考虑加权理论的风电场集群风速预测方法

Wind speed prediction method for wind farm cluster considering weighted theory

风能发电极易受风的不规则波动影响,干预了电力系统的安全稳定运行。为解决该问题,提出一种考虑加权理论的风电场集群风速预测方法。首先采用综合加权理论进行数据预处理,随后采用聚类算法将物理特性相似的风电机组构成集群,最后根据长短期记忆网络模型得到预测结果。针对风电机组运行数据中各指标权重对聚类结果影响相同的现象,基于综合加权理论对数据进行了预处理,在赋予各指标物理意义的同时有效改善了聚类效果,进而提升了预测精度。基于风电场真实监测数据对该方法的验证表明,所提出的方法具有较高的预测精度。

Wind power generation electrodes are susceptible to irregular wind fluctuations, which interferes with the safe and stable operation of the power system. To solve this problem, a wind speed prediction method for wind farm clusters considering weighting theory is proposed. The proposed method first uses comprehensive weighting theory for data preprocessing, then uses a clustering algorithm to form clusters of wind turbines with similar physical characteristics, and finally obtains the prediction results according to the long-and short-term memory network model. Aiming at the phenomenon that the weight of each index in the wind turbine operating data has the same effect on the clustering results, the data is preprocessed based on the comprehensive weighting theory, which effectively improves the clustering effect while giving the indicators physical meaning, thereby improving the prediction accuracy. The verification of this method based on the real monitoring data of wind farms shows that the proposed method has high prediction accuracy.