源–荷互动环境下含高比例风电并网的自适应线性化概率潮流计算  

Adaptive-linearized Probabilistic Power Flow Calculation for Power Grid Integrated With High Proportion Wind Power in Source-load Interactive Environment

作　　者：柳志航 卫志农[1] 高昇宇 成乐祥 谈康 孙国强[1] 臧海祥[1] LIU Zhihang;WEI Zhinong;GAO Shengyu;CHENG Lexiang;TAN Kang;SUN Guoqiang;ZAN G Haixiang(College of Energy and Electrical Engineering,Hohai University,Nanjing 211100,Jiangsu Province,China;Nanjing Power Supply Company,State Grid Jiangsu Electric Power Co.,Ltd.,Nanjing 210019,Jiangsu Province,China)

机构地区：[1]河海大学能源与电气学院,江苏省南京市211100 [2]国网江苏省电力有限公司南京供电分公司,江苏省南京市210019

出　　处：《电网技术》2019年第11期3926-3934,共9页Power System Technology

基　　金：国家自然科学基金项目(51277052);国网江苏省电力有限公司重点科技项目(J2017142)~~

摘　　要：高比例风电大范围波动与柔性负荷随机响应加剧了电力系统不确定性,传统半不变量法计算概率潮流(probabilisticpowerflow,PPF)难以同时处理强随机性和相关性,导致分析结果具有较大误差。提出一种计及源-荷强随机性的自适应线性化半不变量法PPF计算方法。考虑源-荷互动随机响应,采用高斯混合模型准确建立源-荷双侧不确定性模型。针对源-荷强随机波动特性,提出自适应线性化处理方式,采用迭代更新波动子区域的方式实现随机波动范围自适应多区域划分,并在划分区域内部线性化以减小潮流全局线性化误差。同时,计及风功率区域相关性,利用自适应线性化半不变量法的解析法获取区域整合重组后状态变量的各阶半不变量。最后,运用C型Gram-Charlier级数拟合状态变量概率分布。对改进IEEE30节点系统进行测试,验证了算法的准确性及有效性,并定量评估了源-荷双侧扰动对系统运行的影响。Large-scale fluctuation of high proportion wind power and random response of flexible load aggravate uncertainty of power system.It is difficult to use conventional cumulant method based probabilistic power flow(PPF)to deal with high randomness and correlation simultaneously,resulting in large error.An adaptive-linearized cumulant based PPF method considering high randomness is proposed.Considering the stochastic response of source-load interaction,a source-load uncertainty model is accurately established using Gaussian mixture model.An adaptive linearization method for PPF is proposed to deal with high randomness.The sub-regions are updated iteratively to realize adaptive multi-region division of the fluctuation range and linearization within the region to reduce the global linearization error of power flow.Meanwhile,considering the regional correlation of wind power,cumulant after integration is obtained using adaptive linearized cumulant method.Finally,C-type Gram-Charlier series is used to estimate probability distribution of state variables.Accuracy and effectiveness o f the algorithm is verified with simulation of modified IEEE 30-bus system,and the impact o f source-load disturbance on system operation is quantitatively assessed.

关 键 词：源-荷强随机性 高斯混合模型 区域相关性 概率潮流 自适应线性化 半不变量 

分 类 号：TM721[电气工程—电力系统及自动化]