基于<em>N</em>/<em>N</em>-1潮流内嵌图卷积神经网络的电网运行方式智能调整方法

doi:10.19595/j.cnki.1000-6753.tces.241728

Abstract
Figure/Table
References (0)
Related Citation (2)

Download: PDF (1931 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract

The adjustment of the operation mode is crucial for ensuring the safety and stability of the power grid, and manual adjustment is still the main method in current industry practices. However, with the integration of numerous renewable energy sources and the application of power electronic equipment, both the scale and complexity of the power grid continue to increase, resulting in the manual adjustment method that relies on manual experience to adjust repeatedly by trial-and-error method faces the following severe challenges: low efficiency, lack of theoretical guidance. Recently, some methods for operation mode adjustment have been proposed, but most of them cannot provide the transition process and intrinsic connection between the operation modes before and after the adjustment, making it difficult for the staff to explain and analyze the operation mode adjustment issues, which affects the decision-making efficiency. To solve these problems, this paper proposes an intelligent adjustment method of power grid operation mode based on N/N-1 power flow embedded graph convolutional neural network (GCNN).
Firstly, the forward propagation expression of the convolutional module of the design diagram is derived from the N/N-1 power flow physical model, and the convolutional forward propagation strategy based on N/N-1 power flow embedding is proposed to effectively extract the node and branch features of the power system in N/N-1 conditions. Then, taking power flow characteristics in N/N-1 conditions as input/output characteristics, the N/N-1 power flow coupling model based on multi-layer graph convolution and convolutional neural network module collaboration is constructed to characterize the data-driven power flow coupling relationship in N/N-1 conditions. Next, aiming at the operation mode of power flow exceeding the limit under N/N-1 conditions, an intelligent adversarial adjustment method based on N/N-1 power flow coupling relationship is proposed to accurately analyze the adjustment process before and after the adjustment and obtain the effective adjustment strategy for the operation mode, achieving the adjustment of the operation mode of power flow exceeding the limit under N/N-1 conditions to meet the N-1 verification.
The simulation results on the IEEE 30-bus system show that the accuracy of the proposed N/N-1 power flow coupling model for predicting the nodal voltage magnitude, phase angle, branch active power, and reactive power is 99.48%, 99.72%, 99.32%, and 99.81% respectively, which is higher than that of the prediction models constructed based on other neural network architectures (fully connected neural network, convolutional neural network, graph convolutional neural network), indicating that the proposed model can accurately characterize the power flow coupling relationship under N/N-1 conditions. To verify the effectiveness of the proposed intelligent adjustment method for power grid operation mode, the operation mode with power flow limit exceeding in the IEEE 30-bus system and 341-bus system in an area is adjusted, and the adjustment process before and after the adjustment can be accurately analyzed, and the effective adjustment strategy for the operation mode can be obtained. Finally, to prove the convergence of this method, the proposed method is applied to adjust 1000 operation modes with power flow limit exceeding in the two power systems, and the results show that these operation modes can be adjusted to meet the N-1 verification with fewer adjustment times.
The following conclusions can be drawn from the simulation analysis: (1) Compared with the models based on fully connected neural network, convolutional neural network and graph convolutional neural network architectures, the proposed N/N-1 power flow coupling model can more effectively and accurately characterize the power flow coupling relationship under N/N-1 conditions. (2) The proposed intelligent adversarial adjustment method can adjust the operation mode with limit exceeding to meet all static N-1 verification with fewer adjustment times and accurately analyze the adjustment process between the operation modes before and after the adjustment, and it provides the detailed basis for the staff to make the decision of operation mode adjustment.

Key words： Operation mode adjustment N-1 verification graph convolutional neural network power flow embedded adversarial process

Received: 07 October 2024

PACS:

TM761

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Duan Shiqi
	Yu Juan
	Yang Zhifang
	Chen Tao
	Zhu Shengyi

Cite this article:

Duan Shiqi,Yu Juan,Yang Zhifang等. Intelligent Adjustment Method of Power Grid Operation Mode Based on N/N-1 Power Flow Embedded Graph Convolutional Neural Network[J]. Transactions of China Electrotechnical Society, 0, (): 20241728-20241728.

URL:

https://dgjsxb.ces-transaction.com/EN/10.19595/j.cnki.1000-6753.tces.241728 OR https://dgjsxb.ces-transaction.com/EN/Y0/V/I/20241728

[1] 张智刚, 康重庆. 碳中和目标下构建新型电力系统的挑战与展望[J]. 中国电机工程学报, 2022, 42(8): 2806-2819.
Zhang Zhigang, Kang Chongqing.Challenges and prospects for constructing the new-type power system towards a carbon neutrality future[J]. Proceedings of the CSEE, 2022, 42(8): 2806-2819.
[2] 国家能源局.国家能源局综合司关于完善电力系统运行方式分析制度强化电力系统运行安全风险管控的通知:国能综通安全[2023] 13号[EB/OL].(2023-02-17)[2024-06-29] .https://zfxxgk.nea.gov.cn/2023-02/17/c_1310700939.htm.
李军徽,邵岩,朱星旭,等.计及碳排放量约束的多区域互联电力系统分布式低碳经济调度[J].电工技术学报,2023,38(17):4715-4728.
Li Junhui,Shao Yan,Zhu Xingxu,et al.Carbon emissions constraint distributed low-carbon economic dispatch of power system[J].Transactions of China Electrotechnical Society,2023,38(17):4715-4728.
[3] 李军徽, 邵岩, 朱星旭, 等. 计及碳排放量约束的多区域互联电力系统分布式低碳经济调度[J]. 电工技术学报, 2023, 38(17): 4715-4728.
Li Junhui, Shao Yan, Zhu Xingxu, et al.Carbon emissions constraint distributed low-carbon economic dispatch of power system[J]. Transactions of China Electrotechnical Society, 2023, 38(17): 4715-4728.
[4] 杨珺, 侯俊浩, 刘亚威, 等. 分布式协同控制方法及在电力系统中的应用综述[J]. 电工技术学报, 2021, 36(19): 4035-4049.
Yang Jun, Hou Junhao, Liu Yawei, et al. Distributed cooperative control method and application in power system[J]. Transactions of China Electrotechnical Society, 2021, 36(19): 4035-4049.
[5] 高倩, 杨知方, 李文沅. 电力系统混合整数线性规划问题的运筹决策关键技术综述与展望[J]. 电工技术学报, 2024, 39(11): 3291-3307.
Gao Qian, Yang Zhifang, Li Wenyuan. Prospect on operations research for mixed-integer linear programming problems in power systems[J]. Transactions of China Electrotechnical Society, 2024, 39(11): 3291-3307.
[6] 王方雨, 王海云, 于希娟, 等. 考虑越限程度的断面功率灵敏度控制方法[J]. 电网技术, 2023, 47(9): 3847-3855.
Wang Fangyu, Wang Haiyun, Yu Xijuan, et al. Section power sensitivity control considering out-of-limit degrees[J]. Power System Technology, 2023, 47(9): 3847-3855.
[7] 陈海荣, 方健. 一种新能源场站紧急态监控装置的最优化切机算法[J]. 电气技术, 2023, 24(8): 44-49.
Chen Hairong, Fang Jian. An optimal generator tripping method for emergency monitoring and control devices in new energy stations[J]. Electrical Engineering, 2023, 24(8): 44-49.
[8] 吴宇航, 王涛, 范辉, 等. 利用灵敏度分析的规模风电并网系统连锁故障阻断控制[J]. 电网技术, 2023, 47(9): 3743-3755.
Wu Yuhang, Wang Tao, Fan Hui, et al. Blocking control of cascading failures for large-scale wind power system using sensitivity analysis[J]. Power System Technology, 2023, 47(9): 3743-3755.
[9] 陈厚合, 邵俊岩, 姜涛, 等. 基于参数灵敏度的综合能源系统安全控制策略研究[J]. 中国电机工程学报, 2020, 40(15): 4831-4843.
Chen Houhe, Shao Junyan, Jiang Tao, et al. Security control strategy for integrated energy system using parameter sensitivity[J]. Proceedings of the CSEE, 2020, 40(15): 4831-4843.
[10] 曾泓泰, 郭庆来, 周艳真, 等. 面向电网运行方式计算的不收敛潮流无功调整方法[J]. 电力系统保护与控制, 2022, 50(19): 1-12.
Zeng Hongtai, Guo Qinglai, Zhou Yanzhen, et al. Reactive power adjustment method of non-convergent power flow for power system operation mode calculation[J]. Power System Protection and Control, 2022, 50(19): 1-12.
[11] 杨茂, 王金鑫. 考虑可再生能源出力不确定的孤岛型微电网优化调度[J]. 中国电机工程学报, 2021, 41(3): 973-985.
Yang Mao, Wang Jinxin. Optimal scheduling of islanded microgrid considering uncertain output of renewable energy[J]. Proceedings of the CSEE, 2021, 41(3): 973-985.
[12] 李书益. 基于遗传模拟退火粒子群算法的微电网优化运行研究[D]. 南京: 南京邮电大学, 2020.
Li Shuyi. Research on optimal operation of microgrid based on genetic simulated annealing particle swarm optimization algorithm[D]. Nanjing: Nanjing University of Posts and Telecommunications, 2020.
[13] 聂瀚, 杨文荣, 马晓燕, 等. 基于改进鸟群算法的离网微电网优化调度[J]. 燕山大学学报, 2019, 43(3): 228-237.
Nie Han, Yang Wenrong, Ma Xiaoyan, et al. Optimal scheduling of islanded microgrid based on improved bird swarm optimization algorithm[J]. Journal of Yanshan University, 2019, 43(3): 228-237.
[14] 吴熙, 王梦婷, 王亮, 等. 考虑UPFC控制模式的N-1安全约束最优潮流及应用[J]. 电力系统自动化, 2020, 44(9): 43-51.
Wu Xi, Wang Mengting, Wang Liang, et al. N-1 security constrained optimal power flow considering control modes of unified power flow controller and its application[J]. Automation of Electric Power Systems, 2020, 44(9): 43-51.
[15] 张兆毅, 胡浩, 王子江, 等. 基于非线性仿射的风电场电压实时计算和优化方法[J]. 电工技术学报, 2024, 39(13): 3975-3989.
Zhang Zhaoyi, Hu Hao, Wang Zijiang, et al. Real-time voltage calculation and optimization method for wind farms based on nonlinear affine transformation[J]. Transactions of China Electrotechnical Society, 2024, 39(13): 3975-3989.
[16] Wang Licheng, Yang Yu, Gu Huajie, et al. Bottleneck generator identification and the corresponding N-1 frequency security constrained intraday generator dispatch[J]. IEEE Transactions on Power Systems, 2023, 38(1): 739-752.
[17] 林雨眠, 熊厚博, 张笑演, 等. 计及新能源机会约束与虚拟储能的电-热系统分布式多目标优化调度[J]. 电工技术学报, 2024, 39(16): 5042-5059.
Lin Yumian, Xiong Houbo, Zhang Xiaoyan, et al. Distributed multi-objective optimal scheduling of integrated electric-heat system considering chance constraint of new energy and virtual storage[J]. Transactions of China Electrotechnical Society, 2024, 39(16): 5042-5059.
[18] Yang Yan, Yang Zhifang, Yu Juan, et al. Fast economic dispatch in smart grids using deep learning: an active constraint screening approach[J]. IEEE Internet of Things Journal, 2020, 7(11): 11030-11040.
[19] Moreira A, Valenzuela A, Heleno M. Solving market-based large-scale security-constrained AC optimal power flows[J]. IEEE Transactions on Power Systems, 2023, 38(6): 5088-5101.
[20] 王甜婧, 汤涌, 王兵, 等. 传统方法与人工智能:潮流控制优化算法的现状、挑战与未来方向[J]. 中国电机工程学报, 2023, 43(5): 1799-1818.
Wang Tianjing, Tang Yong, Wang Bing, et al.Traditional methods versus artificial intelligence:optimization algorithms for power flow control in state of the art, challenge and future directions[J]. Proceedings of the CSEE, 2023, 43(5): 1799-1818.
[21] 冉晴月, 林伟, 杨知方, 等. 基于可信深度神经网络的最优潮流计算方法[J]. 电工技术学报, 2024, 39(21): 6687-6699.
Ran Qingyue, Lin Wei, Yang Zhifang, et al.Optimal power flow calculation based on a trustworthy deep neural network[J]. Transactions of China Electrotechnical Society, 2024, 39(21): 6687-6699.
[22] 张松涛, 张东霞, 黄彦浩, 等. 基于改进直流潮流算法的潮流计算收敛自动调整方法研究[J]. 电网技术, 2021, 45(1): 86-97.
Zhang Songtao, Zhang Dongxia, Huang Yanhao, et al. Research on automatic power flow convergence adjustment method based on modified DC power flow algorithm[J]. Power System Technology, 2021, 45(1): 86-97.
[23] 杨晓东, 严剑峰, 刘佳霖. 结合深度强化学习与人工经验的电网输电断面功率调整方法[J]. 电力系统自动化, 2023, 47(15): 133-141.
Yang Xiaodong, Yan Jianfeng, Liu Jialin. Power adjustment method for transmission section in power grid combining deep reinforcement learning and artificial experience[J]. Automation of Electric Power Systems, 2023, 47(15): 133-141.
[24] Yan Ziming, Xu Yan. A hybrid data-driven method for fast solution of security-constrained optimal power flow[J]. IEEE Transactions on Power Systems, 2022, 37(6): 4365-4374.
[25] Liu Tingjian, Liu Youbo, Liu Junyong, et al. A Bayesian learning based scheme for online dynamic security assessment and preventive control[J]. IEEE Transactions on Power Systems, 2020, 35(5): 4088-4099.
[26] Duan Jiajun, Shi Di, Diao Ruisheng, et al. Deep-reinforcement-learning-based autonomous voltage control for power grid operations[J]. IEEE Transactions on Power Systems, 2020, 35(1): 814-817.
[27] 胡丹尔, 彭勇刚, 韦巍, 等. 多时间尺度的配电网深度强化学习无功优化策略[J]. 中国电机工程学报, 2022, 42(14): 5034-5045.
Hu Daner, Peng Yonggang, Wei Wei, et al. Multi-timescale deep reinforcement learning for reactive power optimization of distribution network[J]. Proceedings of the CSEE, 2022, 42(14): 5034-5045.
[28] 张剑, 崔明建, 何怡刚. 结合数据驱动与物理模型的主动配电网双时间尺度电压协调优化控制[J]. 电工技术学报, 2024, 39(5): 1327-1339.
Zhang Jian, Cui Mingjian, He Yigang. Dual timescales coordinated and optimal voltages control in distribution systems using data-driven and physical optimization[J]. Transactions of China Electrotechnical Society, 2024, 39(5): 1327-1339.
[29] 王甜婧, 汤涌, 郭强, 等. 基于知识经验和深度强化学习的大电网潮流计算收敛自动调整方法[J]. 中国电机工程学报, 2020, 40(8): 2396-2406.
Wang Tianjing, Tang Yong, Guo Qiang, et al. Automatic adjustment method of power flow calculation convergence for large-scale power grid based on knowledge experience and deep reinforcement learning[J]. Proceedings of the CSEE, 2020, 40(8): 2396-2406.
[30] Gao Maosheng, Yu Juan, Yang Zhifang, et al. A physics-guided graph convolution neural network for optimal power flow[J]. IEEE Transactions on Power Systems, 2024, 39(1): 380-390.
[31] Dong Yinpeng, Liao Fangzhou, Pang Tianyu, et al. Boosting adversarial attacks with momentum[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 2018: 9185-9193.
[32] Duan Shiqi, Yu Juan, Yang Zhifang, et al. An intelligent power flow violation adjustment method based on adversarial process[C]//2023 IEEE 7th Conference on Energy Internet and Energy System Integration (EI2), Hangzhou, China, 2023: 4774-4779.
[33] Gao Maosheng, Yu Juan, Yang Zhifang, et al. Physics embedded graph convolution neural network for power flow calculation considering uncertain injections and topology[J]. IEEE Transactions on Neural Networks and Learning Systems, 2024, 35(11): 15467-15478.
[34] Yang Zhifang, Xie Kaigui, Yu Juan, et al. A general formulation of linear power flow models: basic theory and error analysis[J]. IEEE Transactions on Power Systems, 2019, 34(2): 1315-1324.
[35] Lei Yunwen, Tang Ke. Learning rates for stochastic gradient descent with nonconvex objectives[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, 43(12): 4505-4511.
[36] Goodfellow I J, Shlens J, Szegedy C.Explaining and harnessing adversarial examples[C]//Proceedings of the 3rd International Conference on Learning Representations,San Diego City, CA, USA, 2015.