电工技术学报  2016, Vol. 31 Issue (5): 91-98    DOI:
电力系统及其自动化 |
基于厚尾均值广义自回归条件异方差族模型的短期风电功率预测
陈昊,万秋兰,王玉荣
东南大学电气工程学院 南京 210096
Short-Term Wind Power Forecast Based on Fat-Tailed Generalized Autoregressive Conditional Heteroscedasticity-in-Mean Type Models
Chen Hao,Wan Qiulan,Wang Yurong
School of Electrical Engineering Southeast University Nanjing 210096 China
全文: PDF (2398 KB)  
输出: BibTeX | EndNote (RIS)      
摘要 风电功率预测准确度的提高对提高电力系统调度效率具有重要的作用。基于对风电功率时间序列波动性的研究,推广了一种厚尾均值广义自回归条件异方差(GARCH-M)族短期风电功率预测模型,同时,基于波动补偿项的不同形式,将模型拓展为多种类型的厚尾GARCH-M模型。该类模型能够捕捉风电功率时间序列波动性与其条件均值的直接关系,并能够有效刻画具有高峰度特征的实际风电功率序列的厚尾效应,使风电预测准确度提高。结合江苏地区风电场风电功率实际数据,对所提厚尾GARCH-M模型进行了参数估计,论证了存在于风电时间序列中的GARCH-M效应和厚尾效应,给出了风电功率均值和条件方差的预测方案。算例分析结果验证了所提方法的可行性和有效性,表明了考虑厚尾特征的GARCH-M族模型短期预测效果满意。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
陈昊
万秋兰
王玉荣
关键词 均值广义自回归条件异方差模型风电功率预测厚尾效应波动补偿系数    
Abstract:Improving the precision of wind power forecasting can be helpful to the increase of dispatch efficiency.In this paper,based on the the investigation on the intrinsic volatility of wind power time series,the fat-tail generalized autoregressive conditional Heteroscedasticity (GARCH) in mean type short term wind power forecasting models are generalized.Based on different formulations of volatility compensation items,several types of the fat-tail GARCH-M models are derived.The proposed models can capture the direct relationship between the volatility of wind power time series and its conditional mean.Furthermore,the models can depict the fat-tail effect in the practical wind power time series with leptokurtosis feature to improve the forecasting performance.In the case study,by means of the historical coastal wind power data of Jiangsu wind farm,the parameters of the proposed models are estimated,the GARCH-M effect and the fat-tail effect in the wind power time series are verified,and the conditional mean and conditional variance of the wind power are forecasted.Case study results clearly illustrate the validation and effectiveness of the proposed methods.And it is clearified that the GARCH-M model with the consideration of fat-tail effect is effective to provide satisfying forecasting results.
Key wordsGeneralized auto-regressive conditional Heteroskedasticity (GARCH)-in-mean model    wind power forecast    fat tail effect    volatility compensation coefficient   
收稿日期: 2014-09-30      出版日期: 2016-03-16
PACS: TM714  
基金资助:国家高技术研究发展计划(863计划)资助项目(2011AA05A105)。
通讯作者: 陈 昊,男,1980年生,博士,教授级高工,研究方向为风电功率预测,非线性时间序列分析等。E-mail:pingfengma@126.com     E-mail: qlwan@seu.edu.cn
作者简介: 万秋兰 女,1950年生,博士,教授,研究方向为电力系统分析与仿真。
引用本文:   
陈昊,万秋兰,王玉荣. 基于厚尾均值广义自回归条件异方差族模型的短期风电功率预测[J]. 电工技术学报, 2016, 31(5): 91-98. Chen Hao,Wan Qiulan,Wang Yurong. Short-Term Wind Power Forecast Based on Fat-Tailed Generalized Autoregressive Conditional Heteroscedasticity-in-Mean Type Models. Transactions of China Electrotechnical Society, 2016, 31(5): 91-98.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/Y2016/V31/I5/91