测风塔异常风速数据识别与补齐方法研究

Study on identification and completion method of abnormal wind speed data of wind tower

测风数据能够真实客观地反映该区域内风能情况,数据质量对计算风电场理论出力有重要意义。文章根据异常风速数据产生的原因及特点,提出了一种基于最小二乘滤波-肖维勒组合的风速数据识别算法,并利用测风塔不同高度的风速数据波动关联性的特点,对待剔除的数据进行校正。在数据缺失情况下,提出了基于属性重要度-相似片段的补齐方法,得到完整风速数据。对比分析了常用异常数据识别和补齐方法,结果表明,文章所提方法可有效剔除并重构异常数据,对不同风电场有较强的通用性,具有一定的工程实用价值。

The wind measurement data can truly and objectively reflect the wind energy situation in the area,and the data quality is of great significance for calculating the theoretical output of wind farms.According to the reasons and characteristics of abnormal wind speed data,this paper proposes a wind speed data recognition algorithm based on the least squares filtering-chauvenet criterion combination.And the wind speed data at different heights of the wind measuring tower is characterized by fluctuation correlation,and the data to be eliminated is corrected.In the case of missing data,a complementing method based on attribute importance-similar segments is proposed to obtain complete wind speed data.A comparative analysis of commonly used abnormal data identification and complementation methods shows that the method proposed in the article can effectively eliminate and reconstruct abnormal data,has strong versatility for different wind farms,and has certain engineering practical value.