张若愚;吴俊勇;李宝琴;邵美阳;

• 1:北京交通大学电气工程学院

摘要(Abstract)：

基于人工智能的电力系统暂态稳定预测,通常需要用离线生成的大量暂稳样本对预测模型进行训练,然后根据系统的实时响应进行在线预测。但当系统的运行方式和拓扑结构发生较大变化时,预测模型的精度会显著下降,亟需一种能跟踪系统变化的自适应暂稳预测方法。针对该问题,将迁移学习引入电力系统暂稳预测,基于卷积神经网络提出了一种自适应预测方法。首先利用离线生成的大量暂稳样本训练并得到基于卷积神经网络的预训练模型。当系统运行方式和拓扑结构发生较大变化时,保持预训练模型的网络结构不变,将其中的2个卷积层、2个池化层和全连接层的网络参数迁移至新模型;提出了一种最小均衡样本集的变步长生成方法,用新生成的最小均衡样本集训练分类层参数,从而快速得到新的预测模型。新英格兰10机39节点系统的测试结果表明:所提方法能自适应跟踪系统运行方式和拓扑结构的变化,有效更新预测模型且大幅减少新模型的训练时间,为基于人工智能的电力系统暂态稳定自适应预测提供了一条新思路。

关键词(KeyWords)： 迁移学习;深度学习;卷积神经网络;电力系统;暂态稳定预测

Abstract:

Keywords:

基金项目(Foundation): 国家重点研发计划项目(2018YFB0904500);; 国家电网有限公司科技项目(SGLNDK00KJJS1800236)~~

作者(Author): 张若愚;吴俊勇;李宝琴;邵美阳;

Email:

DOI: 10.13335/j.1000-3673.pst.2019.2376

参考文献(References)：

• [1]典焱．英国大停电事故分析：新能源大量替代传统火电将导致系统惯量水平下降[J]．电力设备管理，2019(9):1-2.Dian Yan.Analysis of the blackout in the UK:A large number of new energy sources to replace traditional thermal power will lead to a decline in system inertia[J].Power Equipment Management,2019(9):1-2(in Chinese).
• [2]段荣华，段平生，黄伟．巴西大停电的分析及对云南电网的启示[J]．云南电力技术，2019,47(3):65-68.Duan Ronghua,Duan Pingsheng,Huang Wei.Analysis of Brazilian blackout and its enlightenment to Yunnan power grid[J].Yunnan Electric Power,2019,47(3):65-68(in Chinese).
• [3]朱朝阳．委内瑞拉大停电事故的背后[J]．国家电网，2019(5):72-74.Zhu Chaoyang.Behind the blackout in Venezuela[J].State Grid,2019(5):72-74(in Chinese).
• [4]于之虹，施浩波，安宁，等．暂态稳定多故障协调预防控制策略在线计算方法[J]．电网技术，2014,38(6):1554-1561.Yu Zhihong,Shi Haobo,An Ning,et al.A computational approach of preventive control strategy with multi-contingency constraints for power grid transient stability[J].Power System Technology,2014,38(6):1554-1561(in Chinese).
• [5]马世英，刘道伟，吴萌，等．基于WAMS及机组对的电网暂态稳定态势在线量化评估方法[J]．电网技术，2013,37(5):1323-1328.Ma Shiying,Liu Daowei,Wu Meng,et al.An online quantitative evaluation method of power system transient stability situation based on WAMS and generator pair[J].Power System Technology,2013,37(5):1323-1328(in Chinese).
• [6]吴为，汤涌，孙华东，等．基于广域量测信息的电力系统暂态稳定研究综述[J]．电网技术，2012,36(9):81-87.Wu Wei,Tang Yong,Sun Huadong,et al.A survey on research of power system transient stability based on wide-area measurement information[J].Power System Technology,2012,36(9):81-87(in Chinese).
• [7]李亚楼，李芳，刘赫川，等．基于PSASP的综合能源仿真分析系统[J]．电网技术，2019,43(7):2498-2505.Li Yalou,Li Fang,Liu Hechuan,et al.Framework design of simulation system for integrated energy systems based on PSASP[J].Power System Technology,2019,43(7):2498-2505(in Chinese).
• [8]朱雪琼，薛禹胜，黄天罡．机组惯量对暂态功角稳定性影响的复杂性[J]．电力系统自动化，2019,43(17):72-77.Zhu Xueqiong,Xue Yusheng,Huang Tiangang.Complexity of influence of generator inertia on transient angle stability[J].Automation of Electric Power Systems,2019,43(17):72-77(in Chinese).
• [9]周艳真，吴俊勇，于之虹，等．基于转子角轨迹簇特征的电力系统暂态稳定评估[J]．电网技术，2016,40(5):1482-1487.Zhou Yanzhen,Wu Junyong,Yu Zhihong,et al.Power system transient stability assessment based on cluster features of rotor angle trajectories[J].Power System Technology,2016,40(5):1482-1487(in Chinese).
• [10]冀鲁豫，吴俊勇，周艳真，等．基于WAMS受扰电压轨迹簇特征的电力系统暂态稳定性预测[J]．高电压技术，2015,41(3):807-814.Ji Luyu,Wu Junyong,Zhou Yanzhen,et al.Transient stability prediction of power system based on WAMS characteristic of perturbed voltage trajectory clusters[J].High Voltage Engineering,2015,41(3):807-814(in Chinese).
• [11] Zhou Yanzhen,Guo Qinglai,Sun Hongbin,et al.A novel data-driven approach for transient stability prediction of power systems considering the operational variability[J].Electrical Power and Energy Systems,2019(107):379-394.
• [12] Zhang Yuchen,Xu Yan,Dong Zhaoyang,et al.Intelligent early warning of power system dynamic insecurity risk:toward optimal accuracy-earliness trade off[J].IEEE Transactions on industrial informatics,2017,13(5):2544-2554.
• [13] Zhu Qiaomu,Chen Jinfu,Zhu Lin,et al.A deep end-to-end model for transient stability assessment with PMU data[J].IEEE Access,2018(6):65474-65487.
• [14]朱乔木，陈金福，李弘毅，等．基于堆叠自动编码器的电力系统暂态稳定评估[J]．中国电机工程学报，2018,38(10):2937-2946.Zhu Qiaomu,Chen Jinfu,Li Hongyi,et al.Transient stability assessment based on stacked autoencoder[J].Proceedings of the CSEE,2018,38(10):2937-2946(in Chinese).
• [15]朱乔木，党杰，陈金福，等．基于深度置信网络的电力系统暂态稳定评估方法[J]．中国电机工程学报，2018,38(3):735-743.Zhu Qiaomu,Dang Jie,Chen Jinfu,et al.A method for power system transient stability assessment based on deep belief networks[J].Proceedings of the CSEE,2018,38(3):735-743(in Chinese).
• [16]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 Kunlun,Yang Shuai,Liu Siyan,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(in Chinese).
• [17] Zhang Ruoyu,Wu Junyong,Xu Yan,et al.A hierarchical selfadaptive method for post-disturbance transient stability assessment of power systems using an integrated CNN-based ensemble classifier[J].Energies,2019,12(17):3217.
• [18] James J Q,David J,Albert Y,et al.Intelligent time-adaptive transient stability assessment system[J].IEEE Transactions on Power Systems,2018,33(1):1049-1058.
• [19] Zhang Rui,Xu Yan,Dong Zhaoyang,et al.Post-disturbance transient stability assessment of power systems by a self-adaptive intelligent system[J].IET Generation,Transmission&Distribution,2015,9(3):296-305.
• [20]张驰名，王庆凤，刘志勤，等．基于深度迁移学习的肺结节良恶性辅助诊断方法[J]．计算机工程，2020(1):271-278.Zhang Chiming,Wang Qingfeng,Liu Zhiqing,et al.A layer-wise progressive fine-tuning strategy based transfer learning for pulmonary nodule auxiliary diagnosis[J].Computer Engineering,2020(1):271-278(in Chinese).
• [21]王俊超，黄浩，徐海华，等．基于迁移学习的低资源度维吾尔语语音识别[J]．计算机工程，2018,44(10):281-291.Wang Junchao,Huang Hao,Xu Haihua,et al.Low-resource Uyghur speech recognition based on transfer learning[J]. Computer Engineering,2018,44(10):281-291(in Chinese).
• [22]王毅星．基于深度学习和迁移学习的电力数据挖掘技术研究[D]．杭州：浙江大学，2019.
• [23] Ren Chao,Xu Yan.Transfer learning-based power system online dynamic security assessment:using one model to assess many unlearned faults[J].IEEE transactions on Power Systems,2020,35(1):821-824.
• [24]田芳，周孝信，史东宇，等．基于卷积神经网络综合模型和稳态特征量的电力系统暂态稳定评估方法[J]．中国电机工程学报，2019,39(14):4025-4032.Tian Fang,Zhou Xiaoxin,Shi Dongyu,et al.Power system transient stability assessment based on comprehensive convolutional neural network model and steady-state features[J].Proceedings of the CSEE,2019,39(14):4025-4032(in Chinese).
• [25]王怀远，陈启凡．基于代价敏感堆叠变分自动编码器的暂态稳定评估方法[J]．中国电机工程学报，2020,40(7):2213-2220+2400.Wang Huaiyuan,Chen Qifan.A transient stability assessment method based on cost-sensitive stacked variational auto-encoder[J].Proceedings of the CSEE,2020,40(7):2213-2220+2400(in Chinese).