基于改进ISODATA算法的变电站负荷特性聚类

doi:10.11930/j.issn.1004-9649.202505079

中国电力 ›› 2026, Vol. 59 ›› Issue (3): 125-133.DOI: 10.11930/j.issn.1004-9649.202505079

基于改进ISODATA算法的变电站负荷特性聚类

蒋达飞¹(), 艾洪克¹(), 孟巧¹(), 董彪¹(), 翁一帆²(), 张谦²()

1. 国网冀北电力有限公司唐山供电公司，河北唐山　063000
2. 输变电装备技术全国重点实验室（重庆大学），重庆　400044

收稿日期:2025-05-28 修回日期:2025-12-16 发布日期:2026-03-16 出版日期:2026-03-28
作者简介:蒋达飞（1987），男，硕士，高级工程师，从事新型电力系统规划研究，E-mail：jdf8020@163.com

艾洪克（1989），女，硕士，高级工程师，从事电气工程研究，E-mail：13060206654@163.com
孟巧（1987），女，高级工程师，从事电力大数据研究，E-mail：202311131110T@stu.cqu.edu.cn
董彪（1987），男，硕士，高级工程师，从事智慧配电网规划研究，E-mail：1137684372@qq.com
翁一帆（2001），男，硕士研究生，从事电动汽车与电网互动研究，E-mail：1084497796@qq.com
张谦（1980），女，通信作者，博士，教授，博士生导师，从事综合能源系统优化调度、电动汽车与电网互动研究，E-mail：zhangqian@cqu.edu.cn
基金资助:
国家自然科学基金资助项目（52277081）。

Clustering of substation load characteristics based on improved ISODATA algorithm

JIANG Dafei¹(), AI Hongke¹(), MENG Qiao¹(), DONG Biao¹(), WENG Yifan²(), ZHANG Qian²()

1. Tangshan Power Supply Company of State Grid Jibei Electric Power Co., Tangshan 063000, China
2. State Key Laboratory of Power Transmission Equipment Technology (Chongqing University), Chongqing 400044, China

Received:2025-05-28 Revised:2025-12-16 Online:2026-03-16 Published:2026-03-28
Supported by:
This work is supported by National Natural Science Foundation of China (No.52277081).

摘要/Abstract

摘要：

新型电力系统高压配电网面临规模化、多元化负荷接入的挑战。变电站负荷聚类是精准识别用户用电规律、优化电网资源配置的核心手段，其分析结果可直接支撑电网规划、需求侧管理及新能源消纳策略制定。因此亟须通过变电站负荷曲线聚类分析，精准解析差异化负荷模式及其动态演化规律，为智能配电网运行决策提供数据支撑。针对迭代式自组织数据分析算法（iterative self organizing data analysis techniques algorithm，ISODATA）存在收敛速度慢和难以捕捉数据高维特征的局限，尤其是负荷数据的动态特性捕捉不足的问题，分别通过优化初始聚类中心选取策略与引入核函数映射机制，以提升算法对变电站负荷曲线高维特征的解析能力。在完成缺失值填补与数据标准化预处理后，本算法首先基于最大距离准则优化初始聚类中心选取，最大化初始中心间异质性以提升聚类稳定性；其次，引入核函数映射机制，映射负荷曲线至高维空间聚类，实现高维特征的显式解耦与聚类分析。仿真结果表明，在特征提取能力方面，改进算法生成的主成分分析（principal component analysis，PCA）特征空间中变电站四季负荷特征呈现显著差异性，能更好地获取负荷高维特征；在算法性能方面，改进算法使执行时间减少32.8%，聚类评价指标戴维斯-布尔丁指数（davies-bouldin index，DBI）降低了29.1%，邓恩指数（dunn index，DI）提高了42.9%，验证了所提算法的有效性和优越性。

关键词: 变电站, 负荷聚类, 聚类效果指标

Abstract:

The new power system's high-voltage distribution grid faces challenges posed by the large-scale and diversified connection of loads. Substation load clustering is a core method for accurately identifying user electricity consumption patterns and optimizing grid resource allocation. Its analysis results can directly support grid planning, demand-side management, and the formulation of renewable energy integration strategies. Therefore, it is urgent to conduct substation load curve clustering analysis to precisely analyze differentiated load patterns and their dynamic evolution patterns, thereby providing data support for intelligent distribution grid operation decisions. Addressing the limitations of the iterative self-organizing data analysis techniques algorithm (ISODATA), such as slow convergence speed and difficulty in capturing high-dimensional data features—particularly the insufficient capture of load data's dynamic characteristics—this study enhances the algorithm's ability to analyze high-dimensional features of substation load curves by optimizing the initial cluster center selection strategy and introducing a kernel function mapping mechanism. After completing missing value filling and data standardization preprocessing, this algorithm first optimizes the selection of initial clustering centers based on the maximum distance criterion to maximize the heterogeneity between initial centers and improve clustering stability. Second, it introduces a kernel function mapping mechanism to map load curves to high-dimensional space clustering, achieving explicit decoupling and clustering analysis of high-dimensional features. Simulation results indicate that in terms of feature extraction capability, the principal component analysis (PCA) feature space generated by the improved algorithm exhibits significant differences in the seasonal load characteristics of substations, enabling better capture of high-dimensional load features; In terms of algorithm performance, the improved algorithm reduces execution time by 32.8%, lowers the Davies-Bouldin Index (DBI) by 29.1%, and increases the Dunn Index (DI) by 42.9%, validating the effectiveness and superiority of the proposed algorithm.

Key words: substation, load clustering, clustering effect indicator

蒋达飞, 艾洪克, 孟巧, 董彪, 翁一帆, 张谦. 基于改进ISODATA算法的变电站负荷特性聚类[J]. 中国电力, 2026, 59(3): 125-133.

JIANG Dafei, AI Hongke, MENG Qiao, DONG Biao, WENG Yifan, ZHANG Qian. Clustering of substation load characteristics based on improved ISODATA algorithm[J]. Electric Power, 2026, 59(3): 125-133.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.202505079

https://www.electricpower.com.cn/CN/Y2026/V59/I3/125

图/表 7

图 1 变电站负荷聚类应用流程

Fig.1 Application process of load clustering in substations

图 2 K-L-ISODATA算法聚类流程

Fig.2 K-L-ISODATA algorithm clustering process

表 1 3种核函数效果对比

Table 1 Comparison of the effects of the three kernel functions

聚类指标	线性核函数	多项式核函数	高斯核函数
DBI	0.58	0.52	0.39
DI	0.25	0.30	0.40

图 3 负荷聚类曲线

Fig.3 Four-season load clustering results

图 4 四季PCA特征空间

Fig.4 Four-season PCA feature space

表 2 跨季节特征维度对比

Table 2 Comparison of Cross-Seasonal Characterization Dimensions

季节	核心维度	关键特征	典型场景
春	PC1	双极对立	供暖-制冷切换期
夏	PC1/PC3	强聚集+尾部发散	空调负荷与极端气候
秋	PC2	核心-外围结构	农业/温变混合负荷
冬	PC3	核心稳定+尾部异常	极寒天气响应

表 3 不同聚类方法所需执行时间

Table 3 Execution time required for different clustering methods

算法	K-means	K-Medoids	ISODATA	K-L-ISODATA
时间/ms	13.5	44.8	19.5	13.1
DBI	0.60	0.52	0.55	0.39
DI	0.22	0.33	0.28	0.40

参考文献 30

1	王杰, 郑飞, 张鹏城, 等. 基于数据驱动的高比例新能源配电网规划模型[J]. 中国电力, 2025, 58 (3): 175- 182.
	WANG Jie, ZHENG Fei, ZHANG Pengcheng, et al. Model of high-proportion new energy distribution network planning based on data-driven approach[J]. Electric Power, 2025, 58 (3): 175- 182.
2	陈仕龙, 吴涛, 郭成, 等. 基于相关性分析的电网非同步监测数据场景谐波责任划分[J]. 中国电力, 2025, 58 (1): 15- 25.
	CHEN Shilong, WU Tao, GUO Cheng, et al. Harmonic responsibility division of grid asynchronous monitoring data scenarios based on correlation analysis[J]. Electric Power, 2025, 58 (1): 15- 25.
3	周洋, 黄德志, 李培栋, 等. 考虑平衡端点相位不对称及光伏接入的低压配电网三相潮流模型[J]. 中国电力, 2024, 57 (10): 190- 198.
	ZHOU Yang, HUANG Dezhi, LI Peidong, et al. A three-phase power flow model for low-voltage distribution networks considering balanced bus phase asymmetry and photovoltaic access[J]. Electric Power, 2024, 57 (10): 190- 198.
4	王伟杰, 黄海宇, 徐远途, 等. 电动汽车参与主动配电网电压调控的策略研究[J]. 广东电力, 2023, 36 (10): 93- 104.
	WANG Weijie, HUANG Haiyu, XU Yuantu, et al. Strategy research on electric vehicles participating in active distribution network voltage regulation[J]. Guangdong Electric Power, 2023, 36 (10): 93- 104.
5	宣羿, 樊立波, 孙智卿, 等. 考虑低碳交通的电动汽车充电站优化配置方法[J]. 浙江电力, 2024, 43 (6): 69- 79.
	XUAN Yi, FAN Libo, SUN Zhiqing, et al. An optimal allocation method for electric vehicle charging stations considering lowcarbon transportation[J]. Zhejiang Electric Power, 2024, 43 (6): 69- 79.
6	李畸勇, 宋子涵, 刘斌. 含垃圾能源化处理及电动汽车负载的区域综合能源系统协调优化[J]. 电测与仪表, 2024, 61 (4): 109- 115.
	LI Jiyong, SONG Zihan, LIU Bin. Coordinated optimization of regional comprehensive energy system with garbage energy treatment and electric vehicle load[J]. Electrical Measurement & Instrumentation, 2024, 61 (4): 109- 115.
7	栾捷, 李梁, 卢峰, 等. 电动汽车负荷聚合商参与调频辅助服务市场的方案研究[J]. 浙江电力, 2023, 42 (3): 18- 27.
	LUAN Jie, LI Liang, LU Feng, et al. Research on the scheme for electrical vehicle aggregators participating in the fre? quency modulation auxiliary service market[J]. Zhejiang Electric Power, 2023, 42 (3): 18- 27.
8	张怡, 郝思鹏. 电动汽车充电站变压器容量及储能优化配置[J]. 电测与仪表, 2023, 60 (7): 19- 25.
	ZHANG Yi, HAO Sipeng. Optimized configuration of transformer capacity and energy storage for electric vehicle charging stations[J]. Electrical Measurement & Instrumentation, 2023, 60 (7): 19- 25.
9	陈业夫, 王钦, 蔡新雷, 等. 计及规模化电动汽车调控潜力的含风电系统优化调度策略[J]. 浙江电力, 2024, 43 (4): 95- 104.
	CHEN Yefu, WANG Qin, CAI Xinlei, et al. An optimal scheduling strategy for wind-integrated systems considering the regula-Tory potential of large-scale electric vehicles[J]. Zhejiang Electric Power, 2024, 43 (4): 95- 104.
10	杨隽雯, 尚磊, 董旭柱, 等. 新型配电系统多层次综合评价指标体系与评价方法[J]. 广东电力, 2024, 37 (10): 13- 24.
	YANG Juan, JUNWEN, SHANG Lei, et al. Multi-level comprehensive evaluation index system and evaluation method of new distribution system[J]. Guangdong Electric Power, 2024, 37 (10): 13- 24.
11	杨彪, 颜伟, 莫静山. 考虑源荷功率随机性和相关性的主导节点选择与无功分区方法[J]. 电力系统自动化, 2021, 45 (11): 61- 67.
	YANG Biao, YAN Wei, MO Jingshan. Pilot-bus selection and network partitioning method considering randomness and correlation of source-load power[J]. Automation of Electric Power Systems, 2021, 45 (11): 61- 67.
12	LIN C, HAN G J, WANG T T, et al. Fast node clustering based on an improved birch algorithm for data collection towards software-defined underwater acoustic sensor networks[J]. IEEE Sensors Journal, 2021, 21 (22): 25480- 25488.
13	KHAN A A, AKHTER M M, MAHESHWARI R, et al. L-ASCRA: a linearithmic time approximate spectral clustering algorithm using topologically-preserved representatives[J]. IEEE Transactions on Knowledge and Data Engineering, 2024, 36 (12): 8643- 8654.
14	ZAKI ABDULHAMEED T, YOUSIF S A, SAMAWI V W, et al. SS-DBSCAN: semi-supervised density-based spatial clustering of applications with noise for meaningful clustering in diverse density data[J]. IEEE Access, 2024, 12, 131507- 131520.
15	谢七月, 刘广帅, 刘瑶, 等. 基于改进灰狼算法的定日镜场布局优化方法[J]. 太阳能学报, 2024, 45 (11): 394- 400.
	XIE Qiyue, LIU Guangshuai, LIU Yao, et al. Improved grey wolf optimization algorithm for heliostats field layout[J]. Acta Energiae Solaris Sinica, 2024, 45 (11): 394- 400.
16	鲁晓天, 唐金锐, 尹昕, 等. 基于多维时频特征的新型配电系统单相接地故障定位方法[J]. 高电压技术, 2025, 51 (2): 903- 914.
	LU Xiaotian, TANG Jinrui, YIN Xin, et al. Single-phase-to-ground fault section location in new-type distribution system based on multidimensional time-frequency distribution characteristics[J]. High Voltage Engineering, 2025, 51 (2): 903- 914.
17	朱昌敏. 基于spark模型的电力异常数据检测和短期负荷预测[D]. 南京: 南京邮电大学, 2018.
	ZHU Changmin. Identifying bad data and power load forecasting of power system based on spark[D]. Nanjing: Nanjing University of Posts and Telecommunications, 2018.
18	张世辉, 金同清, 张运杰, 等. 基于自组织聚类的多机协同编批方法[J]. 工程科学学报, 2024, 46 (7): 1269- 1278.
	ZHANG Shihui, JIN Tongqing, ZHANG Yunjie, et al. Multi-aircraft collaborative batching method based on self-organizing clustering[J]. Chinese Journal of Engineering, 2024, 46 (7): 1269- 1278.
19	姚明明, 张新, 杨培宏, 等. 基于改进风光场景聚类联合虚拟储能的源网荷储低碳优化调度[J]. 电力系统保护与控制, 2024, 52 (15): 115- 130.
	YAO Mingming, ZHANG Xin, YANG Peihong, et al. Low-carbon optimal scheduling of source-grid-load-storage based on improved wind-solar scene clustering combined with virtual energy storage[J]. Power System Protection and Control, 2024, 52 (15): 115- 130.
20	杨天鑫, 黄云辉, 何珍玉, 等. 基于多时间尺度调节的构网型储能电站定容选址优化配置[J]. 电力系统自动化, 2024, 48 (23): 54- 64.
	YANG Tianxin, HUANG Yunhui, HE Zhenyu, et al. Optimal configuration of siting and sizing for grid-forming battery energy storage station based on multi-timescale regulation[J]. Automation of Electric Power Systems, 2024, 48 (23): 54- 64.
21	张文伟, 王维庆, 王海云, 等. 考虑碳捕集技术的风光-富氧燃煤发电系统容量优化配置[J]. 电力系统自动化, 2023, 47 (13): 176- 189.
	ZHANG Wenwei, WANG Weiqing, WANG Haiyun, et al. Optimal capacity configuration for wind power-photovoltaic and oxygen-enriched coal-fired power generation system considering carbon capture technology[J]. Automation of Electric Power Systems, 2023, 47 (13): 176- 189.
22	张施令. 高压SF₆气体绝缘组合电器放电故障模式智能识别[J]. 高电压技术, 2020, 46 (2): 432- 440.
	ZHANG Shiling. Intelligent diagnosis method for discharge fault mode of high voltage SF₆ gas insulated switch-gear[J]. High Voltage Engineering, 2020, 46 (2): 432- 440.
23	YANG B, LIU S W, GAO H D. Energy e-commerce user portrait and multi-agent cooperative game price mechanism design[J]. Frontiers in Energy Research, 2024, 12, 1379612.
24	梁金正, 梁九龄, 熊振冬, 等. 基于全寿命周期成本的变电站设计方案比选方法研究[J]. 湖南电力, 2023, 43 (3): 115- 119.
	LIANG Jinzheng, LIANG Jiuling, XIONG Zhendong, et al. Research on comparison method of substation design scheme based on life cycle cost[J]. Hunan Electric Power, 2023, 43 (3): 115- 119.
25	王泽宇. 基于聚类集成与多重插补的Stacking光伏发电功率预测研究[D]. 南宁: 广西大学, 2023.
	WANG Zeyu. Prediction of photovoltaic power in stacking based on consensus clustering and multiple interpolation[D]. Nanning: Guangxi University, 2023.
26	毋雪雁. 基于遗传算法与粒子群算法混合优化加权K最近邻算法的病脑检测系统[D]. 南京: 南京师范大学, 2018.
	WU Xueyan. Smart pathological brain detection system by weighted K nearest neighbor based on hybrid genetic algorithm and particle swarm optimization algorithm[D]. Nanjing: Nanjing Normal University, 2018.
27	李卓昊, 石琼林, 王康丽, 等. 锂离子电池健康状态估计方法研究现状与展望[J]. 电力系统自动化, 2024, 48 (20): 109- 129.
	LI Zhuohao, SHI Qionglin, WANG Kangli, et al. Research status and prospects of state of health estimation methods for lithium-ion batteries[J]. Automation of Electric Power Systems, 2024, 48 (20): 109- 129.
28	周赣, 茅欢, 冯燕钧, 等. 基于多特征符号聚合近似和层次聚类的户变关系识别方法[J]. 电力系统自动化, 2024, 48 (3): 133- 141.
	ZHOU Gan, MAO Huan, FENG Yanjun, et al. Identification method for user-transformer relationship based on multi-feature symbolic aggregate approximation and hierarchical clustering[J]. Automation of Electric Power Systems, 2024, 48 (3): 133- 141.
29	于志诚, 穆士才, 梁晔, 等. 基于K-means聚类算法和BP神经网络的代理购电量预测模型研究[J]. 湖南电力, 2025, 45 (1): 68- 72.
	YU Zhicheng, MU Shicai, LIANG Ye, et al. Research on electricity prediction model of power purchasing agent based on K-means clustering algorithm and BP neural network[J]. Hunan Electric Power, 2025, 45 (1): 68- 72.
30	梁艳, 郭立, 张丹, 等. 考虑主客观响应能力的电动汽车聚合潜力评估[J]. 综合智慧能源, 2023, 45 (9): 1- 10.
	LIANG Yan, GUO Li, ZHANG Dan, et al. Evaluation on the convergence potential of electric vehicles considering their subjective and objective responsiveness[J]. Integrated Intelligent Energy, 2023, 45 (9): 1- 10.

[1]	陶军, 钟鸣, 张艺, 刘锋, 武玉珠, 夏振兴. 基于网络流算法和深度神经网络的智能变电站二次系统故障定位方法[J]. 中国电力, 2025, 58(9): 68-78.
[2]	叶远波, 王吉文, 孙振兴, 王志强, 赵勇呓. 分区协同下适应新一代智能变电站的通信链路故障定位方法[J]. 中国电力, 2025, 58(9): 79-87.
[3]	吕鹏飞, 裘愉涛, 金盛, 王志华. 面向智能变电站保护的FMM测试及改进SVM校验诊断技术[J]. 中国电力, 2025, 58(6): 76-82.
[4]	叶远波, 王吉文, 孙振兴, 周坤, 毛玉荣. 基于云端DDPG优化与协同的新一代变电站保护高效测试方法[J]. 中国电力, 2025, 58(6): 90-96.
[5]	叶远波, 王吉文, 汪伟, 毛玉荣, 王志华. 基于哈希和编辑距离算法的SCD双层向量化与变更校验技术[J]. 中国电力, 2024, 57(1): 255-262.
[6]	唐旭辰, 潮铸, 段秦尉, 苏炳洪, 陈卉灿. 基于分层测量数据的高压变电站概率负荷预测方法[J]. 中国电力, 2023, 56(8): 143-150.
[7]	皮志勇, 朱益, 廖玄, 李振兴, 方豪, 吴沛. 智能变电站多信息融合建模的通信链路故障定位方法[J]. 中国电力, 2023, 56(8): 207-215.
[8]	皮志勇, 朱益, 廖玄, 李振兴, 方豪, 吴沛. 基于深度学习的智能变电站通信链路故障定位方法[J]. 中国电力, 2023, 56(7): 136-145.
[9]	张杰, 王恒凤, 刘生春, 王华彪, 王沧海, 冉垚, 王献敏, 马勇飞, 颜伟. 基于内点法和邻域搜索解耦动态规划法的区域电网动态无功优化方法[J]. 中国电力, 2023, 56(2): 59-67.
[10]	钟鸣, 陶军, 刘洵宇, 杨逸, 杨炳元. 智能变电站光纤虚实回路映射及故障诊断技术[J]. 中国电力, 2023, 56(10): 171-178.
[11]	严亚兵, 褚旭, 肖豪龙, 梁文武, 李辉, 夏振兴. 基于改进支持向量机的数字化变电站安全措施生成技术[J]. 中国电力, 2023, 56(10): 194-201.
[12]	汤锦慧, 伍发元, 支妍力, 毛梦婷, 代小敏. 基于深度强化学习的户内变电站通风降噪优化设计[J]. 中国电力, 2023, 56(1): 96-105,118.
[13]	李肖博, 于杨, 姚浩, 习伟, 蔡田田. 新一代智能变电站采控装置[J]. 中国电力, 2022, 55(4): 85-92.
[14]	范长俊, 孟飞, 郭添亨, 杜彦锟, 崔晓琳, 张程. 预制舱变电站多状态监测技术探讨[J]. 中国电力, 2022, 55(12): 98-104.
[15]	闫帅, 李朋宇, 王高洁, 李强, 吴凡, 罗林根. 基于特高频无线智能传感阵列的敞开式变电站放电定位方法[J]. 中国电力, 2021, 54(2): 52-57,65.

基于改进ISODATA算法的变电站负荷特性聚类

Clustering of substation load characteristics based on improved ISODATA algorithm

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献 30

相关文章 15

编辑推荐

Metrics

模态框（Modal）标题

基于改进ISODATA算法的变电站负荷特性聚类

Clustering of substation load characteristics based on improved ISODATA algorithm

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献 30

相关文章 15

编辑推荐

Metrics