Page 83 - 《中国电力》2026年第4期

第 59 卷 第 4 期 Vol. 59, No. 4
2026 年 4 月 ELECTRIC POWER Apr. 2026

引用格式：李欣, 宋金金. 计及源荷双重不确定性的高速路域虚拟电厂运行优化策略[J]. 中国电力, 2026, 59(4): 79−93.
Citation: LI Xin, SONG Jinjin. Operation optimization strategy for highway-domain virtual power plants considering dual uncertainties of source and
loads[J]. Electric Power, 2026, 59(4): 79−93.

计及源荷双重不确定性的高速路域

虚拟电厂运行优化策略

李欣，宋金金

(兰州交通大学 新能源与动力工程学院，甘肃 兰州 730070)

Operation optimization strategy for highway-domain virtual power plants

considering dual uncertainties of source and loads
LI Xin, SONG Jinjin

(School of New Energy and Power Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China)

Abstract: The rapid expansion of China's highway network has This work is supported by National Natural Science Foundation
exacerbated the problems of transportation energy consumption of China (No.52567023); Gansu Provincial Department of Trans-
and carbon emissions. It is therefore imperative to achieve portation's Unveiling Leadership Project (No.JT-JJ-2023-008).
efficient local consumption of distributed energy resources in Keywords: transportation-energy integration; source-load
highway domains through virtual power plants (VPPs). To uncertainty; highway-domain VPP; operation optimization;
address the "source-load mismatch" problem caused by the policy optimization algorithm
volatility of distributed energy output and the randomness of
electric vehicle (EV) charging demand, it is of great 摘 要：中国高速公路路网的快速扩张加剧了交通能源
significance to develop an optimal operational strategy for 消耗与碳排放问题，亟需通过虚拟电厂（virtual power plant，
highway-domain VPPs that account for source-load uncertainty, VPP）实现高速路域分布式能源的高效就地消纳，开展计
with the goal of reducing the VPPs' operating costs. To tackle 及源荷不确定性的高速路域 VPP 运行优化策略对降低 VPP
the uncertainty in distributed energy output and EV charging 运行成本具有重要意义。针对分布式能源出力与电动汽
demand, this paper firstly employs the Markov Chain Monte 车充电需求不确定性的问题，基于马尔科夫链蒙特卡洛
Carlo (MCMC) method to generate a multi-timescale scenario 法采样生成多时间尺度场景集，提出马尔科夫链蒙特卡
set. It then proposes a hybrid framework combining MCMC 洛法与基于概率距离削减的方法结合的场景生成-削减混
and the probability distance-based reduction method for 合框架进行场景生成与削减。以高速路域 VPP 运行成本
scenario generation and reduction. Subsequently, an 最小为目标建立高速路域 VPP 运行优化模型，采用基于
optimization model for highway-domain VPPs is established 约束策略优化的高速路域 VPP 智能调度算法进行求解优
with the objective of minimizing its operating costs, and a smart 化。算例结果表明，所提优化方案使高速路域 VPP 运行
scheduling algorithm based on constrained proximal policy 成本降低，验证了所提方案具有良好的经济性。
optimization (C-PPO) is proposed to solve this optimization 关 键 词 ： 交 能 融 合 ； 源 荷 不 确 定 性 ； 高 速 路 域 虚 拟 电
model. Case study results show that the proposed optimization 厂；运行优化；策略优化算法
scheme reduces the operating cost of the highway-domain DOI：10.11930/j.issn.1004-9649.202509067

VPPs, validating the favorable economic performance of the
proposed solution. 0 引言

收稿日期：2025−09−29； 修回日期：2025−12−30。 《“十四五”可再生能源发展规划》提出将
基金项目：国家自然科学基金资助项目（52567023）；甘 交通基础设施与能源统筹布局规划建设，推进交
肃省交通厅揭榜挂帅项目（JT-JJ-2023-008）。 通网与能源网能源融合，提高风电、光伏发电主

79