风电机组功率曲线是风电机组重要的性能指标,表征了机组的实际运行状态.准确的实测风电功率曲线可以为风电机组性能评估、风电功率曲线监测、风电功率预测、风电场数值建模等工作提供重要的参考依据.但采用耗时的功率曲线建模方法会花费大量的建模时间,从而影响建模效率.文章从功率曲线建模的数据筛选和功率曲线拟合入手,选取耗时较短的二维核密度估计模型筛选风速功率散点集中区域内的正常运行数据,并选用5种功率曲线拟合方法对正常风速功率数据进行拟合.5种模型的建模精度和建模效率对比分析表明,多项式拟合方法原理简单,拟合速度最快,且拟合精度较高,比较适用于实际功率曲线的建模工作.%Wind turbine power curve is an important performance index of the wind turbine, which represents the actual running state of the wind turbine unit.Accurate wind power curve can provide an important reference for the performance evaluation of wind turbines, wind power curve monitoring, wind power forecasting, and wind farm equivalent modeling and so on. Due to the large number of wind turbine units in large wind farms, it will be a tedious task to compare the wind turbine power curves of different units, different years and months when the wind farm is running for a long time. If the time-consuming power curve modeling method is used, a large amount of modeling time will be spent. In this paper, a time-saving kernel density estimation model is used to filter the normal operation data in the wind power concentration area, then five power curve fitting methods are adopted to fit the normal wind speed power data. The modeling accuracy and modeling efficiency of the five models are compared and analyzed.Modeling precision and efficiency comparison of the five models show that the polynomial fitting method is easy to understand, and it has the fast fitting speed with high precision. Polynomial fitting power curve model is more suitable for the actual work.
展开▼
