基于爬坡方向状态划分的MCMC风电功率序列建模方法  

Method of MCMC wind power sequence modeling based on climbing direction state division

作　　者：崔黎丽 周云海[1] 石基辰 高怡欣 燕良坤 CUI Lili;ZHOU Yunhai;SHI Jichen;GAO Yixin;YAN Liangkun(College of Electrical Engineering and New Energy,China Three Gorges University,Yichang 443002,China)

机构地区：[1]三峡大学电气与新能源学院,湖北宜昌443002

出　　处：《现代电子技术》2024年第8期113-120,共8页Modern Electronics Technique

摘　　要：由于电网弃风或者灵活性资源不足往往发生在风电大量发电时,故提高风电时间序列模型对大出力状态的建模-抽样精度,有助于后续的电网灵活性资源相关研究。在传统马尔科夫链蒙特卡洛(MCMC)法和持续与波动蒙特卡罗(PV-MC)法基础上,提出一种考虑爬坡方向状态划分的改进方法,以更准确地描述风电出力连续爬坡至大出力状态的过程。该方法以累积分布概率而不是以功率大小均匀划分状态区间,使各个状态区间的样本分布更均匀,提高了风电时间序列模型对大出力状态的建模-抽样精度。通过算例比较所提方法、MCMC法及PV-MC法生成风电功率序列与历史数据的分布特性和统计特性指标,结果表明,所提方法的拟合度较好,且能够有效解决MCMC法和PV-MC法高出力、样本偏少的问题。As the abandonment of wind power or the lack of flexibility resources often occurs when a large amount of wind power is generated,improving the modeling-sampling accuracy of wind power time series model for large output state is helpful for subsequent research on grid flexibility resources.Based on the traditional Markov Chain Monte Carlo(MCMC)method and the persistence and variation-Monte Carlo(PV-MC)method,an improved method considering the state division of climbing direction is proposed,which can more accurately describe the process of continuous climbing of wind power output to large output state.In this method,state intervals is divided uniformly based on cumulative distribution probability rather than power size,making the sample distribution of each state interval more uniform and improving the modeling and sampling accuracy of wind power time series models for high output states.By comparing the distribution characteristics and statistical indicators of the wind power series generated by the proposed method,MCMC method,and PV-MC method with historical data through numerical examples,the results show that the proposed method has a good fit and can effectively improve the high output and small sample size problems of MCMC method and PV-MC method.

关 键 词：风力发电 风电功率时间序列 马尔科夫链蒙特卡洛法 持续与波动蒙特卡洛(PV-MC)法 爬坡方向 状态划分 累积分布概率 

分 类 号：TN911.23-34[电子电信—通信与信息系统] TM614[电子电信—信息与通信工程]