赵亦岚,陈攀,周阮凯,罗平,董雨轩
ZHAO Yilan,CHEN Pan,ZHOU Ruankai,LUO Ping,DONG Yuxuan

• 1:杭州市电力设计院有限公司
• 2:杭州电子科技大学自动化学院

摘要(Abstract)：

由于LHS(拉丁超立方采样)能较大程度覆盖采样区间,因此常作为概率潮流计算中的分层抽样方法。但随着风、光等分布式能源在电力系统中渗透率的提高,以及采样对象维度的增加,使得LHS所得样本的均匀性略显不足。同时,风、光等随机变量之间的相关性也会日趋复杂。相关矩阵作为表现随机变量之间相关性的有效方式,在矩阵维度增加时会出现非正定情况,从而导致基于Cholesky分解排序变换的概率潮流计算方法无法进行。针对上述问题,提出一种考虑风、光相关性为非正定时的改进LHS概率潮流计算方法,并利用改进的IEEE 30与IEEE 118系统验证了所提方法的准确性与有效性。
LHS(Latin hypercube sampling) can extensively cover the sampling range, so it is often used as a stratified sampling method in probabilistic power flow calculation. However, due to the increasing penetration rate of distributed energy such as wind power and photovoltaic power in the power system and the increasing dimension of the sampling object the uniformity of LHS is insufficient. At this time, the correlation of the random variables may become increasingly complex. As an effective way to express the correlation between random variables, the correlation matrix will appear non-positive definite when the matrix dimension increases, which leads to the failure of probabilistic power flow calculation method based on Cholesky decomposition sorting transformation. To solve these problems, an improved LHS probabilistic power flow calculation method considering the non-positive definite correlation relationship between wind power and photovoltaic power is proposed. The improved IEEE 30 and IEEE 118 systems are used to verify the accuracy and effectiveness of the proposed method.

关键词(KeyWords)： 概率潮流;相关性修正;非正定;采样均匀性;改进的拉丁超立方采样
probabilistic power flow;correlation correction;non-positive definite;sampling uniformity;improved LHS

Abstract:

Keywords:

基金项目(Foundation): 国家电网有限公司科技项目（198006）;; 浙江省自然科学基金项目（LY20E070004)

作者(Author): 赵亦岚,陈攀,周阮凯,罗平,董雨轩
ZHAO Yilan,CHEN Pan,ZHOU Ruankai,LUO Ping,DONG Yuxuan

DOI: 10.19585/j.zjdl.202004006

参考文献(References)：

• [1]黄煜,徐青山,卞海红,等.基于拉丁超立方采样技术的半不变量法随机潮流计算[J].电力自动化设备,2016,36(11):112-119.
• [2]YU H,CHUNG C Y,WONG K P,et al.Probabilistic load flow evaluation with hybrid latin hypercube sampling and cholesky decomposition[J].IEEE Transactions on Power Systems,2009,24(2):661-667.
• [3]ZOU B,XIAO Q.Solving probabilistic optimal power flow problem using quasi monte carlo method and ninth-order polynomial normal transformation[J].IEEE Transactions on Power Systems,2014,29(1):300-306.
• [4]张建波,张忠伟,杨洋.改进拉丁超立方蒙特卡洛模拟[J].吉林大学学报(信息科学版),2018,36(4):452-458.
• [5]HELTON J C,DAVIS F J.Latin hypercube sampling and the propagation of uncertainty in analyses of complex sys tems[J].Reliability Engineering&System Safety,2003,81(1):23-69.
• [6]LIEFVENDAHLA M,STOCKI R.A study on algorithms for optimization of latin hypercubes[J].Journal of Statistical Planning and Inference,2006(136):3231-3247.
• [7]CONSTANTINO G,GEORGE C.Explaining the saddlepoint approximation[J].The American Statistician,1999,53(3):216-224.
• [8]CHEN Y,WEN J,CHENG S.Probabilistic load flow method based on nataf transformation and latin hypercube sampling[J].IEEE Transactions on Sustainable Energy,2013,4(2):294-301.
• [9]CHEN C,WU W,ZHANG B,et al.Correlated probabilistic load flow using a point estimate method with Nataf transformation[J].International Journal of Electrical Power&Energy Systems,2015(65):325-333.
• [10]石东源,蔡德福,陈金富,等.计及输入变量相关性的半不变量法概率潮流计算[J].中国电机工程学报,2012,32(28):104-113.
• [11]方斯顿,程浩忠,徐国栋,等.基于Nataf变换含相关性的扩展准蒙特卡洛随机潮流方法[J].电工技术学报,2017,32(2):255-263.
• [12]KUPIEC.P H.Stress testing in a value at risk framework[J].The Journal of Derivatives,1998,6(1):7-24.
• [13]徐青山,杨阳,黄煜,等.基于非正定型相关性控制的拉丁超立方随机潮流计算方法[J]-高电压技术,2018,44(7):2292-2299.
• [14]徐潇源,严正,冯冬涵,等.基于输入变量秩相关系数的概率潮流计算方法[J].电力系统自动化,2014,38(12):54-61.
• [15]XU X,YAN Z.Probabilistic load flow evaluation consider ing correlated input random variables[J].International Transactions on Electrical Energy Systems,2016,26(3):555-572.
• [16]严岩,苏宏升,车玉龙.基于改进LHS的含风电电力系统概率潮流计算[J].太阳能学报,2018,39(4):1111-1118.
• [17]HICKERNELL F J.A generalized discrepancy and quadra ture error bound[J].Mathematics of computation of the American Mathematical Society,1998,67(221):299-322.
• [18]YIN J F,HUANG Y.Modified multiplicative update algorithms for computing the nearest correlation matrix[J].J.appl.math.inform,2012,30(30):201-210.
• [19]ZHANG L,ZHOU W,LI D.Global convergence of a modified fletcher-reeves conjugate gradient method with Armijo-type line search[J].Numerische Mathematik,2006,104(4):561-572.
• [20]王敏.分布式电源的概率建模及其对电力系统的影响[D].合肥:合肥工业大学,2010.
• [21]赵继超,袁越,傅质馨,等.基于Copula理论的风光互补发电系统可靠性评估[J].电力自动化设备,2013,33(1):124-129.
• [22]李韵喻,梁有珍,甄文喜,等.考虑光伏非线性相关性的电力系统概率潮流计算[J].电网与清洁能源,2019,3 5(6):69-75.
• [23]章立宗,张锋明,蒋正威,等.基于并行概率潮流计算的含光伏配电网电压风险评估[J].浙江电力,2018,37(8):7-14.
• [24]叶晨,崔双喜,王维庆.含风电的电力系统概率潮流计算[J].电网与清洁能源,2018,34(2):167-172.