采用混合高斯模型及边缘变换技术的蒙特卡洛随机潮流方法

摘要

A probabilistic load flow method considering the correlation between input variables based on improved Monte Carlo simulation ( MCS) is proposed . Regarding the behaviors of diversity and randomness for variable input , the method establishes the Gaussian mixture model ( GMM ) for variable input and performs parameter estimation by the use of measured data . Uniform design sampling ( UDS) is introduced to improve the sampling efficiency , and correlated samples are generated by marginal transformation and Cholesky decomposition . Moreover , multi‐linearization is applied to reduce the truncated error as well as time consumption . The simulation results of IEEE 30‐bus and IEEE 118‐bus test system verify the effectiveness , accuracy and practicability of the proposed method .%提出一种计及输入变量相关性的改进蒙特卡洛随机潮流方法。该方法针对系统输入变量多样性和随机性的特点，建立其混合高斯模型并根据测量数据进行参数估计。在蒙特卡洛仿真基础上，引入均匀设计抽样技术提高采样效率，并通过边缘变换和 Cholesky 分解产生具有相关性的输入变量样本。采用多重线性化手段在加快计算速度的同时减小潮流方程线性化引起的截断误差。对 IEEE 30和 IEEE 118节点系统的测试分析，验证了所提方法的准确性、快速性和有效性。