基于逐步回归与深度学习的风电功率预测

Wind Power Prediction Based on Stepwise Regression and Deep Learning Model

摘要: 根据各因素对风电场输出功率的影响，本文建立了逐步回归与深度学习超短期风电场输出功率的预测模型。对比分析了风电场实际输出功率、不同高度的风速、风向数据对于预测精度的影响。最终通过逐步回归挑选出对结果有重要影响的变量作为模型的输入数据，提高预测精度。

Abstract: According to the influence of various factors on the output power of wind farms, this paper estab-lishes a stepwise regression and deep learning ultra-short-term wind farm output power prediction model. The effect of wind farm actual output power, wind speed and wind direction data at different heights to the prediction accuracy is comparatively analyzed. Finally, through stepwise regression, the variables that have an important impact on the results are selected as the input data of the model to improve the prediction accuracy.