不变量法的局限性以及MIM法的优良性能。

英文摘要：

Due to the impact of the correlation and uncertainty of photovoltaic（PV） generation on power system, it is necessary to apply probabilistic load flow（PLF） considering correlation of PV generation to obtain the statistics of system characteristic quantities. A modified Nataf transformation was proposed to handle the correlation of PV generation in this paper. Based on the traditional Nataf transformation, a third-order polynomial normal transformation was adopted to reduce computational effort of traditional Nataf transformation. In order to simplify the computational complexity induced by the nonlinear relationship between input and output variables of PLF, multiple integral method（MIM） and Gram-Charlier expansion were introduced. Based on load flow calculation corresponding to specified input variables, high precision results of output statistics can be estimated. An IEEE 39-bus system was tested in the simulation study. The simulation results verified the effectiveness, accuracy and efficiency of the proposed algorithm. Comparison between cumulant method and MIM confirmed the limitation of cumulant method and well performance of MIM.