[1]李 智,孟凡敏,周 蕾,等.考虑源网荷变化的数据驱动潮流样本生成方法[J].机械与电子,2026,44(07):112-119.
 LI Zhi,MENG Fanmin,ZHOU Lei,et al.A Data-driven Method for Generating Trend Samples Considering Changes in Source Network Load[J].Machinery & Electronics,2026,44(07):112-119.
点击复制

考虑源网荷变化的数据驱动潮流样本生成方法()
分享到:

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
44
期数:
2026年07期
页码:
112-119
栏目:
电力控制
出版日期:
2026-07-25

文章信息/Info

Title:
A Data-driven Method for Generating Trend Samples Considering Changes in Source Network Load
文章编号:
1001-2257(2026)07-0112-08
作者:
李 智孟凡敏周 蕾秦子健刘世超亓晓燕
国网山东省电力公司莱芜供电公司,山东 济南 271100
Author(s):
LI ZhiMENG FanminZHOU LeiQIN ZijianLIU ShichaoQI Xiaoyan
(Laiwu Power Supply Company,State Grid Shandong Electric Power Company,Jinan 271100,China)
关键词:
潮流样本生成电力系统数据驱动蒙特卡洛抽样网络拓扑
Keywords:
power flow sample generationpower systemdata driven methodMonte Carlo sampling network topology
分类号:
TM744
文献标志码:
A
摘要:
针对传统数据驱动潮流计算方法未能有效处理因系统网络拓扑变化导致数据模型难以实现高精度潮流计算的问题,提出一种考虑源网荷变化的数据驱动潮流样本生成方法。首先,基于系统“源荷”侧变化特性采用改进的蒙特卡洛抽样方法生成基础样本;其次,基于网络拓扑变化前后的节点电压差,提出一种适用于表征网络拓扑变化的数据驱动潮流样本抽样方法;最后,利用深度神经网络构建数据驱动潮流计算模型,实现对系统运行状态的快速计算与性能评估。算例分析表明,所提方法能够有效提升潮流样本对拓扑变化场景的表征能力,从而提高数据驱动潮流计算模型的精度与泛化性能。
Abstract:
Conventional data-driven power flow computation methods often fail to maintain high accuracy when the system network topology changes,as the underlying data model cannot adequately adapt to such variations.To address this issue,this paper proposes a data-driven power flow sample generation method that explicitly accounts for source-network-load variations.First,an improved Monte Carlo sampling approach is adopted to generate basic samples by taking into account the variation characteristics of source and load sides.Subsequently,by exploiting the nodal voltage difference before and after a topology change,a data-driven sampling strategy is developed to characterize network topology variations.Finally, a deep neural network is employed to construct a data-driven power flow computation model,enabling rapid evaluation of system operating states and performance.Case studies demonstrate that the proposed method can effectively enhance the capability of power flow samples to characterize topology variation scenarios, thereby improving both the accuracy and generalization performance of data-driven power flow models.

参考文献/References:

[1] 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):1799-1817.

[2] 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):2806-2818.
[3] 汤伟,吴昊.基于网络拓扑结构的全网潮流计算模型自动生成[J].现代电力,2013,30(2):13-16.
[4] 唐灿,董树锋,任雪桂,等.用于迭代法潮流计算的改进 Jacobi预处理方法[J].电力系统自动化,2018,42(12):81-86.
[5] 樊哲新,杨知方,冯楠,等.潮流方程线性化误差分析及最优状态空间选取[J].电力系统自动化,2020,44(1):109-117.
[6] 苏晨博,刘崇茹,李至峪,等.基于贝叶斯理论的考虑多维风速之间相关性的概率潮流计算[J].电力系统自动化,2021,45(3):157-165.
[7] 马丁,沈沉,陈颖,等.深度学习应用于潮流计算判敛问题的可行性研究[J].南方电网技术,2020,14(2):46-54.
[8] 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.
[9] 卫志农,张清松,赵静波,等.电力系统线性化模型研究综述与改进[J].电网技术,2017,41(9):2919-2927.
[10] 赵渊,王洁,耿莲,等.电网可靠性非序贯蒙特卡洛仿真的扩展交叉熵法[J].中国电机工程学报,2017,37(7):1963-1974.
[11] 彭显刚,林利祥,刘艺,等.基于纵横交叉拉丁超立方采样蒙特卡洛模拟法的分布式电源优化配置[J].中国电机工程学报,2015,35(16):4077-4085.
[12] Sreenuch T,Alghassi A,Perinpanayagam S,et al. Probabilistic Monte-Carlo method for modelling and prediction of electronics component life [J].International Journal of Advanced Computer Science and Applications,2014,5(1):113.
[13] 韩海腾,高山,吴晨,等.基于Nataf变换的电网不确定性多点估计法[J].电力系统自动化,2015,39(7):8-34.
[14] 张惠娟,冯垚,高建森,等.基于蒙特卡罗抽样半不变量法的微电网随机潮流计算[J].电力系统及其自动化学报,2017,29(1):58-62.
[15] 曾琮,黄强,陈德,等.基于改进纵横交叉算法的电网最优潮流计算[J].中国电力,2021,54(9):9-16.
[16] 郎燕生,李静,罗雅迪,等.基于图划分的大电网拓扑分析[J].电力系统保护与控制,2017,45(23):108-115.
[17] Wang Lei,Zhou Qun,Jin Shuangshuang.Physics-guided deep learning for power system state estimation[ J].Journal of Modern Power Systems and CleanEnergy,2020,8(4):607-615.
[18] Yang Yan,Yang Zhifang,Yu Juan,et al.Fast calculation of probabilistic power flow:a model based deep learning approach[J].IEEE Transactions on Smart Grid,2020,11 (3):2235-2244.
[19] 董昱,董存,于若英,等.基于线性最优潮流的电力系统新能源承载能力分析[J].中国电力,2022,55(3):1-8.
[20] 战鹏祥,黄飞虎,廖思睿,等.基于物理机理引导的数据驱动潮流计算方法[J].电网技术,2024,48(12):5034-5045.
[21] 杨浩,朱宇迪,刘钺铖,等.基于MLSSVR数据驱动的潮流非线性回归及其灵敏度解析[J].中国电机工程学报,2022,42(21):7706-7718.
[22] Wang Dawei,Zheng Kedi,Chen Qixin,et al.A data-driven probabilistic power flow method based on convolutional neural networks[J].International Transactions on Electrical Energy Systems,2020,30(7):e12367.
[23] 周明,夏澍,李琰,等.含风电的电力系统月度机组组合和检修计划联合优化调度[J].中国电机工程学报,2015,35(7):1586-1595.
[24] 米阳,卢长坤,申杰,等.基于条件生成对抗网络的风电功率极端场景生成[J].高电压技术,2023,49(6):2253-2263.
[25] 廖一帆,武志刚.基于迁移学习与Wasserstein生成对抗网络的静态电压稳定临界样本生成方法[J].电网技术,2021,45(9):3722-3728.
[26] 胡家祥,曹迪,胡维昊,等.融入拓扑知识的图神经网络配电网抗差状态估计方法[J].电力系统自动化,2023,47(10):84-97.
[27] 宋智强,黄耀辉,赵化时,等.含LCC-MMC串/并联混合直流的交直流系统潮流计算方法[J].电网技术,2023,47(7):2860-2868.
[28] 徐原,马伟哲,赵利刚,等.基于CIM 模型的多粒度潮流计算数据生成方法[J].南方电网技术,2022,16(3):40-47.

相似文献/References:

[1]陈刚,宁楠,张广梅,等.广域备自投控制系统策略在电网中的应用与研究[J].机械与电子,2017,(08):50.[doi:1001-2257(2017)08-0050-04]
 CHEN Gang,NING Nan,ZHANG Guangmei,et al.On the Application and Research of Wide-Area Self-Cast Control System Strategy in Power Grid[J].Machinery & Electronics,2017,(07):50.[doi:1001-2257(2017)08-0050-04]
[2]潘可佳,王鑫,杨帆,等.面向电力大数据的多源异构数据融合技术研究[J].机械与电子,2017,(09):7.[doi:1001-2257(2017)09-0007-05]
 PAN Kejia,WANG Xin,YANG Fan,et al.Research on Multi-Source Heterogeneous Data Fusion Technology for the Big Data of Power[J].Machinery & Electronics,2017,(07):7.[doi:1001-2257(2017)09-0007-05]
[3]刘燕,王婷婷,高 虹.智能电网环境下负荷响应对系统消纳风电能力影响模型探析[J].机械与电子,2020,(01):10.
 ,,et al.Research on the Model of the Impact of Load Response on Wind Power Absorption in Smart Grid Environment[J].Machinery & Electronics,2020,(07):10.
[4]韦 雅1,文耀宽1,田 珂2,等.基于时间序列线性大数据分析的电力系统潮流计算方法[J].机械与电子,2020,(01):32.
 ,,et al.Power Flow Calculation Method Based on Time Series Linear Big Data Analysis[J].Machinery & Electronics,2020,(07):32.
[5]骆东松,魏義民,张杰锋.基于经验模态分解和优化 BiLSTM 的短期负荷预测[J].机械与电子,2024,42(09):11.
 LUO Dongsong,WEI Yimin,ZHANG Jiefeng.Short-term Load Forecasting Based on Empirical Modal Decomposition and Optimized BiLSTM[J].Machinery & Electronics,2024,42(07):11.
[6]岳聚欣,杨冬梅,李先允,等.基于量测数据的含风电电力系统等效惯量估计[J].机械与电子,2025,(06):9.
 YUE Juxin,YANG Dongmei,LI Xianyun,et al.Equivalent Inertia Estimation of Wind Power Systems Based on Measurement Data[J].Machinery & Electronics,2025,(07):9.

备注/Memo

备注/Memo:
收稿日期:2026-04-14
基金项目:国网山东省电力公司科技项目资助(520612250001)
作者简介:李 智 (1983-),男,山东淄博人,博士,高级工程师,研究方向为电力系统与自动化。
更新日期/Last Update: 2026-08-28