基于不同风切变指数算法推算风机轮毂高度风速

Calculation of Wind Speed at Hub Height of Wind Turbines Based on Different Wind Shear Exponent Algorithms

为了将地面10m风速推算至风机轮毂高度,使用2017—2018年内蒙古华电玫瑰营风电场10、70m高度逐小时测风数据,通过分季节、分时次、分风速段、分风向和大气稳定度等级等分类方法计算风切变指数并检验,得到如下结论:(1)风切变指数的季节变化表现为冬季>春季>秋季>夏季,具有先减后增的日变化特征,在10—16时风切变指数较小。(2)大气越不稳定、风速越大,风切变指数越小,推算效果越好。(3)在6种算法中,分时次计算风切变指数的推算效果最佳,尤其在10—18时计算值和观测值的标准化平均误差不超过10.0%;其次为分风速段算法,中风速段(3.0<V_(10)≤12.0 m·s^(-1))和大风速段(12.0<V_(10)≤25.0 m·s^(-1))推算效果较好,标准化平均误差分别为11.3%和3.6%。建议推算风机轮毂高度风速时,优先考虑分时次和分风速段计算风切变指数。

In order to calculate the wind speed at hub height of the wind turbine from the 10 m wind speed observed on the ground,the hourly wind observation data of Inner Mongolia Huadian Meiguiying Wind Farm from 2017 to 2018 at the height of 10 m and 70 m were used to calculate and test the wind shear exponents by classification algorithms of seasons,times,wind speed,wind direction and atmospheric stability.The conclusions were as follows:(1)The seasonal variation of the wind shear exponents was winter>spring>autumn>summer,with diurnal variation characteristics of first decreasing and then increasing,and wind shear exponents always remained at a relatively low level from 10:00 to 16:00.(2)The more unstable the atmosphere,the greater the wind speed.And the smaller the wind shear exponent,the better the calculation effect.(3)Among these six algorithms,the time-divided classification algorithms of calculating wind shear exponents was the best,especially at 10:0018:00,the standardized average error of the calculated value and the observed value was no more than 10%.The second best algorithm was the wind speed-divided algorithm.The results were better in the medium wind speed segment(3.0<V_(10)≤12.0 m·s^(-1))and the large wind speed segment(12.0<V_(10)≤25.0 m·s^(-1)),with the standardized average errors being 11.3%and 3.6%,respectively.Therefore,when calculating the wind speed at the hub height of the wind turbines,it should be preferred to calculate the wind shear exponents by time and wind speed classification algorithms.