基于偏最小二乘多项式稀疏展开的含风电电力系统概率潮流计算

董晓阳; 梁琛; 马喜平; 李亚昕; 杨军亭

doi:10.19912/j.0254-0096.tynxb.2022-0073

PDF(1667 KB)

太阳能学报 ›› 2023, Vol. 44 ›› Issue (6) : 351-359. DOI: 10.19912/j.0254-0096.tynxb.2022-0073

基于偏最小二乘多项式稀疏展开的含风电电力系统概率潮流计算

董晓阳^1,2, 梁琛^1,2, 马喜平^1,2, 李亚昕^1,2, 杨军亭^1,2

作者信息 +

PROBABILISTIC FLOW CALCULATION OF POWER SYSTEM CONSIDERING WIND POWER BASED ON SPARSE POLYNOMIAL CHAOS EXPANSION WITH PARTIAL LEAST SQUARES METHOD

Dong Xiaoyang^1,2, Liang Chen^1,2, Ma Xiping^1,2, Li Yaxin^1,2, Yang Junting^1,2

Author information +

文章历史 +

摘要

计及新型环保绿色能源如风电、光伏的不确定性,传统确定性潮流计算难以全面描述电力系统的运行情况。针对传统蒙特卡洛概率潮流算法计算量庞大的问题,结合偏最小二乘回归算法和多项式代理模型,提出一种偏最小二乘多项式疏展开的概率潮流算法。利用偏最小二乘回归算法的伪交叉验证误差自适应机制筛选出多项式展开式中的贡献度较大的多项式,得到多项式展开式的稀疏表达形式,可克服多项式展开概率潮流在输入变量较多时面临的维数灾难问题。在改进的IEEE-9,IEEE-30算例中进行仿真计算,并与传统方法作对比,验证了所提方法的有效性。

Abstract

Considering the uncertainty of new environmental protection and green energy such as wind power or photovoltaic power generation, the traditional deterministic calculation method of power flows is difficult to comprehensively reflect the operation of the power system. For the computational amount of traditional Monte Carlo probability power flows algorithm, combining the partial least squares regression algorithm and the multi-class proxy model, this paper proposes a probability power flow algorithm for partial least squarer polynomial sparse expansion. Using the pseudo-cross-correction error adaptive mechanism of the bias minimum square regression algorithm, this paper contributes larger polynomial to the polynomial development, and obtains a sparse expression of the polynomial expansion, overcomes the dimensional disaster facing the multi-term expansion probability power flows of the input variable. The simulation calculation is carried out in an improved IEEE-9 and IEEE-30 examples and compared with the traditional method . The reslts verify the effectiveness of the method.

导出引用

董晓阳, 梁琛, 马喜平, 李亚昕, 杨军亭. 基于偏最小二乘多项式稀疏展开的含风电电力系统概率潮流计算[J]. 太阳能学报. 2023, 44(6): 351-359 https://doi.org/10.19912/j.0254-0096.tynxb.2022-0073

Dong Xiaoyang, Liang Chen, Ma Xiping, Li Yaxin, Yang Junting. PROBABILISTIC FLOW CALCULATION OF POWER SYSTEM CONSIDERING WIND POWER BASED ON SPARSE POLYNOMIAL CHAOS EXPANSION WITH PARTIAL LEAST SQUARES METHOD[J]. Acta Energiae Solaris Sinica. 2023, 44(6): 351-359 https://doi.org/10.19912/j.0254-0096.tynxb.2022-0073

中图分类号： O325/TK79

参考文献

[1] 肖青, 周少武. 基于改进降维法的概率潮流计算[J]. 电网技术, 2018, 42(5): 1565-1575.
XIAO Q, ZHOU S W.Probabilistic load flow calculation based on improved dimension reduction method[J]. Power system technology, 2018, 42(5): 1565-1575.
[2] 柳志航, 卫志农, 孙国强, 等. 计及参数模糊性的含风电场电力系统概率潮流计算[J]. 电网技术, 2017, 41(7): 2308-2315.
LIU Z H, WEI Z N, SUN G Q, et al.Probabilistic power flow calculation of power system with wind farms considering fuzzy parameters[J]. Power system technology, 2017, 41(7): 2308-2315.
[3] 刘宇. 电力系统概率潮流算法综述[J]. 电力系统自动化, 2014, 38(23): 127-135.
LIU Y.Review on Algorithms for probabilistic power flow in power system[J]. Automation of electric power systems, 2014, 38(23): 127-135.
[4] 于晗, 钟志勇, 黄杰波, 等. 采用拉丁超立方采样的电力系统概率潮流计算方法[J]. 电力系统自动化, 2009, 33(21): 32-35.
YU H, ZHONG Z Y, HUANG J B, et al.A probabilistic load flow calculation method with latin hypercube sampling[J]. Automation of electric power systems, 2009, 33(21): 32-35.
[5] 蔡德福, 石东源, 陈金富. 基于多项式正态变换和拉丁超立方采样的概率潮流计算方法[J]. 中国电机工程学报, 2013, 33(13): 92-100.
CAI D F, SHI D Y, CHEN J F.Probabilistic load flow calculation based on polynomial normal transformation and Latin hypercube sampling[J]. Proceedings of the CSEE, 2013, 33(13): 92-100.
[6] 宋晓通, 谭震宇. 改进的重要抽样法在电力系统可靠性评估中的应用[J]. 电网技术, 2005, 29(13): 56-59.
SONG X T, TAN Z Y.Application of improved importance sampling method in power system reliability evaluation[J]. Power system technology, 2005, 29(13): 56-59.
[7] 方斯顿, 程浩忠, 徐国栋, 等. 基于Nataf变换和准蒙特卡洛模拟的随机潮流方法[J]. 电力自动化设备, 2015, 35(8): 38-44.
FANG S D, CHENG H Z, XU G D, et al.Probabilistic load flow method based on nataf transformation and quasi Monte Carlo simulation[J]. Electric power automation equipment, 2015, 35(8): 38-44.
[8] 方斯顿, 程浩忠, 宋越, 等. 基于样条重构和准蒙特卡洛的随机潮流方法[J]. 高电压技术, 2015, 41(10): 3447-3453.
FANG S D, CHENG H Z, SONG Y, et al.Probabilistic load flow method based on spline reconstruction and quasi monte carlo simulation[J]. High voltage engineering, 2015, 41(10): 3447-3453.
[9] 刘小团, 赵晋泉, 罗卫华, 等. 基于TPNT和半不变量法的考虑输入量相关性概率潮流算法[J]. 电力系统保护与控制, 2013, 41(22): 13-18.
LIU X T, ZHAO J Q, LUO W H, et al.A TPNT and cumulants based probabilistic load flow approach considering the correlation variables[J]. Power system protection and control, 2013, 41(22): 13-18.
[10] 余光正, 林涛, 徐遐龄, 等. 基于2m+1点估计法的考虑风力发电接入时电力系统谐波概率潮流算法[J]. 电网技术, 2015, 39(11): 3260-3265.
YU G Z, LIN T, XU X L, et al.An algorithm based on 2m+1 point estimate method for harmonic probabilistic load flow calculation of power systems incorporating wind power[J]. Power system technology, 2015, 39(11): 3260-3265.
[11] 孙鑫, 王博, 陈金富, 等. 基于稀疏多项式混沌展开的可用输电能力不确定性量化分析[J]. 中国电机工程学报, 2019, 39(10): 2904-2914.
SUN X, WANG B, CHEN J F, et al.Sparse polynomial chaos expansion based uncertainty quantification for available transfer capability[J]. Proceedings of the CSEE, 2019, 39(10): 2904-2914.
[12] 鲍海波, 韦化. 考虑风电的电压稳定概率评估的随机响应面法[J]. 中国电机工程学报, 2012, 32(13): 77-85.
BAO H B, WEI H.A stochastic response surface method for probabilistic evaluation of the voltage stability considering wind power[J]. Proceedings of the CSEE, 2012, 32(13): 77-85.
[13] 潘雄, 刘文霞, 徐玉琴, 等. 基于SRSM和Nataf方法的含风电场群电力系统暂态稳定分析[J]. 中国电机工程学报, 2013, 33(16): 56-62.
PAN X, LIU W X, XU Y Q, et al.Transient stability analysis of power system integrated with wind farm groups based on SRSM and nataf method[J]. Proceedings of the CSEE, 2013, 33(16): 56-62.
[14] REN Z, LI W, BILLINTON R, et al.Probabilistic power flow analysis based on the stochastic response surface method[J]. IEEE transactions on power systems, 2016, 31(3): 2307-2315.
[15] 何琨, 徐潇源, 严正, 等. 基于稀疏多项式混沌展开的孤岛微电网概率潮流计算[J]. 电力系统自动化, 2019, 43(2): 95-107.
HE K, XU X Y, YAN Z, et al.Probabilistic power flow calculation of islanded microgrid based on sparse polynomial chaos expansion[J]. Automation of electric power systems, 2019, 43(2): 95-107.
[16] 刘亮亮. 基于任意多项式逼近的不确定量化问题的压缩感知算法的研究[D]. 上海: 上海师范大学, 2017: 19-33.
LIU L L.Research on compressed sensing algorithm for uncertain quantization problem based on arbitrary polynomial approximation[D]. Shanghai: Shanghai Normal University, 2017: 19-33.
[17] JIRUTITIJAROEN P, SINGH C.Comparison of simulation methods for power system reliability indexes and their distributions[J]. IEEE transactions on power systems, 2008, 23(2): 486-493.
[18] 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.
SHI D Y, CAI D F, CHEN J F, et al.Probabilistic load flow calculation based on cumulant method considering correlation between input variables[J]. Proceedings of the CSEE, 2012, 32(28): 104-113.
[19] 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.
[20] 胡军, 张树道. 基于多项式混沌的全局敏感度分析[J]. 计算物理, 2016, 33(1): 1-14.
HU J, ZHANG S D.Global sensitivity analysis based on polynomial chaos[J]. Chinese journal of computational physics, 2016, 33(1): 1-14.
[21] ISUKAPALLI S S, ROY A, GEORGOPOULOS P G.Stochastic response surface methods(SRSMs) for uncertainty propagation: application to environmental and biological systems[J]. Risk analysis, 2010, 18(3): 351-363.
[22] 苏宏升, 董晓阳. 基于改进随机响应面法的含风电电力系统概率潮流计算[J]. 太阳能学报, 2021, 42(6): 289-296.
SU H S, DONG X Y.Probabilistic power flow calculation of power system considering wind power based on improved stochastic response surface method[J]. Acta energiae solaris sinica, 2021, 42(6): 289-296.
[23] 王惠文. 偏最小二乘回归的线性与非线性方法[M]. 北京: 国防工业出版社, 2006: 55-127.
WANG H W.Partial least-squares regression—linear and nonlinear methods[M]. Beijing: National Defense Industry Press, 2006: 55-127.
[24] 卜令泽. 全局灵敏度与结构可靠度分析[D]. 哈尔滨: 哈尔滨工业大学, 2017: 11-48.
BU L Z.Global sensitivity and structural reliability analysis: research on partial least squares-based polynomial chaos expansion method[D]. Harbin: Harbin Institute of Technology, 2017: 11-48.