基于充电桩数据挖掘的EV有序充电策略

Orderly charging strategy of EV based on data mining of charing piles

充电桩的快速发展为EV用户提供便利的同时,也给小区电网的安全稳定运行带来影响。针对居民充电桩使用过程中的痛点问题,提出一种基于数据挖掘技术的充电桩有序充电策略。首先利用基于遗传算法优化的Elman神经网络对小区负荷进行预测,便于提前做好小区电力调度;然后利用K-means算法对EV用户充电习性进行分类,对不同EV用户进行针对性负荷调控;最后基于对EV用户充电习性的分析,提出改进的有序充电策略。所提策略充分利用了小区变压器剩余容量,减少了电网投资,在保证小区电网稳定运行的同时,极大地提高了EV用户的充电体验。

The rapid development of charging piles not only provides convenience for EV users,but also affects the safe and stable operation of the community power grid.Aiming at the pain points in the use of residential charging piles,an orderly charging strategy based on data mining is proposed.Firstly,the Elman neural network optimized based on genetic algorithm is used to predict the cell load,which is convenient to do the power dispatching in advance.Then,K-means algorithm is used to classify the charging habits of EV users.Targeted load regulation for different EV users.Finally,based on the above analysis of EV users’charging habits,an improved orderly charging strategy is proposed.The proposed strategy makes full use of the residual capacity of the transformer in the community,reduces the investment of the grid,and greatly improves the charging experience of EV users while ensuring the stable operation of the grid in the community.