自下而上的群体居民日负荷预测  

Bottom-up daily load profile forecasting of group households

作　　者：朱云毓 高丙团[1] 陈宁 朱振宇[1] 刘晓峰[1] 秦艳辉 Zhu Yunyu;Gao Bingtuan;Chen Ning;Zhu Zhenyu;Liu Xiaofeng;Qin Yanhui(School of Electrical Engineering,Southeast University,Nanjing 210096,China;State Key Laboratory of Operation and Control of Renewable Energy and Storage Systems,China Electric Power Research Institute,Nanjing 210003,China;Electric Power Research Institute of State Grid Xinjiang Electric Power Co.,Ltd.,Urumqi 830011,China)

机构地区：[1]东南大学电气工程学院,南京210096 [2]中国电力科学研究院有限公司新能源与储能运行控制国家重点实验室,南京210003 [3]国网新疆电力有限公司电力科学研究院,乌鲁木齐830011

出　　处：《东南大学学报（自然科学版）》2020年第1期46-55,共10页Journal of Southeast University：Natural Science Edition

基　　金：国家自然科学基金资助项目(51741709);国家电网公司总部科技资助项目(5230HQ19000J)

摘　　要：为了提高居民日负荷预测精度,基于自下而上的建模思想,提出一种模块化的群体居民用户日负荷预测方法.考虑外界因素和用户自身用电行为对负荷的影响,构建相似日提取模块、聚类分析模块和用户用电行为分析模块以实现单户家庭负荷预测.在此基础上,利用蒙特卡洛抽样方法对家用电器组合、家用电器功率和用户用电时间点这3个随机变量进行抽样模拟,构建用户负荷预测模块,实现群体居民日负荷预测.算例仿真结果表明:采用所提方法的居民日负荷预测平均误差为1.3%,最大误差为5.6%.相较于基于灰色模型预测的平均误差2.7%、最大误差7.5%,和基于神经网络模型预测的平均误差2.3%、最大误差6.9%,所提方法显著提高了群体居民日负荷预测的精度.To improve the forecasting accuracy of the daily load of group households, a modular daily load forecasting method for group residents was presented based on the bottom-up modeling idea. Considering the influences of external factors and user’s own behavior on the load, a similar day extraction module, a cluster analysis module and a user electricity behavior analysis module were established to realize daily load forecasting for a single-family household. Consequently, by using Monte Carlo sampling method to simulate three random variables: household appliance combination, household appliance power and user power consumption time point, the user load prediction module was established to realize the daily load forecasting for group households. Simulation results by a case study indicate that the average error and the maximum error of daily load forecasting for group households are 1.3% and 5.6%, respectively. Compared with the forecasting method based on grey model with the average error 2.7% and the maximum error 2.5%, and the forecasting method based on neural network model with the average error 2.3% and the maximum error 6.9%, the proposed method can improve forecasting accuracy of the daily load of group households.

关 键 词：自下而上 用电行为 日负荷预测 模块化 蒙特卡洛模拟 

分 类 号：TM714[电气工程—电力系统及自动化]