In view of the large number of user types and different user power consumption patterns in power industry,a re?al-time load forecasting method based on user clustering strategy for smart meters is proposed. This method takes first order moment of the user load as features,then the clustering algorithm is used to different power mode the user classification,and then regression algorithm is used to predict the load on the user classification. By predicting the load superposition of all kinds of users,real-time load forecasting total power is gotten. The experimental results show that the hybrid algorithm has a high accuracy and has a certain guiding significance for power dispatching.%针对电力行业用户类型种类多,不同用户用电模式差异大的问题,提出基于用户分群策略的智能电表计量电力实时负荷预测方法.该方法首先以用户整点时刻的负荷作为特征,使用聚类算法将不同用电模式的用户进行分类,然后使用回归算法对分类用户的负荷进行预测,通过叠加各类用户的预测负荷,获得电力实时总预测负荷.实验结果表明,该混合算法准确性较高,对电力调度具有一定的指导意义.
展开▼
