基于改进灰色系统模型的风电功率预测研究

Wind turbine power prediction research based on improved grey system model

文章提出了基于灰色模型的超短期风电功率预测方法。灰色预测模型需要的运算数据较少,计算时间短,适于风电系统在线预测。首先,对GM(1,1)模型、灰色Verhulst模型和DGM模型预测步骤及理论公式进行阐述,并分析了相互关系及适用范围;其次,在运用改进模型进行功率预测时提出滚动预测机制,进行等维数据更新;最后,针对模型优化问题,讨论所取不同原始数据点数对预测精度的影响,并且研究了不同初始条件和不同背景值对预测误差的影响。以某风电场实际风机运行数据为研究对象,进行了风电功率预测,结果表明,模型所选数据点数和背景值提高了预测精度,改善了预测性能。

A new method based on grey model for ultra-short term wind turbine power prediction is presented. The grey prediction model is suitable for wind power system online operation because of few modeling data and fast computation. First of all,the procedure and mathematical formulation of GM(1,1), grey Verhulst and discrete grey model are also clarified. The relationship among these models and the scope of application are analysed. Secondly, when the wind power is predicted by the improved model, the rolling mechanism updating data in the way of equal dimension is proposed in order to rapidly track the data change. Finally, as to the problem of the model optimization,the effect on the prediction accuracy is discussed according to the different numbers of raw data used in the modeling.Also the influence on prediction error is studied by setting various boundary conditions and background values. The actual data of wind turbine’s output power in some wind farm is selected for prediction test.The results demonstrate the enhancement of prediction performance due to the selected parameters in the paper such as the number of data points and the background value.