一种基于时段特征的匹配算法在智能电表用电预测中的应用研究

用电需求预测问题对智能电网的稳定运行和用户体验的提升有着非常重要的影响。现有的工作大多是针对长、中期用电需求进行预测,而对短期内细粒度的用电需求的预测效果不佳。针对这一问题,基于特征匹配的思想,提出一种细粒度的用电需求预测算法。基于智能电表中记录的居民用电数据,提取每个用户的用电特征,并在智能电表中记录的实时用电数据上进行特征匹配,利用匹配到的特征进行用电量的预测。实验证明,该方法在细粒度的用电数据上取得了良好的性能。

APPLICATION OF MATCHING ALGORITHM BASED ON TEMPORAL FEATURE IN POWER CONSUMPTION FORECAST OF SMART METER

Prediction of power demand is very important for the stability of the smart grid and the user s experience.Most power demand forecasting algorithms are aimed at the long and medium term power demand forecast.While the prediction effect of fine-grained electricity demand in the short term is not good.In order to solve this problem,this paper proposes a fine-grained electricity demand forecasting algorithm based on the idea of feature matching.It extracted the power consumption characteristics of each user based on the residential power data recorded on smart meters,and matched the characteristics on the real-time power consumption data.And matched characteristics were used to predict the power consumption.Experimental results show that the proposed method achieves the good performance on fine-grained power data.