基于Copula函数的风-电-热相关性及其潜在不确定性分析*

doi:10.11985/2020.02.007

摘要/Abstract

摘要：

在电力系统经济调度中,考虑热负荷、电负荷和风电出力之间的相关性对节省发电成本和实现风电合理消纳是必要且有效的。为此,基于Copula相关性分析理论,提出多元Copula分析工具箱(Multivariate Copula analysis toolbox,MvCAT),它包含各种复杂程度不同的Copula族,采用基于残差高斯似然函数的贝叶斯框架来推断Copula参数和估计潜在的不确定性。在贝叶斯框架内进行模型的推断,并利用混合进化马尔可夫链蒙特卡罗(Markov chain Monte Carlo,MCMC)方法进行后验分布的数值估计和Copula参数的生成。根据模拟的双变量概率与其经验观测值的接近程度来评价Copula的性能,选取最优Copula模型。最后,以某省的数据为样本进行分析,结果表明提出的方法可以很好地反映变量间的秩相关性,并且能定量地评估与数据长度相关的不确定性。

关键词: 相关性, 不确定性, Copula函数, MvCAT

Abstract:

In economic dispatch of power system, it is necessary and effective to consider the correlation between heat load, electric load and wind power output to save power generation costs and achieve reasonable wind power consumption. Therefore, based on the theory of Copula correlation analysis, a multivariate Copula analysis toolbox (MvCAT) is proposed, which contains various Copula families with different degrees of complexity. The Bayesian framework based on residual Gaussian likelihood function is used to infer Copula parameters and estimates potential uncertainty. First, the model is inferred in the Bayesian framework, and the hybrid evolutionary Markov chain Monte Carlo (MCMC) method is used to estimate the posterior distribution and generate the Copula parameters. Then the performance of Copula is evaluated based on how close the simulated bivariate probability is to its empirical observations, and the optimal Copula model is selected. Finally, the data from a certain province is used as a sample for analysis. The results show that the proposed method can well reflect the rank correlation between variables, and the uncertainty associated with data length can be assessed quantitatively.

Key words: Correlation, uncertainty, Copula function, MvCAT

中图分类号:

TM76

杨柳, 杨德友, 张宇时, 许小鹏, 李典阳. 基于Copula函数的风-电-热相关性及其潜在不确定性分析^*[J]. 电气工程学报, 2020, 15(2): 47-53.

YANG Liu, YANG Deyou, ZHANG Yushi, XU Xiaopeng, LI Dianyang. Wind-power-heat Correlation and Its Underlying Uncertainty Analysis Based on Copula Function[J]. Journal of Electrical Engineering, 2020, 15(2): 47-53.

图/表 8

图1

图2

表1

表2

图3

图4

图5

图6

参考文献 17

[1]	谢敏, 熊靖, 刘明波, 等. 基于Copula的多风电场出力相关性建模及其在电网经济调度中的应用[J]. 电网技术, 2016,40(4):1100-1106.
	Xie Min, Xiong Jing, Liu Mingbo, et al. Modeling of multi wind farm output correlation based on Copula and its application in power system economic dispatch[J]. Power System Technology, 2016,40(4):1100-1106.
[2]	田浩, 刘文颖, 王方雨, 等. 基于可调节负荷参与的源荷互动调峰多目标优化方法[J]. 电网与清洁能源, 2018,34(11):32-38.
	Tian Hao, Liu Wenying, Wang Fangyu, et al. A multi- objective optimization method based on source-charge interaction peak-shaping based on adjustable load participation[J]. Power System and Clean Energy, 2018,34(11):32-38.
[3]	Liu Huahan, Jiang Wei, Yang Guanghui, et al. Single failure mode reliability analysis of mechanical com- ponents under dependent random variables with non- normal distribution[J]. Journal of Dalian University of Technology, 2015,55(1):22-30.
[4]	张学功, 薛志超, 吕龙. 基于Pair Copula-SV-t模型的金融市场相关性分析[J]. 武汉金融, 2016,4(9):71-76.
	Zhang Xuegong, Xue Zhichao, Lü Long. Financial market correlation analysis based on Pair Copula-SV-t model[J]. Wuhan Finance, 2016,4(9):71-76.
[5]	Bin L, Shahzad M, Bing Q, et al. Probabilistic computational model for correlated wind farms using copula theory[J]. IEEE Access, 2018(6):14179-14187.
[6]	Jongman B, Hochrainer stigler S, Feyen L, et al. Increasing stress on disaster-risk finance due to large floods[J]. Nature Climate Change, 2014,4(4):264-268.
[7]	Éric P, Favre A C, Bernier J, et al. Copula models for frequency analysis what can be learned from a Bayesian perspective[J]. Advances in Water Resources, 2014,63(2):91-103.
[8]	Cai Fei, Yan Zheng, Zhou Jingbo, et al. Dependence structure models for wind speed and wind power among different wind farms based on Copula theory[J]. Auto- mation of Electric Power Systems, 2013,37(17):9-16.
[9]	段偲默, 苗世洪, 霍雪松, 等 . 基于动态 Copula 的风光联合出力建模及动态相关性分析[J]. 电力系统保护与控制, 2019,47(5):35-42.
	Duan Simo, Miao Shihong, Huo Xuesong, et al. Modeling and dynamic correlation analysis of wind/solar power joint output based on dynamic Copula[J]. Power System Protection and Control, 2019,47(5):35-42.
[10]	马建鹏, 李文升, 安树怀, 等. 基于Copula理论的综合能源系统负荷相关性研究[J]. 青岛大学学报, 2018,33(4):60-64.
	Ma Jianpeng, Li Wensheng, An Shuhuai, et al. Research on load correlation of integrated energy system based on Copula theory[J]. Journal of Qingdao University, 2018,33(4):60-64.
[11]	Yang P, Libao S, Yixin N. Modelling of multiple wind farms output correlation based on Copula theory[J]. The Journal of Engineering, 2017,2017(13):2303-2308.
[12]	Grimaldi S, Petroselli A, Salvadori G, et al. Catchment compatibility via copulas:A non-parametric study of the dependence structures of hydrological responses[J]. Advances in Water Resources, 2016(90):116-133.
[13]	王俊, 蔡兴国, 季峰. 基于Copula理论的相关随机变量模拟方法[J]. 中国电机工程学报, 2013,33(22):75-82.
	Wang Jun, Cai Xingguo, Ji Feng. A simulation method of correlated random variables based on Copula[J]. Proceedings of the CSEE, 2013,33(22):75-82.
[14]	张冬冬, 鲁帆, 刘少华. 基于贝叶斯理论的洪水频率不确定性分析[J]. 中国农村水利水电, 2016(4):96-102.
	Zhang Dongdong, Lu Fan, Liu Shaohua. An analysis of flood frequency uncertainty based on Bayesian theory[J]. China Rural Water and Hydropower, 2016(4):96-102.
[15]	张彦龙, 翟登辉, 许丹, 等. 基于机器学习的换流站故障分析报告智能分类方法[J]. 电气工程学报, 2019,14(1):83-88.
	Zhang Yanlong, Zhai Denghui, Xu Dan, et al. Intelligent classification method of converter station fault analysis report based on machine learning[J]. Journal of Electrical Engineering, 2019,14(1):83-88.
[16]	史会峰. 基于贝叶斯理论MCMC优化参数的负荷预测模型[D]. 北京:华北电力大学, 2013.
[17]	马晓晓. 基于Copula函数的不完全降水序列频率计算方法研究[D]. 杨凌:西北农林科技大学, 2017.

排序	Max-Likelihood	AIC	BIC
1	Linear-Spearman	Linear-Spearman	Linear-Spearman
2	Marshal-Olkin	Marshal-Olkin	Marshal-Olkin
3	Tawn	Tawn	Tawn
4	Cubic	Cubic	Cubic
5	Raftery	Raftery	Raftery