基于数据修正的概率稀疏自注意短期风电功率预测

Short-Term Wind Power Prediction Based on Data Correction with Probabilistic Sparse Self-Attention

风电功率的短期精确预测对提高风能利用效率具有重大意义。提出了一种基于数据修正的概率稀疏自注意风电功率预测方法。首先,将来自风电场采集的风电数据进行清洗、缺失值重构;然后采用Informer预测模型,加入概率稀疏自注意、蒸馏机制,使模型的训练速度与预测的精度都得到了有效提升。结果表明,本文所提出的方法能够提高数据质量,且拥有更高的预测效率和精度。

Accurate short-term wind power forecasting is of great importance for enhancing the efficiency of wind energy utilization.The paper proposes a wind power forecasting method that incorporates data correction and probabilistic sparse self-attention.Firstly,wind power data from wind farms is cleaned and the missing values are reconstructed.Then the Informer forecasting model is used to enhance both the training velocity and prediction accuracy of the model by integrating the probabilistic sparse self-attention and a distillation mechanism.The case analysis shows that the proposed method can improve the data quality while simultaneously achieving high prediction efficiency and precision.