TRANSIENT VOLTAGE STABILITY ASSESSMENT CONSIDERING SAMPLE IMBALANCE AND SPATIO-TEMPORAL FEATURE FUSION

He Jiacheng; Zhang Xuexia

doi:10.19912/j.0254-0096.tynxb.2025-1136

PDF(2313 KB)

Acta Energiae Solaris Sinica ›› 2026, Vol. 47 ›› Issue (12) : 299-309. DOI: 10.19912/j.0254-0096.tynxb.2025-1136

TRANSIENT VOLTAGE STABILITY ASSESSMENT CONSIDERING SAMPLE IMBALANCE AND SPATIO-TEMPORAL FEATURE FUSION

He Jiacheng, Zhang Xuexia

Author information +

History +

Abstract

To improve the performance of transient voltage stability assessment models for power systems, this paper proposes a spatiotemporal feature modeling approach that integrates an improved graph attention network （GAT） with a Transformer. Considering the inherent sample imbalance that may lead to missed detection in assessment, a Wasserstein generative adversarial network with gradient penalty （WGAN-GP） is further introduced for data augmentation. The proposed model extracts the temporal evolutionary features of transient voltage via the Transformer module, and incorporates an impedance-aware edge prior into the graph attention branch, thereby enhancing the spatial representation's ability to capture the physical coupling strength of the power grid. Meanwhile, the WGAN-GP network generates high-quality unstable samples to alleviate the class imbalance issue. Simulation results on the modified New England 39-bus system demonstrate that the proposed model effectively reduces the missed detection of instability and improves assessment accuracy. Furthermore, the model exhibits strong robustness even under noise interference and extreme class imbalance conditions.

Key words

deep learning / Transformer / generative adversarial networks / transient voltage stability assessment / graph attention network

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

He Jiacheng, Zhang Xuexia. TRANSIENT VOLTAGE STABILITY ASSESSMENT CONSIDERING SAMPLE IMBALANCE AND SPATIO-TEMPORAL FEATURE FUSION[J]. Acta Energiae Solaris Sinica. 2026, 47(12): 299-309 https://doi.org/10.19912/j.0254-0096.tynxb.2025-1136

References

[1] 李汐, 高阳, 马斌, 等. 基于PEF-PinSVM的含可再生能源并网电力系统电压暂态稳定裕度评估[J]. 太阳能学报, 2024, 45(8): 240-248.
Li X, Gao Y, Ma B, et al.Voltage transient stability margin assessment of grid connected power systems containing renewable energy based on PEF-PinSVM[J]. Acta Energiae Solaris Sinica, 2024, 45(8): 240-248.
[2] 杨银国, 林旭, 吴国炳, 等. 构网型、跟网型非同步机电源并网系统暂态同步稳定性分析[J]. 太阳能学报, 2024, 45(8): 218-228.
Yang Y G, Lin X, Wu G B, et al.Analysis of transient synchroniiation stability of power system with grid-forming/grid-following nonsynchronous-machine sources[J]. Acta Energiae Solaris Sinica, 2024, 45(8): 218-228.
[3] 李永康, 刘宝柱, 胡俊杰. 基于数据驱动与时域仿真融合的电力系统暂态稳定快速评估[J]. 电网技术, 2023, 47(11): 4386-4395.
Li Y K, Liu B Z, Hu J J.Rapid evaluation of power system transient stability based on fusion of data-driven and time-domain simulation[J]. Power System Technology, 2023, 47(11): 4386-4395.
[4] 王长江, 姜涛, 陈厚合, 等. 基于相位校正李雅普诺夫指数的电力系统暂态电压稳定评估[J]. 电工技术学报, 2021, 36(15): 3221-3236.
Wang C J, Jiang T, Chen H H, et al.Transient voltage stability assessment of power systems based on phase correction maximum Lyapunov exponent[J]. Transactions of China Electrotechnical Society, 2021, 36(15): 3221-3236.
[5] 符益华, 卢国强, 王怀远. 基于定向对抗迁移的暂态稳定评估模型[J]. 太阳能学报, 2025, 46(2): 226-234.
Fu Y H, Lu G Q, Wang H Y.Transient stability assessment model based on directional adversarial transfer learning[J]. Acta Energiae Solaris Sinica, 2025, 46(2): 226-234.
[6] 高昆仑, 杨帅, 刘思言, 等. 基于一维卷积神经网络的电力系统暂态稳定评估[J]. 电力系统自动化, 2019, 43(12): 18-26.
Gao K L, Yang S, Liu S Y, et al.Transient stability assessment for power system based on one-dimensional convolutional neural network[J]. Automation of Electric Power Systems, 2019, 43(12): 18-26.
[7] Shi Z T, Yao W, Zeng L K, et al.Convolutional neural network-based power system transient stability assessment and instability mode prediction[J]. Applied Energy, 2020, 263: 114586.
[8] 李楠, 朱嫄, 崔莹. 考虑代价敏感的AC-LSTM暂态稳定评估[J]. 电力系统保护与控制, 2022, 50(22): 160-169.
Li N, Zhu Y, Cui Y.AC-LSTM transient stability assessment considering cost-sensitivity[J]. Power System Protection and Control, 2022, 50(22): 160-169.
[9] Li Y, Zhang M, Chen C.A Deep-Learning intelligent system incorporating data augmentation for short-term voltage stability assessment of power systems[J]. Applied Energy, 2022, 308: 118347.
[10] 姜涛, 董雨, 王长江, 等. 基于图卷积和双向长短期记忆网络的受端电力系统暂态电压稳定评估[J]. 电网技术, 2023, 47(12): 4937-4947.
Jiang T, Dong Y, Wang C J, et al.Transient voltage stability assessment of receiving-end power system based on graph convolution and bidirectional long/short-term memory networks[J]. Power System Technology, 2023, 47(12): 4937-4947.
[11] 王铮澄, 周艳真, 郭庆来, 等. 考虑电力系统拓扑变化的消息传递图神经网络暂态稳定评估[J]. 中国电机工程学报, 2021, 41(7): 2341-2349.
Wang Z C, Zhou Y Z, Guo Q L, et al.Transient stability assessment of power system considering topological change: a message passing neural network-based approach[J]. Proceedings of the CSEE, 2021, 41(7): 2341-2349.
[12] Shi F S, Wu J Y, Wang Y, et al.Integrated advance assessment of power system transient voltage and transient angle stability based on two-stage ensemble spatio-temporal graph neural network[J]. Measurement, 2023, 221: 113447.
[13] Zhang D, Yang Y, Shen B, et al.Transient stability assessment in power systems: a spatiotemporal graph convolutional network approach with graph simplification[J]. Energies, 2024, 17(20): 5095.
[14] 卢锦玲, 郭鲁豫. 基于改进深度残差收缩网络的电力系统暂态稳定评估[J]. 电工技术学报, 2021, 36(11): 2233-2244.
Lu J L, Guo L Y.Power system transient stability assessment based on improved deep residual shrinkage network[J]. Transactions of China Electrotechnical Society, 2021, 36(11): 2233-2244.
[15] 胡力涛, 王怀远, 党然, 等. 基于梯度范数的暂态稳定评估模型的不平衡修正方法[J]. 电力自动化设备, 2024(4): 158-163, 177.
Hu L T, Wang H Y, Dang R, et al.Imbalance correction method of transient stability assessment model based on gradient norm[J]. Electric Power Automation Equipment, 2024(4): 158-163, 177.
[16] 时纯, 刘君, 梁卓航, 等. 基于GAN和多通道CNN的电力系统暂态稳定评估[J]. 电网技术, 2022, 46(8): 3191-3202.
Shi C, Liu J, Liang Z H, et al.Transient stability assessment of power system based on GAN and multi-channel CNN[J]. Power System Technology, 2022, 46(8): 3191-3202.
[17] Sun Z Y, Li M P, Zhang J, et al.Transient voltage stability assessment method based on GCForest[J]. Journal of Physics (Conference Series), 2021(1): 7.
[18] Goodfellow I J, Pouget-Abadie J, Mirza M, et al.Generative adversarial nets[J]Advances in Neural Information Processing Systems. 2014, 27: 2672-2680.
[19] Arjovsky M, Chintala S, Bottou L.Wasserstein generative adversarial networks[C]//Proceedings of the 34th International Conference on Machine Learning. Pmlr, 2017: 214-223.
[20] DL/T 1234—2013 电力系统安全稳定计算技术规范[S]. 北京: 中国电力科学研究院, 2021.
[21] Gretton A, Borgwardt K M, Rasch M, et al.A kernel method for the two-sample-problem[M]//Advances in Neural Information Processing Systems 19. Cambridge, MA, USA: The MIT Press, 2007: 513-520.
[22] Tran N T, Tran V H, Nguyen N B, et al.On data augmentation for GAN training[J]. IEEE Transactions on Image Processing, 2021, 30: 1882-1897.
[23] Martin K E, Hamai D, Adamiak M G, et al.Exploring the IEEE standard C37.118-2005 synchrophasors for power systems[J]. IEEE Transactions on Power Delivery, 2008, 23(4): 1805-1811