Page 144 - 《中国电力》2026年第4期
P. 144
第 59 卷 第 4 期 Vol. 59, No. 4
2026 年 4 月 ELECTRIC POWER Apr. 2026
引用格式:曹望璋, 陆泳昊, 靳丰源, 等. 基于分布式梯度投影的居民区电动汽车均衡充电策略[J]. 中国电力, 2026, 59(4): 140−149.
Citation: CAO Wangzhang, LU Yonghao, JIN Fengyuan, et al. Equilibrium charging strategy for electric vehicles in residential areas based on
distributed gradient projection[J]. Electric Power, 2026, 59(4): 140−149.
基于分布式梯度投影的居民区电动汽车均衡充电策略
曹望璋 ,陆泳昊 ,靳丰源 ,阳浩 ,金鑫 ,王宗义 ,赵勃扬 2
1
2
2
1
1
3
(1. 南方电网科学研究院有限责任公司,广东 广州 510640;2. 西安交通大学 电气工程学院,陕西 西安 710049;
3. 南方电网有限责任公司,广东 广州 510623)
Equilibrium charging strategy for electric vehicles in residential areas
based on distributed gradient projection
2
1
2
3
CAO Wangzhang , LU Yonghao , JIN Fengyuan , YANG Hao ,
1
1
JIN Xin , WANG Zongyi , ZHAO Boyang 2
(1. China Southern Power Grid Electric Power Research Institute Co., Ltd., Guangzhou 510640, China; 2. School of Electrical Engineering,
Xi'an Jiaotong University, Xi'an 710049, China; 3. China Southern Power Grid Co., Ltd., Guangzhou 510623, China)
Abstract: Residential charging load management is a key of China (Theory on User Equilibrium Analysis and Regulation
means to mitigate the adverse impacts of large-scale electric in Large-Scale Demand Response for Further Electrification
vehicle (EV) integration on the power grid. However, Revolution, No.52177113), Science and Technology Project of
centralized management compromises users' privacy, and the China Southern Power Grid Co., Ltd. (No.ZBKJXM20232273).
autonomous response of users under time-of-use electricity Keywords: electric vehicles; demand response; aggregative
prices tends to induce new load peaks. Firstly, a dynamic game; Nash equilibrium; dynamic price
electricity pricing mechanism is introduced in residential
communities, and an aggregative game model is constructed to 摘 要:居民充电负荷管理是抑制大规模电动汽车对电
guide users to spontaneously adjust their charging behaviors 网 不 利 影 响 的 重 要 举 措 , 但 集 中 式 管 理 对 用 户 隐 私 不
and avoid peak-valley reversal. Then, a distributed gradient 利 , 分 时 电 价 下 用 户 的 自 主 响 应 易 引 发 新 的 负 荷 高 峰 。
projection algorithm is proposed to rapidly achieve the Nash 首先,在居民小区引入动态电价机制,构建聚合博弈模
equilibrium of the aggregate game, which fully protects the 型 , 引 导 用 户 自 发 调 整 充 电 行 为 , 规 避 峰 谷 倒 置 。 然
privacy of vehicle owners. Finally, simulation analysis is 后,提出分布式梯度投影算法实现聚合博弈纳什均衡的
carried out based on the electricity consumption data of a 快速求解,充分保护车主隐私。最后,基于中国南方某
residential community in southern China. The results show that 居民小区用电数据开展仿真分析。结果表明:所提方法
the proposed method can effectively smooth the peak load 可有效平抑电动汽车充电引发的尖峰负荷,效果随电动
caused by EV charging, and the effect is enhanced with the 汽车数量增加而提升;相比于分时电价机制,可有效避
increase in the number of EVs. Compared with the time-of-use 免新的负荷高峰产生,满足电动汽车不断增长下的充电
pricing mechanism, it can effectively avoid the emergence of 负荷管理要求。
new load peaks and meet the requirements of charging load 关 键 词 : 电 动 汽 车 ; 需 求 响 应 ; 聚 合 博 弈 ; 纳 什 均 衡 ;
management under the growing penetration of electric vehicles. 动态电价
This work is supported by National Natural Science Foundation DOI:10.11930/j.issn.1004-9649.202512027
收稿日期:2025−12−15; 修回日期:2026−03−17。 0 引言
基金项目:国家自然科学基金资助项目(面向再电气化的
大规模可控负荷均衡分析与调控理论研究,52177113); 随着电动汽车(electric vehicle,EV)数量的
中国南方电网有限责任公司科技项目(ZBKJXM20232273)。 增加,公共充电站、办公园区和居民区等场所的
140

