基于转译模型的风力发电机风速时间序列模拟  

Wind Speed Time Series Simulation of Wind-driven Generator Based on Translation Model

作　　者：孙明涛 朱子莉 肖青 彭嘉兴 SUN Mingtao;ZHU Zili;XIAO Qing;PENG Jiaxing(School of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China;School of Information and Electrical Engineering,Hunan University of Science and Technology,Xiangtan 411201,China)

机构地区：[1]华北电力大学控制与计算机工程学院,北京102206 [2]湖南科技大学信息与电气工程学院,湖南湘潭411201

出　　处：《湖南科技大学学报(自然科学版)》2024年第4期69-76,共8页Journal of Hunan University of Science And Technology:Natural Science Edition

基　　金：国家自然科学基金资助项目(12271155)。

摘　　要：风速是一个有着较强非正态性的时间序列,在拟合风速历史观测样本时,难以兼顾非正态性和自相关性,特别是当自相关函数的振荡性较强时,现有方法的拟合精度不高.选用转译模型模拟风速时间序列,将概率分布函数和自相关函数的拟合解耦成2个独立的部分.首先,基于分位数函数和原点矩抽取风速样本的统计信息,选用Tadikamalla分布重构风速的累积分布函数和分位数函数;然后,基于边际概率转换,将风速样本映射到标准正态空间,采用余弦函数级数和来拟合样本的自相关函数;最后,提出一种三角级数展开式的方法生成高斯时间序列样本,利用转译模型将高斯时间序列样本转换成风速时间序列样本,复现风速的非正态累积分布函数和振荡自相关函数.对3组风速观测样本进行拟合测试,结果验证了所提方法的有效性和准确性.The wind speed is a time series with strong non-normality.When fitting historical wind speed observation samples,it is difficult to give consideration to both non-normality and autocorrelation;especially,when the autocorrelation function has a strong oscillation,the fitting accuracy of existing methods is not high.In this paper,a translation model is used to simulate wind speed time series,whereby the fitting of the probability distribution function and the autocorrelation function are decoupled into two independent parts.Firstly,the quantile and raw moment are employed to extract statistical information of wind speed samples,and Tadikamalla distribution is introduced to recover the cumulative distribution function and quantile function of wind speed.Then,based on marginal transformation,wind speed samples are mapped to standard normal space,where a finite cosine series sum is proposed to fit the autocorrelation function of samples.Finally,a trigonometric series expansion is developed to generate samples of a Gaussian time series,which is transformed to samples of wind speed time series by translation model.A fitting test is performed on three sets of wind speed observation samples,and the results of case study demonstrate the effectiveness and accuracy of the proposed method.

关 键 词：风速时间序列 转译模型 Tadikamalla分布 自相关函数 

分 类 号：TM753[电气工程—电力系统及自动化]