风电场短期风电功率的神经网络方法预测研究

Study of Wind Power Short-Term Prediction of Wind Farm Based on Neural Network

对风力发电进行有效的预测,则可降低电网调度的难度。根据从风电场获得的相关风速、温度、风向、风电功率等数据,建立基于BP神经网络的短期风电功率预测模型,预测提前1,2,4,24h的风电功率。对所得预测结果进行比较,从而判断各种短期预测模型的优劣。从对比的结果可知,神经网络模型预测不超过24h的风电功率时具有一定的可靠性。

Effective forecasts on wind power can reduce the difficulty of the power grid dispatching. According to some history data from a wind farm such as wind speed, temperature, wind direction, wind power and so on, a short-term forecast model based on BP neural network was set up in order to forecast wind power ahead 1 hour, 2 hours, 4 hours and 24 hours. The results were compared with each other in order to determine whether the short-term prediction model is good or not. From the prediction results, it shows that the neural network model has certain reliability in predicting no more than 24 hours＇ wind power.