考虑网内外功率特性的含风场电网日内发电需求滚动预测

摘要:

由于自然风力、风向的不确定，大规模风场并网运行使得电网和风电调度的难度增大，在最大功率跟踪模式 下，更加准确地预测发电出力是重要课题。考虑风场传统日前负荷预测时间尺度较大导致风场实时调度困难的实 际，提出了一种在基于日内风场出力的发电需求滚动预测方法，以多级时间尺度为预测依据，建立了考虑风电接入的 理想发电需求预测模型，提出了该模型下的滚动预测算法，并结合实际风场数据进行了实际预测，结果表明，提出的 预测方法、模型和算法能明显提高发电需求预测精度，为大型风场的科学调度提供了决策依据，具有一定理论价值和 工程应用价值。

Abstract:

Large － scale wind power integration makes the grid and wind power dispatching extremely difficult，due to the uncertainty of natural wind and its direction． More accurate power generation prediction is an important issue under the maximum power tracking mode． Considering the day － ahead time scale is somehow long for wind power prediction which makes it difficult for real － time power dispatching，a rolling forecasting method of generation demand based on the intraday output of wind farm is proposed． Firstly，taking multi － level time scale as the forecasting basis，an ideal power generation demand forecasting model is built and the rolling forecasting algorithm using this model is given． Then，based on the actual operation data of wind farms，the computer simulation results show that the proposed method can improve the forecasting accuracy of power generation demand，which provides a decision criterion for the scientific dispatching of large － scale wind farms and may make some contributions to the research and engineering work on wind power dispatching．