基于K-means MCMC算法的中长期风电时间序列建模方法研究
|
摘要
构建风电功率时间序列模型对电力系统中长期规划、年/月调度和安全稳定运行具有重要意义。针对传统马尔科夫链-蒙特卡洛法(Markovchain-MonteCarlo,MCMC)法存在的缺陷,提出一种基于粒子群优化的K-means MCMC风电时间序列建模新方法。首先,对历史风电功率数据进行聚类,并对聚类后的不同类别风电功率序列选取最优状态数,分别建立状态转移矩阵;其次,用拟合度较好的混合高斯分布拟合多时间尺度的风电最大波动率的概率分布特性;最后,采用基于类间转移概率矩阵的MCMC方法依次生成模拟风电出力时间序列;同时,在生成模拟序列过程中叠加高频波动分量,使模拟序列延续历史风电序列的波动特性。通过对比本所提方法和传统MCMC法分别生成的模拟风电出力序列以及历史风电功率序列,验证了所提方法的有效性和准确性。
Establishment of wind power time series model is of great significance for medium-and long-term planning, annual/monthly dispatch and safe and stable operation of power system. Aiming at the shortcomings of traditional Markov chain-Monte Carlo(MCMC) method, a new wind power time series modeling method based on K-means clustering MCMC is proposed. Firstly, the historical wind power data are clustered, and the optimal number of states is selected for different types of wind power series after clustering, and state transition matrices are established respectively. Secondly, the probability distribution characteristics of the maximum wind power volatility in multi-time scales are fitted with the better fitting mixed Gauss distribution, and the simulation sequence is improved. Finally, the MCMC method based on the transition probability matrices between classes is used to generate the simulated wind power output time series in turn. At the same time, the high-frequency fluctuation components are superimposed in the process of generating the simulation sequence, so that the simulation sequence can continue the fluctuation characteristics of the historical wind power series. Validity and accuracy of the proposed method are verified by comparing wind power series generated with the proposed method and traditional MCMC method.
Establishment of wind power time series model is of great significance for medium-and long-term planning, annual/monthly dispatch and safe and stable operation of power system. Aiming at the shortcomings of traditional Markov chain-Monte Carlo(MCMC) method, a new wind power time series modeling method based on K-means clustering MCMC is proposed. Firstly, the historical wind power data are clustered, and the optimal number of states is selected for different types of wind power series after clustering, and state transition matrices are established respectively. Secondly, the probability distribution characteristics of the maximum wind power volatility in multi-time scales are fitted with the better fitting mixed Gauss distribution, and the simulation sequence is improved. Finally, the MCMC method based on the transition probability matrices between classes is used to generate the simulated wind power output time series in turn. At the same time, the high-frequency fluctuation components are superimposed in the process of generating the simulation sequence, so that the simulation sequence can continue the fluctuation characteristics of the historical wind power series. Validity and accuracy of the proposed method are verified by comparing wind power series generated with the proposed method and traditional MCMC method.
引文
[1]刘波,贺志佳,金昊.风力发电现状与发展趋势[J].东北电力大学学报,2016,36(2):7-13.Liu Bo,He Zhijia,Jin Hao.Present situation and development trend of wind power generation[J].Journal of Northeast Dianli University,2016,36(2):7-13(in Chinese).
[2]李驰.基于波动特性的新能源出力时间序列建模方法研究[D].北京:中国电力科学研究院,2015.
[3]罗钢,石东源,陈金富,等.风光发电功率时间序列模拟的MCMC方法[J].电网技术,2014,38(2):321-327.Luo Gang,Shi Dongyuan,Chen Jinfu,et al.A Markov chain Monte Carlo method for simulation of wind and solar power time series[J].Power System Technology,2014,38(2):321-327(in Chinese).
[4]钱政,裴岩,曹利宵,等.风电功率预测方法综述[J].高电压技术,2016,42(4):1047-1060.Qian Zheng,Pei Yan,Cao Lixiao,et al.Review of wind power forecasting methods[J].High Voltage Engineering,2016,42(4):1047-1060(in Chinese).
[5]Karaki S H,Salim B A,Chedid R B.Probabilistic model of a two site wind energy conversion system[J].IEEE Transactions on Energy Conversion,2002,17(8):63-63.
[6]Papaefthymiou G,Klockl B.MCMC for wind power simulation[J].IEEE Transactions on Energy Conversion,2008,23(1):234-240.
[7]郭璋,司刚全,黄越辉,等.考虑波动趋势和相关性的多风电场模拟出力数据生成方法[J].西安交通大学学报,2018,52(4):7-14.Guo Zhang,Si Gangquan,Huang Yuehui,et al.Wind power time series modeling method for multiple wind farms considering fluctuation tendency and correlation characteristics[J].Journal of Xi’an Jiaotong University,2018,52(4):7-14(in Chinese).
[8]沈小军,周冲成,吕洪.基于运行数据的风电机组间风速相关性统计分析[J].电工技术学报,2017,32(16):265-274.Shen Xiaojun,Zhou Chongcheng,LüHong.Statistical analysis of wind speed correlation among wind turbines based on operational data[J].Transactions of China Electrotechnical Society,2017,32(16):265-274(in Chinese).
[9]李驰,刘纯,黄越辉,等.基于波动特性的风电出力时间序列建模方法研究[J].电网技术,2015,39(1):208-214.Li Chi,Liu Chun,Huang Yuehui,et al.Study on the modeling method of wind power time series based on fluctuation characteristics[J].Power System Technology,2015,39(1):208-214(in Chinese).
[10]于鹏,黎静华,文劲宇,等.含风电功率时域特性的风电功率序列建模方法[J].中国电机工程学报,2014,34(22):3715-3723.Yu Peng,Li Jinghua,Wen Jinyu,et al.A wind power time series modeling method based on its time domain characteristics[J].Proceedings of the CSEE,2014,34(22):3715-3723(in Chinese).
[11]吴涤单.基于改进粒子群的K-means算法[J].电脑知识与技术,2014(6):1238-1241.Wu Dishan.Improved K-means clustering algorithm based on particle swarm optimization[J].Computer Knowledge and Technology,2014(6):1238-1241(in Chinese).
[12]李湃,管晓宏,吴江,等.基于天气分类的风电场群总体出力特性分析[J].电网技术,2015,39(7):1866-1872.Li Pai,Guan Xiaohong,Wu Jiang,et al.Analyzing characteristics of aggregated wind power generation based on weather regime classification[J].Power System Technology,2015,39(7):1866-1872(in Chinese).
[13]Dong J,Qi M.A new algorithm for clustering based on particle swarm optimization and K-means[C]//International Conference on Artificial Intelligence&Computational Intelligence.IEEE Computer Society,2009.
[14]赵宇,肖白,顾兵,等.基于改进马尔科夫链的风电功率时间序列模型[J].电力建设,2017,38(7):18-24.Zhao Yu,Xiao Bai,Gu Bing,et al.Wind power times series model based on improved Markov chain[J].Electric Power Construction,2017,38(7):18-24(in Chinese).
[15]蒋增林,叶江明,陈昊.基于时间序列分析的负荷预测方法的比较研究[J].南京工程学院学报:自科版,2018,16(2):26-31.Jiang Zenglin,Ye Jiangming,Chen Hao.A comparative study of load forecasting methods based on time series analysis[J].Journal of Nanjing Institute of Technology(Natural Science Edition),2018,16(2):26-31(in Chinese).
[16]黄越辉,张鹏,李驰,等.基于波动划分及时移技术的多风电场出力相关性研究[J].电力自动化设备,2018,38(4):162-168.Huang Yuehui,Zhang Peng,Li Chi,et al.Research on output correlation of multi-wind farm based on wave partition and time-shift Technology[J].Electric Power Automation Equipment,2018,38(4):162-168(in Chinese).
[17]吴桐.风电功率的特性分析及其时间序列生成方法研究[D].武汉:华中科技大学,2013.
[18]崔杨,杨海威,李鸿博.基于高斯混合模型的风电场群功率波动概率密度分布函数研究[J].电网技术,2016,40(4):1107-1112.Cui Yang,Yang Haiwei,Li Hongbo.Study on the probabilistic density distribution function of power fluctuation of wind farm groups based on Gaussian mixture model[J].Power System Technology,2016,40(4):1107-1112(in Chinese).
[19]Tian P C.Estimation of wind energy potential using different probability density functions[J].Applied Energy,2011,88(5):1848-1856.
[20]Billinton R,Chen H,Ghajar R.A sequential simulation technique for adequacy evaluation of generating systems including wind energy[J].IEEE Transactions on Energy Conversion,1996,11(4):728-734.
[21]Liu C,Li C,Huang Y,et al.A novel stochastic modeling method of wind power time series considering the fluctuation process characteristics[J].Journal of Renewable and Sustainable Energy,2016,8(3):033304.
[22]吕颖,鲁广明,谢昶,等.考虑大规模集中接入风电功率波动相关性的在线概率安全评估[J].电网技术,2018,42(4):1140-1148.LüYing,Lu Guangming,Xie Chang,et al.Online probabilistic security assessment considering centralized integration of large scale wind power[J].Power System Technology,2018,42(4):1140-1148(in Chinese).
[23]刘纯,吕振华,黄越辉,等.长时间尺度风电出力时间序列建模新方法研究[J].电力系统保护与控制,2013,41(1):7-13.Liu Chun,Lu Zhenhua,Huang Yuehui,et al.A new method to simulate wind power time series of large time scale[J].Power System Protection and Control,2013,41(1):7-13(in Chinese).
[2]李驰.基于波动特性的新能源出力时间序列建模方法研究[D].北京:中国电力科学研究院,2015.
[3]罗钢,石东源,陈金富,等.风光发电功率时间序列模拟的MCMC方法[J].电网技术,2014,38(2):321-327.Luo Gang,Shi Dongyuan,Chen Jinfu,et al.A Markov chain Monte Carlo method for simulation of wind and solar power time series[J].Power System Technology,2014,38(2):321-327(in Chinese).
[4]钱政,裴岩,曹利宵,等.风电功率预测方法综述[J].高电压技术,2016,42(4):1047-1060.Qian Zheng,Pei Yan,Cao Lixiao,et al.Review of wind power forecasting methods[J].High Voltage Engineering,2016,42(4):1047-1060(in Chinese).
[5]Karaki S H,Salim B A,Chedid R B.Probabilistic model of a two site wind energy conversion system[J].IEEE Transactions on Energy Conversion,2002,17(8):63-63.
[6]Papaefthymiou G,Klockl B.MCMC for wind power simulation[J].IEEE Transactions on Energy Conversion,2008,23(1):234-240.
[7]郭璋,司刚全,黄越辉,等.考虑波动趋势和相关性的多风电场模拟出力数据生成方法[J].西安交通大学学报,2018,52(4):7-14.Guo Zhang,Si Gangquan,Huang Yuehui,et al.Wind power time series modeling method for multiple wind farms considering fluctuation tendency and correlation characteristics[J].Journal of Xi’an Jiaotong University,2018,52(4):7-14(in Chinese).
[8]沈小军,周冲成,吕洪.基于运行数据的风电机组间风速相关性统计分析[J].电工技术学报,2017,32(16):265-274.Shen Xiaojun,Zhou Chongcheng,LüHong.Statistical analysis of wind speed correlation among wind turbines based on operational data[J].Transactions of China Electrotechnical Society,2017,32(16):265-274(in Chinese).
[9]李驰,刘纯,黄越辉,等.基于波动特性的风电出力时间序列建模方法研究[J].电网技术,2015,39(1):208-214.Li Chi,Liu Chun,Huang Yuehui,et al.Study on the modeling method of wind power time series based on fluctuation characteristics[J].Power System Technology,2015,39(1):208-214(in Chinese).
[10]于鹏,黎静华,文劲宇,等.含风电功率时域特性的风电功率序列建模方法[J].中国电机工程学报,2014,34(22):3715-3723.Yu Peng,Li Jinghua,Wen Jinyu,et al.A wind power time series modeling method based on its time domain characteristics[J].Proceedings of the CSEE,2014,34(22):3715-3723(in Chinese).
[11]吴涤单.基于改进粒子群的K-means算法[J].电脑知识与技术,2014(6):1238-1241.Wu Dishan.Improved K-means clustering algorithm based on particle swarm optimization[J].Computer Knowledge and Technology,2014(6):1238-1241(in Chinese).
[12]李湃,管晓宏,吴江,等.基于天气分类的风电场群总体出力特性分析[J].电网技术,2015,39(7):1866-1872.Li Pai,Guan Xiaohong,Wu Jiang,et al.Analyzing characteristics of aggregated wind power generation based on weather regime classification[J].Power System Technology,2015,39(7):1866-1872(in Chinese).
[13]Dong J,Qi M.A new algorithm for clustering based on particle swarm optimization and K-means[C]//International Conference on Artificial Intelligence&Computational Intelligence.IEEE Computer Society,2009.
[14]赵宇,肖白,顾兵,等.基于改进马尔科夫链的风电功率时间序列模型[J].电力建设,2017,38(7):18-24.Zhao Yu,Xiao Bai,Gu Bing,et al.Wind power times series model based on improved Markov chain[J].Electric Power Construction,2017,38(7):18-24(in Chinese).
[15]蒋增林,叶江明,陈昊.基于时间序列分析的负荷预测方法的比较研究[J].南京工程学院学报:自科版,2018,16(2):26-31.Jiang Zenglin,Ye Jiangming,Chen Hao.A comparative study of load forecasting methods based on time series analysis[J].Journal of Nanjing Institute of Technology(Natural Science Edition),2018,16(2):26-31(in Chinese).
[16]黄越辉,张鹏,李驰,等.基于波动划分及时移技术的多风电场出力相关性研究[J].电力自动化设备,2018,38(4):162-168.Huang Yuehui,Zhang Peng,Li Chi,et al.Research on output correlation of multi-wind farm based on wave partition and time-shift Technology[J].Electric Power Automation Equipment,2018,38(4):162-168(in Chinese).
[17]吴桐.风电功率的特性分析及其时间序列生成方法研究[D].武汉:华中科技大学,2013.
[18]崔杨,杨海威,李鸿博.基于高斯混合模型的风电场群功率波动概率密度分布函数研究[J].电网技术,2016,40(4):1107-1112.Cui Yang,Yang Haiwei,Li Hongbo.Study on the probabilistic density distribution function of power fluctuation of wind farm groups based on Gaussian mixture model[J].Power System Technology,2016,40(4):1107-1112(in Chinese).
[19]Tian P C.Estimation of wind energy potential using different probability density functions[J].Applied Energy,2011,88(5):1848-1856.
[20]Billinton R,Chen H,Ghajar R.A sequential simulation technique for adequacy evaluation of generating systems including wind energy[J].IEEE Transactions on Energy Conversion,1996,11(4):728-734.
[21]Liu C,Li C,Huang Y,et al.A novel stochastic modeling method of wind power time series considering the fluctuation process characteristics[J].Journal of Renewable and Sustainable Energy,2016,8(3):033304.
[22]吕颖,鲁广明,谢昶,等.考虑大规模集中接入风电功率波动相关性的在线概率安全评估[J].电网技术,2018,42(4):1140-1148.LüYing,Lu Guangming,Xie Chang,et al.Online probabilistic security assessment considering centralized integration of large scale wind power[J].Power System Technology,2018,42(4):1140-1148(in Chinese).
[23]刘纯,吕振华,黄越辉,等.长时间尺度风电出力时间序列建模新方法研究[J].电力系统保护与控制,2013,41(1):7-13.Liu Chun,Lu Zhenhua,Huang Yuehui,et al.A new method to simulate wind power time series of large time scale[J].Power System Protection and Control,2013,41(1):7-13(in Chinese).
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |