基于K-means聚类的源网荷储一体化系统容量需求研究  

Research on System Capacity Requirements of Integrated Source Grid Load Storage System Based on K-means Clustering

作　　者：张家杭 朱朱 郑沛沛 高丽萍 ZHANG Jiahang;ZHU Zhu;ZHENG Peipei;GAO Liping(PowerChina Fujian Electric Power Engineering Co.,Ltd.,Fuzhou 350003,China)

机构地区：[1]中国电建集团福建省电力勘测设计院有限公司,福建福州350003

出　　处：《电工技术》2025年第6期25-27,共3页Electric Engineering

摘　　要：源网荷储一体化系统充分挖掘系统灵活性调节能力,实现源网荷储各环节间协调互动,但是一体化系统中由于新能源出力的波动性需要系统提供备用容量,因此首先基于K-means算法对风光出力数据进行聚类分析,选取典型出力场景,采用轮廓系数来评价聚类效果,确定最佳聚类数;然后基于聚类结果建立源网荷储一体化系统容量优化模型,采用蒙特卡洛遗传算法对优化模型进行求解,并通过实例验证该方法的可行性和有效性。The source grid load storage integration system fully explores the flexibility and regulation ability of the system,achieving coordinated interaction between various links of source grid,load and storage.However,in integrated system,the fluctuation of new energy output requires the system to provide reserve capacity.Firstly,the paper conducts clustering on wind and solar power output based on the K-means algorithm,selects typical scenarios,and the silhouette coefficient is used to evaluate the clustering effect and determine the optimal number of clusters.Based on the scenario clustering results,a source grid load storage integration system reserve capacity optimization model is established.A Monte Carlo genetic algorithm is used to solve the established model,and the feasibility and effectiveness of the method are verified through example.

关 键 词：源网荷储一体化 K-MEANS 聚类 备用容量 蒙特卡洛遗传算法 

分 类 号：TM715[电气工程—电力系统及自动化]