基于Transformer-GAT的新型电力系统宽频振荡源定位

张清源; 周波; 池建飞; 赵妍; 陶亮; 胡枭

doi:10.12204/j.issn.1000-7229.2025.10.008

PDF(3429 KB)

电力建设 ›› 2025, Vol. 46 ›› Issue (10) : 88-98. DOI: 10.12204/j.issn.1000-7229.2025.10.008

调度运行

基于Transformer-GAT的新型电力系统宽频振荡源定位

张清源 ¹ ,
周波 ² ,
池建飞 ² ,
赵妍 ³ ,
陶亮 ⁴ ,
胡枭 ¹

作者信息 +

Wide-Frequency Oscillatory Source Localization of a New Power System Based on Transformer-GAT

ZHANG Qingyuan ¹ ,
ZHOU Bo ² ,
CHI Jianfei ² ,
ZHAO Yan ³ ,
TAO Liang ⁴ ,
HU Xiao ¹

Author information +

文章历史 +

摘要

【目的】为解决大规模新能源并网导致新型电力系统安全问题由传统工频段扩展到中高频段的问题，推动基于人工智能的宽频振荡源定位方法系统性研究，提出了一种基于Transformer和图注意力神经网络（graph attention neural network， GAT）相结合的宽频振荡源定位方法。【方法】首先在子站侧，利用Transformer网络的强大信号处理和特征提取能力，对电力系统的量测信号进行高效的编码压缩，以保证在有限带宽条件下有效传输关键的宽频振荡信息，并减少数据的冗余。其次在主站侧，结合压缩信号特征与系统网络拓扑结构利用GAT对振荡源进行定位。最后，利用含风电场的四机两区系统进行验证。【结果】仿真和实验结果表明，所提Transformer编码器可以从宽频振荡信号中提取有效特征以实现对子站信号的压缩；所提GAT模型可以实现宽频振荡源准确定位。在与其他算法进行的对比实验中，GAT模型在定位振荡源时具有较低的误报成本，并且保持灵敏度和特异度之间的平衡。【结论】Transformer-GAT方法通过信号特征与网络拓扑的协同分析，有效提升宽频振荡源定位精度与鲁棒性，为新型电力系统稳定运行提供技术支撑。

Abstract

[Objective] To solve the problem of new power system security caused by large-scale new-energy-grid connections from traditional industrial frequency bands to medium- and high-frequency bands， and to promote systematic research on a broadband oscillation source location method based on artificial intelligence， a wide-frequency oscillation source location method based on a transformer and graph attention neural network （GAT） is proposed. [Methods] First， on the substation side， the powerful signal processing and feature extraction capabilities of the transformer network were used to efficiently encode and compress the measurement signals of the power system to ensure the effective transmission of key broadband oscillation information under limited bandwidth conditions and reduce data redundancy. Subsequently， on the main-station side， combined with the compressed signal characteristics and system network topology， the GAT was used to locate the oscillation source. Finally， a four-machine， two-area system with a wind farm was used for verification. [Results] Simulation and experimental results show that the proposed transformer encoder can extract effective features from wide-frequency oscillation signals to realize the compression of substation signals. The proposed GAT model can achieve the accurate positioning of a wide-frequency oscillation source. In the comparison experiments with other algorithms， the GAT model had a lower false alarm cost when locating the oscillation source and maintained a balance between sensitivity and specificity. [Conclusions] Through the collaborative analysis of signal characteristics and network topology， the Transformer-GAT method effectively improves the positioning accuracy and robustness of wide-frequency oscillation sources. It provides technical support for the stable operation of new power systems.

导出引用

张清源, 周波, 池建飞, 等. 基于Transformer-GAT的新型电力系统宽频振荡源定位[J]. 电力建设. 2025, 46(10): 88-98 https://doi.org/10.12204/j.issn.1000-7229.2025.10.008

ZHANG Qingyuan, ZHOU Bo, CHI Jianfei, et al. Wide-Frequency Oscillatory Source Localization of a New Power System Based on Transformer-GAT[J]. Electric Power Construction. 2025, 46(10): 88-98 https://doi.org/10.12204/j.issn.1000-7229.2025.10.008

中图分类号： TM712

参考文献

列表( 原文顺序 | 文献年度倒序 | 文中引用次数倒序 ) 可视化分析

[1]	谢小荣, 贺静波, 毛航银, 等. “双高” 电力系统稳定性的新问题及分类探讨[J]. 中国电机工程学报, 2021, 41(2): 461-475. XIE Xiaorong, HE Jingbo, MAO Hangyin, et al. New issues and classification of power system stability with high shares of renewables and power electronics[J]. Proceedings of the CSEE, 2021, 41(2): 461-475. 本文引用 [1]

[2]	陈露洁, 徐式蕴, 孙华东, 等. 高比例电力电子电力系统宽频带振荡研究综述[J]. 中国电机工程学报, 2021, 41(7): 2297-2310. CHEN Lujie, XU Shiyun, SUN Huadong, et al. A survey on wide-frequency oscillation for power systems with high penetration of power electronics[J]. Proceedings of the CSEE, 2021, 41(7): 2297-2310. 本文引用 [1]

[3]	卢有飞, 邹时容, 刘璐豪, 等. 基于知识共享的高比例可再生能源系统发电控制[J]. 高压电器, 2024, 60(10): 33-45. LU Youfei, ZOU Shirong, LIU Luhao, et al. Generation control of high percentage renewable energy systems based on knowledge sharing[J]. High Voltage Apparatus, 2024, 60(10): 33-45. 本文引用 [1]

[4]

罗国敏, 赵祎薇, 吴梦宇, 等. 适应高比例新能源接入的配电网改进型自适应电流保护方法[J]. 供用电, 2024, 41(9): 12-23.

LUO

Guomin

, ZHAO

Yiwei

, WU

Mengyu

, et al. Improved adaptive current protection method suitable for distribution network adapted to high proportion of new energy source accessed[J]. Distribution & Utilization, 2024, 41(9): 12-23.

本文引用 [1]

[5]	唐文虎, 聂欣昊, 钱瞳, 等. 面向新型电力系统安全稳定的储能应用技术研究综述与展望[J]. 广东电力, 2024, 37(12): 3-15. TANG Wenhu, NIE Xinhao, QIAN Tong, et al. Review and prospect on application technologies of energy storage for safety and stability of new power system[J]. Guangdong Electric Power, 2024, 37(12): 3-15. 本文引用 [1]

[6]

张东辉, 陈新, 杨舒婷, 等. 含静止无功补偿装置的光伏电站高频谐振分析及抑制策略研究[J]. 中国电机工程学报, 2023, 43(24): 9580-9594.

ZHANG

Donghui

, CHEN

Xin

, YANG

Shuting

, et al. Analysis of high-frequency resonance and suppression strategy of photovoltaic power plant with static reactive power compensation device[J]. Proceedings of the CSEE, 2023, 43(24): 9580-9594.

本文引用 [1]

[7]	LIU H K, XIE X R, HE J B, et al. Subsynchronous interaction between direct-drive PMSG based wind farms and weak AC networks[J]. IEEE Transactions on Power Systems, 2017, 32(6): 4708-4720. 本文引用 [1]

[8]	冯双, 崔昊, 陈佳宁, 等. 人工智能在电力系统宽频振荡中的应用与挑战[J]. 中国电机工程学报, 2021, 41(23): 7889-7905. FENG Shuang, CUI Hao, CHEN Jianing, et al. Applications and challenges of artificial intelligence in power system wide-band oscillations[J]. Proceedings of the CSEE, 2021, 41(23): 7889-7905. 本文引用 [1]

[9]	姚力, 许灵洁, 李舜, 等. 基于数字比较法的宽频标准电流互感器溯源技术研究[J]. 电测与仪表, 2024, 61(8): 97-103. YAO Li, XU Lingjie, LI Shun, et al. Research on traceability technology of broadband standard current transformer based on digital comparison method[J]. Electrical Measurement & Instrumentation, 2024, 61(8): 97-103. 本文引用 [1]

[10]	赵东生, 周原, 范亚洲, 等. 输电线路在线监测设备供能技术研究综述[J]. 广东电力, 2024, 37(7): 107-117. ZHAO Dongsheng, ZHOU Yuan, FAN Yazhou, et al. Overview of energy supply technology for online monitoring equipment of transmission lines[J]. Guangdong Electric Power, 2024, 37(7): 107-117. 本文引用 [1]

[11]	樊陈, 姚建国, 常乃超, 等. 电网宽频振荡实时监测技术方案[J]. 电力系统自动化, 2021, 45(11): 152-159. FAN Chen, YAO Jianguo, CHANG Naichao, et al. Technical scheme for real-time monitoring of wide-frequency oscillation in power grid[J]. Automation of Electric Power Systems, 2021, 45(11): 152-159. 本文引用 [1]

[12]

刘瑞阔, 李伟, 武文, 等. 适用于新能源场站的同步测量方法与装置研制[J]. 电力建设, 2024, 45(4): 156-162.

https://doi.org/10.12204/j.issn.1000-7229.2024.04.015

摘要

现有测量装置无法实时同步监测新能源场站内部电气量,难以为新能源场站动态过程监测、惯量评估、频率控制等应用提供数据支撑。为此,提出了适用于新能源场站的同步测量算法,并研制了可安装于风机出口的新能源场站同步测量装置。针对现有相量测量算法测量频带窄,无法跟踪新能源场站次/超同步振荡等动态过程的问题,提出了基于复带通滤波器的宽频域基波相量测量算法,将基波测量频带拓宽至20~80 Hz;针对快速傅里叶变换(fast Fourier transform, FFT)的栅栏效应和频谱泄露问题,提出了基于FFT谱线插值拟合的宽频测量算法;基于上述算法,研制了适用于新能源场站的同步测量装置,并已将首批80余台装置安装于内蒙古乌兰察布吉红风光储4号电站的80余台风机。静动态测试和录波数据测试验证了所提算法和所研制装置的有效性。

LIU

Ruikuo

, LI

Wei

, WU

Wen

, et al. Synchronized measurement method and device development for renewable energy stations[J]. Electric Power Construction, 2024, 45(4): 156-162.

本文引用 [1]

[13]

程启明, 周伟成, 程尹曼, 等. 基于线性自抗扰控制的双馈风机次同步振荡抑制研究[J]. 电力建设, 2024, 45(4): 134-146.

https://doi.org/10.12204/j.issn.1000-7229.2024.04.013

摘要

双馈风力发电机(doubly-fed induction generator,DFIG)并网时的次同步振荡(sub-synchronous oscillation,SSO)问题,严重威胁系统的平稳运行。系统运行工况发生变化时,现有的传统控制策略可能效果不佳。为了解决这个问题,提出了基于线性自抗扰控制(linear active disturbance rejection control,LADRC)的SSO抑制策略,在转子侧传统双环PI控制的基础上,原有功内环的PI控制部分替换为线性ADRC,其中线性扩张状态观测器实时估计补偿次同步控制相互作用(sub-synchronous control interaction,SSCI)分量,该方法参数整定简单且鲁棒性较强。通过搭建DFIG并网模型,并与PI控制对比,验证该控制策略在稳态和暂态拥有良好的控制性能;通过与附加阻尼控制和非线性ADRC对比,得出线性ADRC可以在多工况下抑制次同步振荡,相比于现有控制,在参数整定更简单的情况下能更有效地使畸变电流收敛,进而稳定DFIG风机系统的输出功率,最后在RT-LAB实验平台进一步验证其有效性。

CHENG

Qiming

, ZHOU

Weicheng

, CHENG

Yinman

, et al. Research on suppressing sub-synchronous oscillation of doubly-fed induction generator based on linear active disturbance rejection control[J]. Electric Power Construction, 2024, 45(4): 134-146.

https://doi.org/10.12204/j.issn.1000-7229.2024.04.013

本文引用 [1] 摘要

Sub-synchronous oscillation (SSO) in a grid-connected doubly fed induction generator (DFIG) threatens the stable operation of the system. Traditional control strategies may not be effective when the operating conditions of a system change. To solve this problem, a sub-synchronous suppression strategy based on linear active disturbance rejection control (ADRC) is proposed. On the basis of the traditional dual loop proportional-integral (PI) control on the rotor side, an improved auto disturbance rejection control is used to replace the PI part of the original inner loop, in which a linear extended state observer is used to estimate the part of compensation sub-synchronous control interaction (SSCI) in real time. The proposed method involves simple parameter tuning and is strongly robust. By building a DFIG grid connection model and comparing it with PI control, it is verified that the control strategy has good control performance in both steady-state and transient states. By comparing with additional damping control and nonlinear ADRC, it is concluded that improved active disturbance rejection control can suppress sub-synchronous oscillations under multiple operating conditions, converging the distorted current more effectively compared to the simple tuning of existing control parameters, thereby stabilizing the output power of the doubly fed fan system. Finally, the effectiveness of the proposed method is verified using an RT-LAB experimental platform.

[14]

蒋奇良, 郑宗生, 史云翔, 等. 基于压缩感知与ISTA的宽频振荡扰动源分级定位方法[J]. 高电压技术, 2024, 50(8): 3725-3736.

JIANG

Qiliang

, ZHENG

Zongsheng

, SHI

Yunxiang

, et al. Hierarchical localization method of broadband oscillation disturbance source based on compressive sensing and ISTA[J]. High Voltage Engineering, 2024, 50(8): 3725-3736.

本文引用 [1]

[15]

冯双, 崔昊, 吴熙, 等. 基于对角化LDPC压缩感知和k-近邻算法的广域系统宽频振荡监测方法[J]. 电网技术, 2021, 45(8): 3025-3033.

FENG

Shuang

, CUI

Hao

, WU

, et al. Monitoring method of wide-band oscillation based on compressed sensing of diagonalizable LDPC and k-nearest neighbour algorithm[J]. Power System Technology, 2021, 45(8): 3025-3033.

本文引用 [1]

[16]

崔昊, 冯双, 陈佳宁, 等. 基于自编码器与长短期记忆网络的宽频振荡广域定位方法[J]. 电力系统自动化, 2022, 46(12): 194-201.

CUI

Hao

, FENG

Shuang

, CHEN

Jianing

, et al. Wide-area location method of wide-band oscillations based on autoencoder and long short-term memory network[J]. Automation of Electric Power Systems, 2022, 46(12): 194-201.

本文引用 [2]

[17]	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[J/OL]. arXiv, 2017.[2024-07-11].https://doi.org/10.48550/arXiv.1706.03762 https://doi.org/10.48550/arXiv.1706.03762 本文引用 [1]

[18]

黄悦华, 张子豪, 陈庆, 等. 基于多头注意力机制的CNN-BiLSTM高海拔多因素输电线路可听噪声预测[J]. 高压电器, 2024, 60(12): 160-169.

HUANG

Yuehua

, ZHANG

Zihao

, CHEN

Qing

, et al. Prediction of audible noise of CNN-BiLSTM high altitude multi-factor transmission line based on multi-head attention mechanism[J]. High Voltage Apparatus, 2024, 60(12): 160-169.

本文引用 [1]

[19]	黄碧月. 直驱风电并网的电力系统动态特性分析与次同步振荡研究[D]. 武汉: 华中科技大学, 2021. HUANG Biyue. Dynamic characteristics analysis and subsynchronous oscillation research of power system with direct-drive wind power connected to grid[D]. Wuhan: Huazhong University of Science and Technology, 2021. 本文引用 [1]

[20]

高本锋, 王义, 曾四鸣, 等. 直驱风电场并入弱交流电网的次同步分量通路及阻尼特性分析[J]. 中国电机工程学报, 2022, 42(14): 5089-5103.

GAO

Benfeng

, WANG

, ZENG

Siming

, et al. Analysis of sub-synchronous component path and damping characteristics of D-PMSG-based wind farm incorporated into weak AC grid[J]. Proceedings of the CSEE, 2022, 42(14): 5089-5103.

本文引用 [1]

[21]

马宁宁, 谢小荣, 亢朋朋, 等. 高比例风电并网系统次同步振荡的广域监测与分析[J]. 中国电机工程学报, 2021, 41(1): 65-74, 398.

Ningning

, XIE

Xiaorong

, KANG

Pengpeng

, et al. Wide-area monitoring and analysis of subsynchronous oscillation in power systems with high-penetration of wind power[J]. Proceedings of the CSEE, 2021, 41(1): 65-74, 398.

本文引用 [1]

[22]

吴琛, 刘威, 张丹, 等. 基于宽频振荡稳定约束的风电接入容量分析[J]. 电力建设, 2023, 44(2): 83-91.

https://doi.org/10.12204/j.issn.1000-7229.2023.02.008

摘要

风电的大规模接入可能引发严重的宽频振荡。宽频振荡与风电场的容量、接入点阻抗以及机组的运行工况等密切相关。文章采用阻抗模型方法分析风电并网系统的宽频振荡特性,明确宽频振荡约束下的风电接入容量与电网阻抗之间的关系。首先,建立了风电机组的全工况阻抗模型,该模型以风电机组端口工频电压和输出电流为变量;然后,基于全工况阻抗模型分析了风电机组输出功率、接入点短路比等对风电并网系统宽频振荡的影响;进而,分析了风电场容量与接入点之间的关系,为风电场的建设和运行提供参考;最后,通过时域仿真验证了全工况阻抗模型分析结果的准确性。结果表明,基于全工况阻抗模型可以确定在不同电网条件下考虑宽频振荡稳定时风电场的最大接入容量。

Chen

, LIU

Wei

, ZHANG

Dan

, et al. Wind power penetration capacity analysis based on wide-band oscillation constraint[J]. Electric Power Construction, 2023, 44(2): 83-91.

https://doi.org/10.12204/j.issn.1000-7229.2023.02.008

本文引用 [1] 摘要

The large-scale penetration of wind power may cause serious wide-band oscillations. Wide-band oscillation is closely related to the capacity of the wind farm, the interface impedance, and the operating conditions of wind turbines generators (WTGs). In this paper, the impedance model method is adopted to analyze the wide-band oscillatory characteristics of the wind power system, so as to clarify the relationship between the penetration capacity of the wind farm and the grid impedance under the constraint of wide-band oscillation. Firstly, the full-operating-condition impedance model (FOCIM) of the WTG is established, which takes the voltage and output current at the terminal as variables. Then, on the basis of the FOCIM, the effects of output power and short-circuit ratio (SCR) on the wide-band oscillation are analyzed. Furthermore, how the practicable maximum capacity of the wind farm varies with the grid impedance is studied, which is of reference value for the construction and operation of the wind farm. Finally, the accuracy of the analysis results of the FOCIM is verified by time-domain simulations. The results show that the practicable maximum capacity of the wind farm considering the constraint of wide-band oscillation can be determined on the basis of the FOCIM under different grid conditions.

[23]

袁慧涛, 袁文涛, 王文敬, 等. 基于阻抗分析法的双馈风电机组经串补线路并网系统的次同步振荡特性研究[J]. 山东电力技术, 2024, 51(11): 48-60.

YUAN

Huitao

, YUAN

Wentao

, WANG

Wenjing

, et al. Research on subsynchronous oscillation of doubly-fed wind turbines connected by series compensating lines based on impedance analysis method[J]. Shandong Electric Power, 2024, 51(11): 48-60.

本文引用 [1]

[24]

王岑峰, 王蕾, 孙飞飞, 等. 基于模糊逻辑控制的混合储能辅助风电调频的双层优化配置模型[J]. 高压电器, 2024, 60(10): 54-63.

WANG

Cenfeng

, WANG

Lei

, SUN

Feifei

, et al. Two-layer optimal configuration model of hybrid energy storage assisted wind power frequency regulation based on fuzzy logic control[J]. High Voltage Apparatus, 2024, 60(10): 54-63.

本文引用 [1]

[25]	冯双, 陈佳宁, 汤奕, 等. 基于SPWVD图像和深度迁移学习的强迫振荡源定位方法[J]. 电力系统自动化, 2020, 44(17): 78-87. FENG Shuang, CHEN Jianing, TANG Yi, et al. Location method of forced oscillation source based on SPWVD image and deep transfer learning[J]. Automation of Electric Power Systems, 2020, 44(17): 78-87. 本文引用 [1]

[26]

陈剑, 杜文娟, 王海风. 采用深度迁移学习定位含直驱风机次同步振荡源机组的方法[J]. 电工技术学报, 2021, 36(1): 179-190.

CHEN

Jian

, DU

Wenjuan

, WANG

Haifeng

. A method of locating the power system subsynchronous oscillation source unit with grid-connected PMSG using deep transfer learning[J]. Transactions of China Electrotechnical Society, 2021, 36(1): 179-190.

本文引用 [1]

[27]

陈剑, 杜文娟, 王海风. 基于对抗式迁移学习的含柔性高压直流输电的风电系统次同步振荡源定位[J]. 电工技术学报, 2021, 36(22): 4703-4715.

CHEN

Jian

, DU

Wenjuan

, WANG

Haifeng

. Location method of subsynchronous oscillation source in wind power system with VSC-HVDC based on adversarial transfer learning[J]. Transactions of China Electrotechnical Society, 2021, 36(22): 4703-4715.

本文引用 [1]

[28]	VELIKOVI P, CUCURULL G, CASANOVA A, et al. Graph attention networks[J/OL]. arXiv, 2017.[2024-07-11]. https://doi.org/10.48550/arXiv.1710.10903. https://doi.org/10.48550/arXiv.1710.10903. 本文引用 [1]

[29]

王渝红, 李晨鑫, 周旭, 等. 压缩感知和图卷积神经网络相结合的宽频振荡扰动源定位方法[J]. 高电压技术, 2024, 50(3): 1080-1089.

WANG

Yuhong

, LI

Chenxin

, ZHOU

, et al. Localization method of wide-band oscillation disturbance sources based on compressed sensing and graph convolutional neural networks[J]. High Voltage Engineering, 2024, 50(3): 1080-1089.

本文引用 [1]

[30]	朱壮华, 张卫平, 王文彦, 等. 基于图神经网络的光伏组串故障诊断[J]. 分布式能源, 2024, 9(4): 78-85. ZHU Zhuanghua, ZHANG Weiping, WANG Wenyan, et al. Fault diagnosis of photovoltaic string based on graph neural network[J]. Distributed Energy, 2024, 9(4): 78-85. 本文引用 [1]

[31]	李佳玮, 王小君, 和敬涵, 等. 基于图注意力网络的配电网故障定位方法[J]. 电网技术, 2021, 45(6): 2113-2121. LI Jiawei, WANG Xiaojun, HE Jinghan, et al. Distribution network fault location based on graph attention network[J]. Power System Technology, 2021, 45(6): 2113-2121. 本文引用 [1]

[32]	徐少博, 徐永海, 陶顺, 等. 谐波扰动下PWM边带谐波特性分析[J]. 电测与仪表, 2024, 61(9): 172-181. XU Shaobo, XU Yonghai, TAO Shun, et al. Analysis of PWM sideband harmonic characteristics with harmonic disturbance[J]. Electrical Measurement & Instrumentation, 2024, 61(9): 172-181. 本文引用 [1]

[33]	吴熙, 关雅静, 宁威, 等. 双馈风机转子侧变换器参数对次同步振荡的交互影响机理及其应用研究[J]. 电网技术, 2018, 42(8): 2536-2544. WU Xi, GUAN Yajing, NING Wei, et al. Mechanism of interactive effect of RSC parameters in DFIG on SSO and its application[J]. Power System Technology, 2018, 42(8): 2536-2544. 本文引用 [1]

[34]	吴楠, 李尚盛, 查晓明. 改进的同步电机阻尼绕组电流观测器[J]. 电力系统保护与控制, 2015, 43(1): 67-72. WU Nan, LI Shangsheng, ZHA Xiaoming. Improved synchronous generator damper current observer[J]. Power System Protection and Control, 2015, 43(1): 67-72. 本文引用 [1]

[35]	BANNA H U, SOLANKI S K, SOLANKI J. Data-driven disturbance source identification for power system oscillations using credibility search ensemble learning[J]. IET Smart Grid, 2019, 2(2): 293-300. 本文引用 [1]

[36]

李景一, 毕天姝, 于钊, 等. 直驱风机变流控制系统对次同步频率分量的响应机理研究[J]. 电网技术, 2017, 41(6): 1734-1740.

Jingyi

, BI

Tianshu

, YU

Zhao

, et al. Study on response characteristics of grid converter control system of permanent magnet synchronous generators(PMSG) to subsynchronous frequency component[J]. Power System Technology, 2017, 41(6): 1734-1740.

本文引用 [1]

[37]	LU B L, LIU Z H, WEI H L, et al. A deep adversarial learning prognostics model for remaining useful life prediction of rolling bearing[J]. IEEE Transactions on Artificial Intelligence, 2021, 2(4): 329-340. 本文引用 [1]