结合TTAO-VMD和CNN-SE-Attention-ITCN的电力系统次同步振荡参数辨识

马燕峰; 王树颜; 王子建; 赵书强; 徐伟阔; 韩珊珊

doi:10.19912/j.0254-0096.tynxb.2025-0408

PDF(1485 KB)

太阳能学报 ›› 2026, Vol. 47 ›› Issue (7) : 726-735. DOI: 10.19912/j.0254-0096.tynxb.2025-0408

结合TTAO-VMD和CNN-SE-Attention-ITCN的电力系统次同步振荡参数辨识

马燕峰, 王树颜, 王子建, 赵书强, 徐伟阔, 韩珊珊

作者信息 +

IDENTIFICATION OF SUB-SYNCHRONOUS OSCILLATION PARAMETERS OF POWER SYSTEM BASED ON TTAO-VMD AND CNN-SE-ATTENTION-ITCN

Ma Yanfeng, Wang Shuyan, Wang Zijian, Zhao Shuqiang, Xu Weikuo, Han Shanshan

Author information +

文章历史 +

摘要

针对传统神经网络在辨识含噪次同步振荡（SSO）时存在的泛化性差及时序特征提取能力不足的问题,提出一种引入SE-Attention机制的卷积神经网络（CNN）和改进时间卷积网络（ITCN）结合的CNN-SE-Attention-ITCN混合辨识模型。首先,采用三角拓扑聚合优化算法（TTAO）优化变分模态分解（VMD）的模态数K和惩罚因子a,分解得到SSO的若干模态,然后根据皮尔逊相关系数（PCC）筛选有效模态分量进行重构并提取时域特征;最后采用训练好的CNN-SE-Attention-ITCN模型辨识SSO参数。通过含噪理想信号、仿真信号和实际信号的测试,验证该模型具有较高的辨识准确率和良好的鲁棒性。

Abstract

Aiming at the problems of poor generalization, and insufficient temporal feature extraction capability existing in the identification of noisy sub-synchronous oscillations （SSO） by traditional neural networks, a CNN-SE-Attention-ITCN hybrid identification model combining convolutional neural network（CNN）with self-attention mechanism and improved temporal convolutional network（ITCN）is proposed. Firstly, the mode number K and penalty factor α of variational mode decomposition（VMD）were optimized using TTAO algorithm, and several modes of SSO were decomposed. Then, effective modal components were selected according to Pearson correlation coefficient（PCC）, and time domain features were extracted. Finally, the trained CNN-SE-Attention-ITCN model was used to identify SSO parameters. The model is proved to have good identification accuracy and robustness through the testing of noisy ideal signal, simulated signal and real signal.

导出引用

马燕峰, 王树颜, 王子建, 赵书强, 徐伟阔, 韩珊珊. 结合TTAO-VMD和CNN-SE-Attention-ITCN的电力系统次同步振荡参数辨识[J]. 太阳能学报. 2026, 47(7): 726-735 https://doi.org/10.19912/j.0254-0096.tynxb.2025-0408

Ma Yanfeng, Wang Shuyan, Wang Zijian, Zhao Shuqiang, Xu Weikuo, Han Shanshan. IDENTIFICATION OF SUB-SYNCHRONOUS OSCILLATION PARAMETERS OF POWER SYSTEM BASED ON TTAO-VMD AND CNN-SE-ATTENTION-ITCN[J]. Acta Energiae Solaris Sinica. 2026, 47(7): 726-735 https://doi.org/10.19912/j.0254-0096.tynxb.2025-0408

中图分类号： TK712

参考文献

[1] 马宁宁, 谢小荣, 贺静波, 等. 高比例新能源和电力电子设备电力系统的宽频振荡研究综述[J]. 中国电机工程学报, 2020, 40(15): 4720-4731.
Ma N N, Xie X R, He J B, et al.Review of wide-band oscillation in renewable and power electronics highly integrated power systems[J]. Proceedings of the CSEE, 2020, 40(15): 4720-4731.
[2] 肖湘宁, 罗超, 廖坤玉. 新能源电力系统次同步振荡问题研究综述[J]. 电工技术学报, 2017, 32(6): 85-97.
Xiao X N, Luo C, Liao K Y.Review of the research on subsynchronous oscillation issues in electric power system with renewable energy sources[J]. Transactions of China Electrotechnical Society, 2017, 32(6): 85-97.
[3] 吴熙, 陈曦, 吕万, 等. 电力系统次同步振荡检测与在线定位技术综述[J]. 电力自动化设备, 2020, 40(9): 129-141.
Wu X, Chen X, Lyu W, et al.Review of detection and online localization technology of sub-synchronous oscillation in power system[J]. Electric Power Automation Equipment, 2020, 40(9): 129-141.
[4] 况且, 李娟, 白淑华. 基于深度残差网络的SSO模态参数辨识[J]. 电子测量技术, 2022, 45(11): 57-63.
Kuang Q, Li J, Bai S H.Modal parameter identification of SSO based on deep residual network[J]. Electronic Measurement Technology, 2022, 45(11): 57-63.
[5] 邱衍江, 张新燕, 张超. 风电汇集地区次/超同步谐波分析方法研究[J]. 太阳能学报, 2021, 42(1): 286-293.
Qiu Y J, Zhang X Y, Zhang C.Study on sub/super-synchronous harmonic detection method in dense wind power areas[J]. Acta Energiae Solaris Sinica, 2021, 42(1): 286-293.
[6] 杨京, 王彤, 唐俊刺. 基于滑窗FFT的次同步振荡时变幅频在线监测方法[J]. 中国电力, 2020, 53(11): 139-146.
Yang J, Wang T, Tang J C.Subsynchronous oscillation time-varying amplitude frequency on-line monitoring method based on sliding window FFT[J]. Electric Power, 2020, 53(11): 139-146.
[7] Chen L, Zhao W, Wang F P, et al.An interharmonic phasor and frequency estimator for subsynchronous oscillation identification and monitoring[J]. IEEE Transactions on Instrumentation and Measurement, 2019, 68(6): 1714-1723.
[8] Li K, Zhang W J, Wang H, et al.Parameter identification of subsynchronous oscillation based on FastICA and MP algorithms[J]. The Journal of Engineering, 2019, 2019(16): 2454-2457.
[9] 刘克天, 胡新宇, 李想, 等. 基于EMD和TLS-Prony的次同步振荡模态辨识方法及电压反馈抑制策略[J]. 太阳能学报, 2023, 44(11): 294-302.
Liu K T, Hu X Y, Li X, et al.Sub-synchronous oscillation modal identification method based on EMD and TLS-Prony algorithm and voltage feedback suppression strategy[J]. Acta Energiae Solaris Sinica, 2023, 44(11): 294-302.
[10] Mohanty, Gupta K K, Raju K S. Bearing fault analysis using variational mode decomposition[C]//2014 9th International Conference on Industrial and Information Systems (ICIIS). Gwalior, India, 2015: 1-6.
[11] 王杨, 晁苗苗, 谢小荣, 等. 基于同步相量数据的次同步振荡参数辨识与实测验证[J]. 中国电机工程学报, 2022, 42(3): 899-908.
Wang Y, Chao M M, Xie X R, et al.Identification of subsynchronous oscillation parameters and field tests based on PMU data[J]. Proceedings of the Chinese Society for Electrical Engineering, 2022, 42(3): 899-908.
[12] 李文博, 钱伟荣, 李淑蓉, 等. 改进自适应VMD和TLS-ESPRIT的风电系统次/超同步振荡参数辨识[J]. 高电压技术, 2025, 51(1): 146-157, I0013-I0017.
Li W B, Qian W R, Li S R, et al. Improved adaptive VMD and TLS-ESPRIT sub/sup-synchronous oscillation parameter identification for wind power systems[J]. High Voltage Engineering, 2025, 51(1): 146-157, I0013-I0017.
[13] 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): 7889-7904.
Feng S, Cui H, Chen J N, et al.Applications and challenges of artificial intelligence in power system wide-band oscillations[J]. Proceedings of the CSEE, 2021, 41(23): 7889-7904.
[14] 张文康. 基于宽频测量的次同步振荡检测方法研究[D]. 北京: 华北电力大学, 2023.
Zhang W K.Research on sub-synchronous oscillation detection method based on broadband measurement[D]. Beijing: North China Electric Power University, 2023.
[15] 秦心筱, 张昌华, 徐子豪, 等. 基于卷积神经网络的电力系统低频振荡主导模态特征定性辨识[J]. 电力系统保护与控制, 2021, 49(10): 51-58.
Qin X X, Zhang C H, Xu Z H, et al.Research on qualitative identification of a low frequency oscillations dominant mode in power system based on a convolutional neural network[J]. Power System Protection and Control, 2021, 49(10): 51-58.
[16] 陆友文, 崔昊, 陈佳宁, 等. 基于RA-CNN和同步相量的风电场次/超同步振荡参数智能辨识方法[J]. 中国电力, 2023, 56(4): 46-55, 67.
Lu Y W, Cui H, Chen J N, et al.Intelligent identification method of wind farm sub-synchronous/super-synchronous oscillation parameters based on RA-CNN and synchrophasor[J]. Electric Power, 2023, 56(4): 46-55, 67.
[17] 李青. 基于注意力时间卷积神经网络的光伏功率概率预测[J]. 太阳能学报, 2025, 46(2): 326-332.
Li Q.Photovoltaic power probability prediction based on attention time convolutional neural network[J]. Acta Energiae Solaris Sinica, 2025, 46(2): 326-332.
[18] 金秀章, 赵术善, 畅晗, 等. 基于金豺优化变分模态分解与时间卷积网络的过热汽温特性建模[J]. 中国电机工程学报, 2025, 45(12): 4759-4767
Jin X Z, Zhao S S, Chang H, et al.Superheated steam temperature characteristics modeling based on golden jackal optimization variational mode decomposition and temporal convolutional network[J]. Proceedings of the Chinese Society for Electrical Engineering, 2025, 45(12): 4759-4767.
[19] Zhao S J, Zhang T R, Cai L, et al.Triangulation topology aggregation optimizer: a novel mathematics-based meta-heuristic algorithm for continuous optimization and engineering applications[J]. Expert Systems with Applications, 2024, 238: 121744.
[20] Hu J, Shen L, Sun G.Squeeze-and-excitation networks[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA, 2018: 7132-7141.
[21] 秦浩. 基于改进时间卷积网络的短期电力负荷预测研究[D]. 南昌: 南昌大学, 2023.
Qin H.Research on short-term load forecasting based on improved temporal convolutional network[D]. Nanchang: Nanchang University, 2023.
[22] Liu H K, Xie X R, Zhang C Y, et al.Quantitative SSR analysis of series-compensated DFIG-based wind farms using aggregated RLC circuit model[J]. IEEE Transactions on Power Systems, 2017, 32(1): 474-483.