| 404 | 0 | 104 |
| 下载次数 | 被引频次 | 阅读次数 |
为提高暂态电压稳定评估模型性能并增强评估结果的可解释性，提出一种将时序卷积与多头自注意力机制相结合的方法。首先，通过嵌入时序卷积模块改进Transformer编码器，捕获暂态过程电气参数间的全局和局部信息，以准确评估电力系统暂态电压稳定状态。然后，提出自适应阈值焦点损失函数，有效缓解样本不平衡对模型训练的影响。其次，采用基于多头自注意力机制的可解释分析方法，在时间与空间维度上基于注意力权重的计算，辅助分析评估模型决策过程。最后，通过IEEE-39和IEEE-300节点系统进行仿真验证，结果表明所提方法具有可解释性、更高的评估精度及较强的鲁棒性。
Abstract：To enhance the performance of transient voltage stability assessment models and improve the interpretability of assessment results,a method integrating temporal convolutional and a multi-head self-attention mechanism is proposed in this paper. First,to accurately assess the power system's transient voltage stability,the Transformer encoder is modified by embedding a temporal convolutional module,thus capturing the global and local information about electrical parameters throughout the transient process. Second,an adaptive threshold focal loss function is put forward to mitigate the adverse impact of sample imbalance on model training. Third,an interpretability analysis method based on the multi-head self-attention mechanism is employed to calculate attention weights across both the temporal and spatial dimensions,providing insights into the decision-making process of the assessment model. Finally,the proposed method is verified by simulations conducted on an IEEE-39 bus system and an IEEE-300 bus system,and results demonstrate that it achieves higher assessment accuracy,exhibits strong robustness,and offers interpretability.
[1]高炜伦，武志刚，苗璐，等.暂态电压稳定薄弱区域辨识的Louvain分区算法[J].电力系统及其自动化学报，2024,36(10):99-107.Gao Weilun,Wu Zhigang,Miao Lu,et al. Louvain partition algorithm for identifying weak region of transient voltage stability[J]. Proceedings of the CSU-EPSA,2024,36(10):99-107.
[2]Zhu Lipeng,Wen Weijia,Li Jiayong,et al. Structureaware recurrent learning machine for short-term voltage trajectory sensitivity prediction[J]. IEEE Internet of Things Journal,2024,11(9):15128-15139.
[3]汤涌，仲悟之，孙华东，等.电力系统电压稳定机理研究[J].电网技术，2010,34(4):24-29.Tang Yong,Zhong Wuzhi,Sun Huadong,et al. Study on mechanism of power system voltage stability[J]. Power System Technology,2010,34(4):24-29.
[4]张文朝，张博，潘捷，等.基于感应电动机网荷互馈特性的暂态电压失稳机理探析[J].电力系统自动化，2017,41(7):8-14.Zhang Wenchao,Zhang Bo,Pan Jie,et al. Mechanism analysis for transient voltage instability based on crossfeed characteristics between power network and load of induction motor[J]. Automation of Electric Power Systems,2017,41(7):8-14.
[5]Kamwa I,Samantaray S R,Joos G. Compliance analysis of PMU algorithms and devices for wide-area stabilizing control of large power systems[J]. IEEE Transactions on Power Systems,2013,28(2):1766-1778.
[6]Vanfretti L,Arava V S N. Decision tree-based classification of multiple operating conditions for power system voltage stability assessment[J]. International Journal of Electrical Power and Energy Systems,2020,123:106251.
[7]Nguyen Duc H,Kamwa I,Dessaint L A,et al. A novel approach for early detection of impending voltage collapse events based on the support vector machine[J]. International Transactions on Electrical Energy Systems,2017,27(9):e2375.
[8]Hagmar H,Tong Lang,Eriksson R,et al. Voltage instability prediction using a deep recurrent neural network[J].IEEE Transactions on Power Systems,2021,36(1):17-27.
[9]杨维全，朱元振，刘玉田.基于卷积神经网络的暂态电压稳定快速评估[J].电力系统自动化，2019,43(22):46-51,136.Yang Weiquan,Zhu Yuanzhen,Liu Yutian. Fast assessment of transient voltage stability based on convolutional neural network[J]. Automation of Electric Power Systems,2019,43(22):46-51,136.
[10]Cui Mingjian,Li Fangxing,Cui Hantao,et al. Data-driven joint voltage stability assessment considering load uncertainty:a variational Bayes inference integrated with multiCNNs[J]. IEEE Transactions on Power Systems,2022,37(3):1904-1915.
[11]张哲，秦博宇，高鑫，等.融合注意力机制和卷积神经网络的电网暂态电压稳定评估及可解释性分析[J].电网技术，2024,48(11):4648-4657.Zhang Zhe,Qin Boyu,Gao Xin,et al. Transient voltage stability assessment and interpretability analysis of power grids based on the fusion of attention mechanism and convolutional neural network[J]. Power System Technology,2024,48(11):4648-4657.
[12]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 Tao,Dong Yu,Wang Changjiang,et al. Transient voltage stability assessment of receiving-end power system based on graph convolution and bidirectional long/shortterm memory networks[J]. Power System Technology,2023,47(12):4937-4947.
[13]Chen Liukai,Guan Lin. Static information,K-neighbor,and self-attention aggregated scheme:a transient stability prediction model with enhanced interpretability[J]. Protection and Control of Modern Power Systems,2023,8(1):5.
[14]Zhao Yinghao,Yang Xu,Huang Jian,et al. Improved weighted extreme learning machine with adaptive costsensitive strategy for imbalanced fault diagnosis of rotating machinery[J]. Mechanical Systems and Signal Processing,2024,217:111526.
[15]Vaswani A,Shazeer N,Parmar N,et al. Attention is all you need[C]//31st International conference on neural information Processing Systems. Long Beach,USA,2017:5999-6009.
[16]陆旦宏，范文尧，杨婷，等.基于生成对抗Transformer的电力负荷数据异常检测[J].电力工程技术，2024,43(1):157-164.Lu Danhong,Fan Wenyao,Yang Ting,et al. Anomaly detection of power load data based on transformer and generative adversarial networks[J]. Electric Power Engineering Technology,2024,43(1):157-164.
[17]纪鹏志，李光肖，王琳，等.基于Transformer深度学习网络的主动配电网多元源荷灾损辨识方法[J].电力建设，2023,44(3):56-65.Ji Pengzhi,Li Guangxiao,Wang Lin,et al. Transformerbased evaluation method of power outage in active distribution networks with multiple sources and loads[J]. Electric Power Construction,2023,44(3):56-65.
[18]Khalil I U,Ul Haq A,ul Islam N. A deep learning-based transformer model for photovoltaic fault forecasting and classification[J]. Electric Power Systems Research,2024,228:110063.——
[19]Lin Tsung-Yi,Goyal P,Girshick R,et al. Focal loss for dense object detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence ,2020,42(2):318-327.
[20]朱利鹏，陆超，孙元章，等.基于数据挖掘的区域暂态电压稳定评估[J].电网技术，2015,39(4):1026-1032.Zhu Lipeng,Lu Chao,Sun Yuanzhang,et al. Data mining based regional transient voltage stability assessment[J].Power System Technology,2015,39(4):1026-1032.
[21]史法顺，吴俊勇，季佳伸，等.基于深度学习的电力系统暂态电压与暂态功角稳定-体化超前评估[J].电网技术，2023,47(2):741-754.Shi Fashun,Wu Junyong,Ji Jiashen,et al. Integrated advance assessment of power system transient voltage and transient angle stability based on deep learning[J]. Power System Technology,2023,47(2):741-754.
[22]谢桦，陈俊星，赵宇明，等.基于SMOTE和决策树算法的电力变压器状态评估知识获取方法[J].电力自动化设备，2020,40(2):137-142.Xie Hua,Chen Junxing,Zhao Yuming,et al. Knowledge acquisition method of power transformer condition assessment based on SMOTE and decision tree algorithm[J].Electric Power Automation Equipment,2020,40(2):137-142.
[23]Saner C B,Kesici M,Yaslan Y,et al. Improving the performance of transient stability prediction using resampling methods[C]//11th International Conference on Electrical and Electronics Engineering. Bursa,Turkey,2019.
[24]Ng W W Y,Zhang Jianjun,Lai Chun Sing,et al. Cost-sensitive weighting and imbalance-reversed bagging for streaming imbalanced and concept drifting in electricity pricing classification[J]. IEEE Transactions on Industrial Informatics,2019,15(3):1588-1597.
基本信息:
DOI：10.19635/j.cnki.csu-epsa.001692
中图分类号:TM712;TP18
引用信息:
[1]李欣,张耀为,赵乔,等.融合时序卷积与多头自注意力的暂态电压稳定评估[J].电力系统及其自动化学报,2026,38(04):12-24.DOI:10.19635/j.cnki.csu-epsa.001692.
基金信息:
国家自然科学基金资助项目（52107107）
2025-05-30
2025
2025-07-16
2026-04-27
2026
1
2025-08-01
2025-08-01
2025-08-01