太阳能Issue(7)：62-70,9.DOI:10.19911/j.1003-0417.tyn20240703.01
基于Bootstrap方法对测风数据插补不确定性的分析研究
ANALYSIS AND RESEARCH ON UNCERTAINTY IN WIND MEASUREMENT DATA INTERPOLATION BASED ON Bootstrap METHOD
摘要
Abstract
In order to accurately evaluate the wind energy resources in the wind farm area,it is necessary to interpolate and extend the missing and unreasonable data of the anemometer tower.The measure-correlate-predict(MCP)interpolation method uses short-term wind measurement data and long-term reference data from anemometer towers to interpolate and extend missing and unreasonable data.However,this can result in data processing uncertainty and affect the accuracy of subsequent wind farm power generation capacity calculations.Therefore,it is necessary to analyze the interpolation uncertainty.This paper collected raw wind measurement data from two anemometer towers located in different regions and terrains of China.The commonly used LLS method in MCP interpolation is used to interpolate the wind measurement data,a quantitative analysis is conducted on the interpolation uncertainty of wind measurement data brought by LLS method based on Bootstrap method.The impact of different wind measurement durations on the uncertainty of wind measurement data interpolation is also analyzed.The research results show that:1)The residual self-help method can effectively estimate the uncertainty of the interpolation parameters of the LLS method,and the uncertainty of the interpolation parameters decreases continuously with the increase of wind measurement durations.2)The interpolation results of LLS method are greatly affected by the duration of wind measurement.As the duration of wind measurement increases,the interpolation uncertainty of wind measurement data brought by LLS method gradually decreases.The higher the correlation coefficient between the wind measurement data from the anemometer tower and the reference data,the lower the interpolation uncertainty of the wind measurement data brought by the LLS method.3)The uncertainty of the interpolation parameters in the LLS method contributes relatively little to the overall prediction uncertainty,resulting in a large number of measured wind speeds falling outside the 95%confidence interval.Therefore,further analysis is needed on the uncertainty of wind speed measurement and the impact of different MCP interpolation methods on the interpolation uncertainty of wind measurement data.关键词
测风数据/风电场/测量-关联-预测插补方法/Bootstrap方法/线性最小二乘法/不确定性Key words
wind measurement data/wind farms/MCP interpolation method/Bootstrap method/LLS method/uncertainty分类
能源科技引用本文复制引用
刘松,李伟岩..基于Bootstrap方法对测风数据插补不确定性的分析研究[J].太阳能,2025,(7):62-70,9.基金项目
国家自然科学基金资助项目(61803393) （61803393）
湖南省战略性新兴产业-科技攻关与重大科技成果转化专项项目(2018GK4002) （2018GK4002）
