基于气象大数据的城市电力负荷预测

Prediction of urban electricity load based on meteorological big data

为克服利用气象因素对用电量预测任务中必须先观测到气象条件再进行预测的困境,提升用电量预测准确性,提出一种基于时序卷积网络与循环神经网络的用电量预测方法。使用时序卷积网络基于历史气象数据对未来气象条件进行预测,结合历史用电量数据对未来用电量数据进行预测。算法在预测当前用电量时只依赖于过去的特征,因此无需先观测到当前气象特征。在真实的气象与用电量数据集上的实验结果表明,在仅使用气象因素这一外部变量时,算法对用电量的预测准确性超出了传统方法,有较高的实用性。

In order to overcome the difficulty in electricity consumption forecasting based on meteorological data where observed present meteorological characters are necessary,and improve the accuracy of forecasting,an electricity consumption prediction method based on temporal convolution network(TCN) and recurrent neural network(RNN) was proposed in this paper.Firstly,future meteorological condition is predicted based on historical meteorological data with TCN,and then,the future electricity consumption is predicted based on historical electricity consumption data.During this process,the proposed method depends only on historical data;therefore,it is not necessary to observe the current meteorological features.Experimental results on real meteorological dataset and electricity consumption dataset demonstrates that the proposed method shows an advantage in electricity consumption prediction accuracy when only the meteorological factor is used as an external variable,and of high applicability.