引用本文: | 廖星星,吴奕,卫志农,胡伟,杨梓俊,孙国强,臧海祥.基于GMM及多点线性半不变量法的电-热互联综合能源系统概率潮流分析[J].电力自动化设备,2019,39(8): |
| LIAO Xingxing,WU Yi,WEI Zhinong,HU Wei,YANG Zijun,SUN Guoqiang,ZANG Haixiang.Probabilistic power flow analysis of integrated electricity-heat energy system based on GMM and multi-point linear cumulant methond[J].Electric Power Automation Equipment,2019,39(8): |
|
摘要: |
提出了一种计及相关性的基于高斯混合模型(GMM)的多点线性半不变量法电-热互联综合能源系统概率潮流(PPF)计算方法。构建了电-热互联综合能源系统潮流模型,采用GMM建立电-热互联综合能源系统源荷不确定模型,针对其输出变量的不确定性提出了基于半不变量法的电-热互联综合能源系统PPF计算方法,定量计算出相关输出状态变量的概率密度函数。由于电-热互联综合能源系统中存在非线性问题导致线性化误差增大,采用多点线性的方法来克服这一难题。并且所提PPF方法进一步考虑了电-热负荷之间的相关性。最后,采用Cornish-Fisher级数展开拟合状态变量概率分布,在改进的巴厘岛电-热互联综合能源系统算例中验证了所提方法的快速性、准确性和实用性。 |
关键词: 电-热互联综合能源系统 高斯混合模型 概率潮流 半不变量法 多点线性 相关性 |
DOI:10.16081/j.epae.201908004 |
分类号:TM73 |
基金项目:国家自然科学基金资助项目(51877071);国网江苏省电力有限公司科技项目(区域综合能源系统协同运行与优化调控关键技术研究及系统研发)(J2019024) |
|
Probabilistic power flow analysis of integrated electricity-heat energy system based on GMM and multi-point linear cumulant methond |
LIAO Xingxing1, WU Yi2, WEI Zhinong1, HU Wei2, YANG Zijun2, SUN Guoqiang1, ZANG Haixiang1
|
1.College of Energy and Electrical Engineering, Hohai University, Nanjing 211100, China;2.State Grid Jiangsu Electric Power Co.,Ltd.,Nanjing 210024, China
|
Abstract: |
A multi-point linear cumulant method based on GMM(Gaussian Mixture Model) considering correlation is put forward to calculate the integrated electricity-heat energy system PPF(Prababilistic Power Flow). The power flow model of integrated electricity-heat energy system is constructed, and the source-load uncertainty model of the system is proposed using GMM. In view of the uncertain output variables, the calculation method of integrated electricity-heat energy system PPF based on cumulant method is proposed, thus the probability density functions of related output state variables are calculated quantitatively. The multi-point linear method is adopted to overcome the large linearization error owing to the nonlinear problem of integrated electricity-heat energy system. Moreover, the proposed PPF method further considers the correlation between electric and thermal loads. Finally, the Cornish-Fisher series expansion is used to fit the probability distribution of state variables. The proposed method is proved to be fast, accurate and practical in the modified Bali integrated electricity-heat energy system case. |
Key words: integrated electricity-heat energy system Gaussian mixture model prababilistic power flow cumulant method multi-point linear correlation |