基于生成对抗网络的中压配网供电单元自动成图方法

doi:10.16183/j.cnki.jsjtu.2023.626

上海交通大学学报 ›› 2025, Vol. 59 ›› Issue (10): 1431-1441.doi: 10.16183/j.cnki.jsjtu.2023.626

• 新型电力系统与综合能源 • 上一篇下一篇

基于生成对抗网络的中压配网供电单元自动成图方法

陈锦铭¹^,², 蒋玮¹(), 王志伟¹, 朱振汉³, 陈烨², 赵彦超³

¹ 东南大学电气工程学院, 南京 210096
² 国网江苏省电力有限公司电力科学研究院, 南京 211103
³ 南京航空航天大学计算机科学与技术学院, 南京 211106

收稿日期:2023-12-15 修回日期:2024-03-11 接受日期:2024-04-17 出版日期:2025-10-28 发布日期:2025-10-24
通讯作者: 蒋玮,教授,博士生导师;E-mail:jiangwei@seu.edu.cn.
作者简介:陈锦铭(1985—),博士生,研究员级高级工程师,从事配电网大数据及人工智能应用研究.
基金资助:
国家重点研发计划资助项目(2022YFB2404200)

Automatic Mapping Method for Power Supply Units in Medium-Voltage Distribution Networks Based on Generative Adversarial Network

CHEN Jinming¹^,², JIANG Wei¹(), WANG Zhiwei¹, ZHU Zhenhan³, CHEN Ye², ZHAO Yanchao³

¹ School of Electrical Engineering, Southeast University, Nanjing 210096, China
² Electric Power Research Institute, State Grid Jiangsu Electric Power Co., Ltd., Nanjing 211103, China
³ College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China

Received:2023-12-15 Revised:2024-03-11 Accepted:2024-04-17 Online:2025-10-28 Published:2025-10-24

摘要/Abstract

摘要：

随着配电网“单元制”规划方法的逐步推广,区域配电网被划分为多个相对独立的供电单元,然而,由于供电单元内部依然存在多条互相联络的线路,其结构的复杂性导致供电单元成图较为困难.现有基于规则或力导向等启发式的自动化成图方法效率较低,且依赖人工干预,难以适应复杂多变的配电网应用场景.为此,提出一种基于均方差条件生成对抗网络的中压配网供电单元自动成图方法.该方法通过布局生成器和遗传变异算法,细粒度地生成并优化配电网节点布局与连线,实现了多种节点规模下的自动成图.同时,设计了面向节点布局生成器的评估函数,以节点聚散程度、连线交叉与拐点数量等拓扑可视化效果作为关键评价指标,用于在循环迭代中优化布局生成器,从而提升成图效果.实验结果表明,所提方法在成图效果上优于其他启发式布局方法.

关键词: 中压配电网, 供电单元, 自动成图, 生成对抗网络

Abstract:

With the gradual promotion of the “unit based” planning method for distribution networks, regional distribution networks have been divided into several relatively independent power supply units. However, there are multiple interconnecting lines within the power supply units, and the complexity of the structure makes the mapping of power supply units more difficult. The heuristic automated mapping methods based on rules and force orientation are inefficient and relies on manual intervention, which cannot adapt to the complex and changing distribution network application scenarios. Therefore, this paper proposes an automatic mapping method for medium-voltage distribution network power supply units based on the mean square error condition generation adversarial network. The layout generator and genetic mutation algorithm in this method can generate and optimize the layout and connection of distribution network nodes at a fine-grained level, achieving automatic mapping at various node scales. Then, it designs an evaluation function for node layout generators, which takes topology visualization performances such as node clustering degree, line crossing, and inflection points as key evaluation indicators. This function can be used to iteratively optimize the layout generator and thereby improve the mapping effect. The experimental results show that the method proposed outperforms other heuristic layout methods in terms of mapping effect.

Key words: medium-voltage distribution network, power supply unit, automatic mapping, generative adversarial network (GAN)

中图分类号:

TM715
TM732

陈锦铭, 蒋玮, 王志伟, 朱振汉, 陈烨, 赵彦超. 基于生成对抗网络的中压配网供电单元自动成图方法[J]. 上海交通大学学报, 2025, 59(10): 1431-1441.

CHEN Jinming, JIANG Wei, WANG Zhiwei, ZHU Zhenhan, CHEN Ye, ZHAO Yanchao. Automatic Mapping Method for Power Supply Units in Medium-Voltage Distribution Networks Based on Generative Adversarial Network[J]. Journal of Shanghai Jiao Tong University, 2025, 59(10): 1431-1441.

图/表 12

图1

图2

图3

图4

图5

图6

图7

图8

图9

表1

表2

图10

参考文献 20

[1]	陈璐, 陈连杰, 欧阳文, 等. 基于环形结构的配电网联络图布局算法[J]. 电力系统自动化, 2016, 40(24): 151-156.
	CHEN Lu, CHEN Lianjie, OUYANG Wen, et al. Ring structure based layout algorithm for distribution network contact-graph[J]. Automation of Electric Power Systems, 2016, 40(24): 151-156.
[2]	徐遐龄, 田国辉, 刘涛, 等. 基于神经网络的电网潮流图自动拓扑布线方法[J]. 电力系统自动化, 2022, 46(6): 102-108.
	XU Xialing, TIAN Guohui, LIU Tao, et al. Automatic topology wiring method of power flow diagram based on neural network[J]. Automation of Electric Power Systems, 2022, 46(6): 102-108.
[3]	国家电网有限公司. 配电网规划设计技术导则: Q/GDW 10738—2020[S]. 北京: 中国电力出版社, 2020.
	State Grid Corporation of China. The guide for planning and design of distribution network: Q/GDW 10738—2020[S]. Beijing: China Electric Power Press, 2020.
[4]	唐广瑜, 金鑫琨. 改进引力-斥力算法在配电网单线图自动生成中的应用研究[J]. 电力系统保护与控制, 2022, 50(18): 150-156.
	TANG Guangyu, JIN Xinkun. Research on application of improved gravity-repulsion algorithm in automatic generation of distribution network one-line diagram[J]. Power System Protection and Control, 2022, 50(18): 150-156.
[5]	岳东. 人工智能技术在新型电力系统中的分析、诊断和控制研究[J]. 电力工程技术, 2021(6): 1.
	YUE Dong. Research on analysis, diagnosis and control of artificial intelligence technology in new power system[J]. Electric Power Engineering Technology, 2021(6): 1.
[6]	李珂, 邰能灵, 张沈习. 基于改进粒子群算法的配电网综合运行优化[J]. 上海交通大学学报, 2017, 51(8): 897-902. doi: 10.16183/j.cnki.jsjtu.2017.08.001
	LI Ke, TAI Nengling, ZHANG Shenxi. Comprehensive optimal dispatch of distribution network based on improved particle swarm optimization algorithm[J]. Journal of Shanghai Jiao Tong University, 2017, 51(8): 897-902.
[7]	章坚民, 邱程峰, 王锋华, 等. 变电站为中心配电网单线图(二)初始自动布局[J]. 电力系统自动化, 2019, 43(19): 124-131.
	ZHANG Jianmin, QIU Chengfeng, WANG Fenghua, et al. Single-line diagram of substation centralized distribution network part two initial automatic layout[J]. Automation of Electric Power Systems, 2019, 43(19): 124-131.
[8]	章坚民, 楼坚. 基于CIM/SVG和面向对象的配电单线图自动生成[J]. 电力系统自动化, 2008, 32(22): 61-65.
	ZHANG Jianmin, LOU Jian. Automatic generation of single-line diagram for distribution feeder systems based on CIM/SVG and object orientation[J]. Automation of Electric Power Systems, 2008, 32(22): 61-65.
[9]	程远, 严伟, 李晓明. 基于斥力-张力模型的网络拓扑图布局算法[J]. 计算机工程, 2004, 30(3): 104-105.
	CHENG Yuan, YAN Wei, LI Xiaoming. Network topology autolayout algorithm based on repulsion-tension model[J]. Computer Engineering, 2004, 30(3): 104-105.
[10]	邱程峰, 章坚民, 刘理峰, 等. 变电站为中心配电网单线图(三)美化布局计算[J]. 电力系统自动化, 2019, 43(20): 170-175.
	QIU Chengfeng, ZHANG Jianmin, LIU Lifeng, et al. Single-line diagram of substation centralized distribution network part three beautifying layout calculation[J]. Automation of Electric Power Systems, 2019, 43(20): 170-175.
[11]	陈兵, 赵肖旭, 施伟成, 等. 配电网网格化自动成图的实现[J]. 电力工程技术, 2017, 36(6): 100-105.
	CHEN Bing, ZHAO Xiaoxu, SHI Weicheng, et al. Realization of automatic grid generation in distribution network[J]. Electric Power Engineering Technology, 2017, 36(6): 100-105.
[12]	ZHANG H, SINDAGI V, PATEL V M. Image de-raining using a conditional generative adversarial network[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2020, 30(11): 3943-3956.
[13]	盛志文, 王镜毓, 石东源. 基于图自编码器的输电网单线图布局自适应生成方法[J]. 中国电机工程学报, 2022, 42(6): 2133-2144.
	SHENG Zhiwen, WANG Jingyu, SHI Dongyuan. An adaptive generation method for transmission network one-line diagram layouts based on graph autoencoder[J]. Proceedings of the CSEE, 2022, 42(6): 2133-2144.
[14]	CHENG Y B, YU N P, FOGGO B, et al. Online power system event detection via bidirectional generative adversarial networks[J]. IEEE Transactions on Power Systems, 2022, 37(6): 4807-4818.
[15]	黎寿涛, 夏成军, 钟明明, 等. 基于故障链聚类算法的电网关键线路辨识[J]. 电力工程技术, 2022, 41(1): 84-92.
	LI Shoutao, XIA Chengjun, ZHONG Mingming, et al. Critical line identification of power grid based on fault chain clustering algorithm[J]. Electric Power Engineering Technology, 2022, 41(1): 84-92.
[16]	肖峻, 祖国强, 贺琪博, 等. 馈线联络对配电网最大供电能力的作用机理[J]. 电力系统自动化, 2015, 39(8): 38-44.
	XIAO Jun, ZU Guoqiang, HE Qibo, et al. Influence mechanism of feeder links on total supply capability for distribution networks[J]. Automation of Electric Power Systems, 2015, 39(8): 38-44.
[17]	CHEN J M, JIANG W, XU Z Q, et al. Hierarchical distribution network topology formulation and dimensionality reduction using homeomorphism transformation[J]. IEEE Access, 2022, 10: 33320-33331.
[18]	BALDAN M, DI BARBA P. Discovering Pareto-optimal magnetic-design solutions via a generative adversarial network[J]. IEEE Transactions on Magnetics, 2022, 58(9): 7200704.
[19]	LU K D, WU Z G. Genetic algorithm-based cumulative sum method for jamming attack detection of cyber-physical power systems[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 9004810.
[20]	HAGBERG A A, SCHULT D A, SWART P J. Exploring network structure, dynamics, and function using NetworkX[C]// Proceedings of the 7th Python in Science Conference. Pasadena, California: SciPy, 2008: 11-15.

模型	最低分数	平均分数
MLP	1.32	1.97
CGAN	1.43	2.07
MSE-CGAN	1.70	2.24
启发式 (Heuristic)	1.80	2.37
MSE-CGAN+GA	1.91	2.50

种群规模	交叉率	变异率	布局平均得分
100	0.5	0.01	2.29
100	0.5	0.05	2.32
100	0.5	0.10	2.31
100	0.6	0.05	2.36
100	0.6	0.10	2.34
500	0.5	0.05	2.42
500	0.5	0.10	2.41
500	0.6	0.05	2.50
500	0.6	0.10	2.47
500	0.7	0.05	2.49
800	0.7	0.10	2.51
1000	0.7	0.10	2.52

基于生成对抗网络的中压配网供电单元自动成图方法

Automatic Mapping Method for Power Supply Units in Medium-Voltage Distribution Networks Based on Generative Adversarial Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 20

相关文章 6

编辑推荐

Metrics

本文评价

[1]	毕金茂, 张朋, 张洁, 赵春财, 崔利. 不完备数据下的聚酯熔体特性黏度预测方法[J]. 上海交通大学学报, 2024, 58(4): 534-544.
[2]	鄢丛强1,2, 郭正玉3,4, 蔡云泽 1,2. 基于改进CycleGAN的SAR图像舰船尾迹数据增强[J]. J Shanghai Jiaotong Univ Sci, 2024, 29(4): 702-711.
[3]	郭俊锋, 王淼生, 王智明. 基于对不平衡数据集进行二次迁移学习的滚动轴承剥落类故障诊断方法[J]. 上海交通大学学报, 2023, 57(11): 1512-1521.
[4]	李钰, 杨道勇, 刘玲亚, 王易因. 利用生成对抗网络实现水下图像增强[J]. 上海交通大学学报, 2022, 56(2): 134-142.
[5]	何新林, 戚宗锋, 李建勋. 基于隐变量后验生成对抗网络的不平衡学习[J]. 上海交通大学学报, 2021, 55(5): 557-565.
[6]	祁生勇, 臧月进, 吕国云, 杜明. 基于生成对抗网络的空中目标图像生成算法研究[J]. 空天防御, 2021, 4(2): 67-73.