Page 73 - 电力与能源2024年第三期

第 45 卷 第 3 期 电力与能源
2024 年 6 月 347

DOI：10.11973/dlyny202403014

计及风光资源不确定性的独立型微电网

容量优化配置

张晶晶，陈 俊，曹 辉，茅伟杰，沈珉峰，姜 雯

（国网上海市电力公司松江供电公司，上海 201699）

摘 要：围绕风光柴储独立型微网的容量优化配置问题展开研究，为考虑风光资源不确定性给微网带来的运
行风险，提出了一种基于条件风险价值（CVaR）的微网容量随机优化配置模型。首先，利用拉丁超立方抽样方
法模拟出大量风光出力场景，并用 K-medoids 聚类算法进行场景消减，得到特征明显、出现概率大的风光场景
数据 ；其次，以微网系统供电可靠性为约束条件，综合考虑微网的经济性和可再生能源利用率指标，结合
CVaR 建立以年均综合费用最小为目标的随机优化模型；最后，采用二进制粒子群优化算法对这一整数优化
模型进行求解，利用电源损失风险指标和负荷风险损失指标分别对确定性优化模型和随机优化模型得到的配
置结果进行评价。仿真算例结果表明，相比确定性优化模型，所提出的随机优化模型得到的配置方案具有更
强的鲁棒性，能够适应未来多种可能的运行场景。
关键词：独立型微电网；风光资源不确定性；拉丁超立方抽样；条件风险价值；随机优化模型
作者简介：张晶晶（1991—），女，工程师，从事电网变电运维工作。
中图分类号：TM61；TM732 文献标志码：A 文章编号：2095-1256（2024）03-0347-08
Capacity Optimization Configuration of Stand-Alone Microgrid Considering Uncertainty

of Wind and Solar Resources

ZHANG Jingjing，CHEN Jun，CAO Hui，MAO Weijie，SHEN Minfeng，JIANG Wen

（State Grid Songjiang Power Supply Company，SMEPC，Shanghai 201699，China）

Abstract：This study focuses on the capacity optimization configuration of a wind-solar-diesel-battery stand-
alone microgrid. To address the operational risks brought by the uncertainty of wind and solar resources, we pro⁃
pose a stochastic optimization configuration model for microgrid capacity based on Conditional Value at Risk
(CVaR). Firstly, a large number of wind and solar output scenarios are simulated using Latin hypercube sam ⁃
pling, and then scenario reduction is conducted using the K-medoids clustering algorithm to obtain data on wind
and solar scenarios with distinct features and high occurrence probabilities. Secondly, considering the reliability of
microgrid system supply as a constraint, we comprehensively consider the economic indicators and renewable en⁃
ergy utilization rates of the microgrid, combined with CVaR, to establish a stochastic optimization model with the
objective of minimizing the average annual comprehensive cost. Finally, a binary particle swarm optimization algo⁃
rithm is employed to solve this integer optimization model, and the deterministic and stochastic optimization re⁃
sults of the configuration are evaluated using power loss risk indicators and load loss risk indicators, respectively.
Simulation results demonstrate that the proposed stochastic optimization model for configuration yields more ro⁃
bust solutions, capable of adapting to various potential operational scenarios in the future.

Key words：stand-alone microgrid，uncertainty of wind and solar resources，Latin hypercube sampling，condi⁃

tional value at risk，stochastic optimization model
独立型微电网（以下简称微网）可以看作是小 在一起，具有供电灵活、可近距离输电等优点，已
型的电力系统，具备完整的发电和配电功能，可将 成为解决偏远地区（如孤岛、边防哨所等）供电问
各类分布式电源、储能装置以及负荷有效地整合 题的关键技术 ［1-2］ 。容量优化配置是设计微网时