基于人工神经网络的风电功率预测

风电场输出功率预测对接入大量风电的电力系统运行有重要意义。对风速和风电场输出功率预测的方法进行了分类。根据风电场输出功率的影响因素，建立了风电功率预测的神经网络模型。分析了实测功率数据、不同高度的大气数据对预测结果的影响。建立了基于神经网络的误差带预测模型，实现了误差带预测。研究结果表明，神经网络的结构和输入样本对预测结果有一定的影响；实测功率数据作为输入可以提高提前量为30 min的预测精度，而对提前量为1 h的预测精度会降低；把不同高度的数据都作为神经网络的输入比只采用轮毂高度数据的预测精度高；设计的神经网络能够对误差带进行预测。

Wind Power Prediction Based on Artificial Neural Network

Wind power prediction is important to the operation of power system with comparatively large mount of wind power. The wind power prediction methods were classified into several kinds. An artificial neural network (ANN) model for wind power prediction was constructed according to the wind power influence factors. Then the impacts of real time measured power and the atmospheric data at different heights on prediction results were analyzed. Besides, another ANN model for error band prediction was also built. The results indicate that the ANN structure and the training sample have some impact on the prediction precision. The real time measured power as input will improve the precision of 30 min ahead prediction, however will decrease the precision of 1h ahead prediction. The results which using the atmospheric data at all different heights as input have a higher accuracy when compared with the results using hub height data only. The designed ANN can forecast the error band.