摘要
为进一步发挥用户用电细粒度负荷数据的内在价值并提高用户聚合体的电力负荷预测精度,提出一种考虑用户用电行为聚类的电力负荷预测方法。方法先计算用户用电数据的分位数自协方差作为用户用电行为特征;使用层次聚类算法对用户用电行为特征聚类;在用户聚类的基础上建立分位数回归森林电力负荷预测模型,通过对各类用户负荷预测结果求和,得到用户聚合体负荷预测结果。采用伦敦用户用电数据集进行仿真,结果表明提出方法能够提高用户聚合体负荷预测的精度。
In order to further utilize the intrinsic value of the fine-grained load data of users’ electricity consumption and improve the accuracy of the power load forecasting of the user aggregates, a power load forecasting method considering the clustering of the electricity consumption behavior of the users is proposed. This method first calculates the quantile autocovariance of the user’s electricity consumption data as the user’s electricity consumption behavior characteristics;then uses a hierarchical clustering algorithm to cluster the user’s electricity consumption behavior characteristics;finally, a quantile regression forest power load forecasting model is established on the basis of user clustering, and the user aggregate load forecasting results are obtained by summing the load forecasting results of various users. Using the London user electricity data set for experimental simulation, the results show that the proposed method can improve the accuracy of user aggregate load forecasting.
作者
黄薇
温蜜
张照贝
HUANG Wei;WEN Mi;ZHANG Zhao-bei(State Grid Shanghai Electric Power Company,Shanghai 200122,China;College of Computer Science and Technology,Shanghai University of Electric Power,Shanghai 201303,China)
出处
《计算机仿真》
北大核心
2022年第12期148-153,共6页
Computer Simulation
基金
国家自然科学基金(61872230)
国网上海市电力公司科技项目(5209001900PL)。
关键词
电力负荷预测
分位数自协方差
用电行为聚类
分位数回归森林
Power load forecast
Quantile autocovariance
Electricity use behavior clustering
Quantile regression forest
