基于风特征分析的风电机组异常数据识别算法

An Algorithm for Abnormal Data Identification of Wind Turbine Based on Wind Characteristic Analysis

对风电的研究往往要依托于历史功率数据,而风电机组采集到的历史数据中往往含有大量的异常数据,这严重影响了对风电功率规律特性的分析。针对风电机组的实测功率数据进行研究,分析风速升降特征与风向特征对风电机组输出功率的影响。将不同的风特征的数据分开讨论,分别利用Copula函数得到概率功率曲线,结合异常数据的时序特征归纳出三类异常数据,建立异常数据识别模型。利用风电机组的实际数据和人工生成数据进行仿真分析,结果表明,该方法能够高效地识别各类异常数据,对风电研究有着重要的意义。

The study of wind power often depends on historical power data, and the historical data collected by wind turbine often contains a lot of abnormal data, which seriously affects the analysis of wind power characteristics. According to the measured power data of wind turbines, the influence of RISE-FALL-Feature of wind speed and wind direction characteristics on the output power of wind turbine was analyzed, and the data of different wind characteristics were discussed separately. Using Copula function to get the probability power curve, three types of anomaly data were summed up according to the timing characteristics of anomaly data and the exception data recognition model was established. The actual data and artificial date of wind turbine were used for simulation analysis. The results show that the proposed method can identify all kinds of anomaly data efficiently, which is of great significance to wind power research.