基于D-vine分位点回归的风电功率短期概率预测  

Short-term probability prediction of offshore wind power based on D-vine quantile regression

作　　者：李强 张伟[2] 汪惟源[1] 汪成根 郝思鹏 LI Qiang;ZHANG Wei;WANG Weiyuan;WANG Chenggen;HAO Sipeng(State Grid Jiangsu Electric Power Company Electric Power Research Institute,Nanjing 211103,China;Nanjing Instituteof Technology,Nanjing 211167,China)

机构地区：[1]国网江苏省电力公司电力科学研究院,江苏南京211103 [2]南京工程学院,江苏南京211167

出　　处：《供用电》2022年第11期93-99,共7页Distribution & Utilization

基　　金：江苏省高等学校自然科学重大项目(21KJA470005)。

摘　　要：风电存在随机性和波动性,大规模风电并网会对系统运行调度造成影响,需进行精确的风功率短期预测。在考虑风电出力时序相关性的基础上,提出一种D-vine分位点回归模型预测短期风电功率发生概率。采用D-vine结构及copula函数进行风电时序相关性建模以获得风电出力联合概率模型。采用条件分位点回归算法确定时序预测点位数并推求后序风电出力概率。针对某海上风电场进行数据测试,利用该预测方法进行验证,表明该方法对于风电出力进行概率预测是有效的。Wind power has randomness and volatility,and large-scale wind power grid integration affects system operation and dispatch,so the accurate short-term wind power forecasting is required.Based on the time series correlation of wind power output,a short-term wind power probability prediction method based on the D-vine quantile regression model is proposed.The D-vine structure and copula function are used to model the time series correlation of wind power,and the joint probability model of wind power output is obtained.The conditional quantile regression algorithm is used to determine the number of points for time series forecasting,and to calculate the probability of subsequent wind power output.Based on the data of an offshore wind farm,the proposed prediction method was tested,and the result showed the effectiveness of predicting the probability of wind power output.

关 键 词：海上风电 概率预测 分位点回归 COPULA VINE 

分 类 号：TM614[电气工程—电力系统及自动化]