基于时间序列分析的风电功率预测模型

风电功率的精确预测对于电网合理调度,降低电网运行成本和保证电网系统安全性等方面有重要的意义。基于风电功率历史数据具有的时间序列特性,应用时间序列法建模,对风电场的发电功率进行短期和中长期预测,同时分析单一机组和汇聚机组对预测效果的影响。实证研究结果表明,利用时间序列分析方法,能比较精确地预测风电场发电功率,且短期内预测平均绝对百分比误差不超过15%,预测的合格率已超过85%,并且集中开发的方式可以减少预测误差。

Wind Power Forecasting Model Based on Time Series Analysis

Wind power reasonable accurate forecasting is an important part to reduce the power grid operation cost and ensure its safety in operation.It establishes a wind power prediction model based on time series analysis.Also the short-term and the mid-long-term wind farm power output are forecasted.Meanwhile,the error effect of four wind power units and fifty-eight wind turbines for wind power forecasting is analyzed respectively.The empirical research result shows that the shortterm and the mid-long-term wind power are more accurately forecasted with the method of time series analysis.The prediction of mean absolute percentage error is less than 15% in the short term and the prediction of qualified rate is more than 85%as well.It is easy to see that the prediction error of wind power will reduce in the condition of the concentrated development of the wind turbines.