摘要
随着风电在现代电网的渗透率越来越高,电力系统优化运行对风电功率区间预测的可靠性提出了更高要求。现有的风电功率区间预测通常针对历史数据整体的误差,或者基于不同的出力水平进行分类误差建模,难以反映预测模型对于不同风况下的适应性。鉴于此,提出了一种基于数值天气预报(NWP)风速和蒙特卡洛法的短期风电功率区间预测模型。首先,按照NWP风速对历史时段的点预测误差进行层次聚类,利用经验分布模型对不同风况下的误差进行概率分布拟合。然后,对待预测时刻的NWP风速所对应的累计经验分布概率值进行蒙特卡洛抽样,并在给定的置信水平下求取短期内各个待预测时点可能发生的功率波动区间。最后,以中国吉林省某风电场运行数据为例,与常用的概率预测方法相比,验证了所提方法的可靠性。
With the increasing penetration rate of wind power in modern power grid,the optimal operation of power systems has higher requirements for the reliability of wind power interval prediction.The existing wind power interval prediction usually aims at the overall error of the historical data or different output levels to perform classification error modeling,which is difficult to reflect the adaptability of the prediction model to different wind conditions.Based on this,this paper proposes a short-term wind power interval prediction model based on numerical weather prediction(NWP)wind speed and Monte Carlo method.First,according to the NWP wind speed,the point prediction error of the historical period is hierarchically clustered,and the empirical distribution model is used to fit the probability distribution of errors with different wind conditions.Then,Monte Carlo sampling is performed on the value of the cumulative empirical distribution probability corresponding to the NWP wind speed at the prediction moment,and at a given confidence level,the power fluctuation interval that may occur at each predicted point in the short term is solved.Finally,taking the operation data of a wind farm in Jilin Province of China as an example,compared with the commonly used probability prediction method,the reliability of the proposed method is verified.
作者
杨茂
董昊
YANG Mao;DONG Hao(Key Laboratory of Modem Power System Simulation and Control&Renewable Energy Technology,Ministry of Education(Northeast Electric Power University),Jilin 132012,China)
出处
《电力系统自动化》
EI
CSCD
北大核心
2021年第5期79-85,共7页
Automation of Electric Power Systems
基金
国家重点研发计划资助项目(2018YFB0904200)。
关键词
层次聚类
经验分布
蒙特卡洛法
短期风电功率区间预测
hierarchical clustering
empirical distribution
Monte Carlo method
short-term wind power interval prediction
