考虑海上风电多机组时空特性的超短期功率预测模型

Ultra-short-term Power Prediction Model Considering Spatial-Temporal Characteristics of Offshore Wind Turbines

海上风电单机预测无法快速预测海上风电场整体功率,且各机组间波动不均会造成集群功率曲线质量差、预测精度低的问题。因此,文中提出考虑海上风电机组时空特性的超短期功率预测模型。首先,利用改进动态时间弯曲算法量化度量海上风电机组时空特性相似度,分析海上风电机组的时空特性;然后,采用基于深度学习的Transformer模型建立海上风电功率预测模型;最后,综合考虑海上风电机组时空特性相似度与母线位置信息,聚类海上风电机组并进行超短期功率预测。通过对海上风电机组实测数据的分析表明,所提出的方法可有效量化与度量海上风电机组间时空特性并及时预测超短期海上风电机组群功率。

The single-unit prediction of offshore wind power cannot quickly predict the overall power of an offshore wind farm,and uneven fluctuations among units lead to poor quality of cluster power curves and low prediction accuracy.Therefore,this paper proposes an ultra-short-term power prediction model considering the spatial-temporal characteristics of offshore wind turbines.First,the improved dynamic time warping(DTW)algorithm is used to quantify the spatial-temporal characteristic similarity of offshore wind turbines and analyze the spatial-temporal characteristics of offshore wind turbines.Then,the Transformer model based on deep learning is used to establish the offshore wind power prediction model.Finally,the spatial-temporal characteristic similarity of offshore wind turbines and the bus position information are comprehensively considered to cluster the offshore wind turbines and conduct the ultra-short-term power prediction.The analysis of measured data of offshore wind turbines show that the proposed method can effectively quantify and measure the spatial-temporal characteristics of offshore wind turbines and timely predict the ultra-short-term power of offshore wind turbine groups.