叶琳,杨滢,洪道鉴,陈新建
YE Lin,YANG Ying,HONG Daojian,CHEN Xinjian

• 1:国网浙江省电力有限公司
• 2:国网浙江省电力有限公司台州供电公司

摘要(Abstract)：

近年来,人工智能特别是深度学习技术的迅速发展,给当今社会带来了巨大变革。首先梳理了人工智能尤其是机器学习的关键及前沿技术,阐述了包括强化学习、迁移学习、生成对抗神经网络、胶囊网络和引导学习等几种典型机器学习方法的特点。然后分析了机器学习在电力系统稳定性分析领域、协调调度领域以及负荷预测领域的典型应用场景,对比了其在解决特定问题时的优势。最后对应用情况进行了概括总结,展望了其在电力系统运行领域的应用前景。
In recent years, artificial intelligence, especially deep learning technology, has developed rapidly,bringing about tremendous innovation in social technology. This paper summarizes the key and cutting-edge technologies of artificial intelligence and expounds the characteristics of typical machine learning such as reinforcement learning, migration learning, generation of anti-neural network, capsule network and guided learning; then, it analyzes the deep learning technology in the field of stability analysis, coordination scheduling and load forecasting in the power system operation, which shows its advantages in solving specific problems.Finally, the paper summarizes the application and elaborates on the future application in electric power system operation.

关键词(KeyWords)： 人工智能;机器学习;强化学习;迁移学习;对抗神经网络;胶囊网络;引导学习
artificial intelligence;machine learning;reinforcement learning;migration learning;anti-neural network;capsule network;guided learning

Abstract:

Keywords:

基金项目(Foundation): 浙江省电力有限公司科技项目(5211TZ170006)

作者(Author): 叶琳,杨滢,洪道鉴,陈新建
YE Lin,YANG Ying,HONG Daojian,CHEN Xinjian

DOI: 10.19585/j.zjdl.201905013

参考文献(References)：

• [1]黄安埠.深入浅出深度学习[M].北京:电子工业出版社,2017.
• [2]余凯,贾磊,陈雨强,等.深度学习的昨天、今天和明天[J].计算机研究与发展,2013,50(9):1799-1804.
• [3]焦李成,杨淑媛,刘芳,等.神经网络七十年:回顾与展望[J].计算机学报,2016,39(8):1697-1716.
• [4]郭勤.基于深度强化学习的视频游戏决策模型研究与应用[D].赣州:江西理工大学,2018.
• [5]王怀智.智能发电控制的多目标优化策略及其均衡强化学习理论[D].广州:华南理工大学,2015.
• [6]SUTTON R S.An introduction to reinforcement learning[M].[S.l.]:The MIT Press,1998.
• [7]BUSONIU L,ROBERT B,BART D S,et al.Reinforcement Learning and Dynamic Programming using Function Approximators[M].[S.l.]:Taylor&Francis CRC Press,2010.
• [8]张孝顺.电力系统的迁移强化学习优化算法研究[D].广州:华南理工大学,2017.
• [9]PAN S J,YANG Q.A survey on transfer learning[J].IEEETransactions on Knowledge&Data Engineering,2010,22(10):1345-1359.
• [10]庄福振,罗平,何清,等.迁移学习研究进展[J].软件学报,2015,26(1):26-39.
• [11]WEI F M,ZHANG J P,CHU Y,et al.Transfer learning from long texts to the short[J].Applied Mathematics&Information Sciences,2014,8(4):2033-2044.
• [12]DU B,ZHANGL,TAO D,et al.Unsupervised transfer learning for target detection from hyperspectral images[J].Neurocomputing,2013,120(10):72-82.
• [13]ARNOLD A,NALLAPATI R,COHEN W W.A compara tive study of methods for transductive transfer learning[C]//IEEE International Conference on Data Mining Workshops,2007.ICDM Workshops.IEEE,2007:77-82.
• [14]张倩,李明,王雪松,等.一种面向多源领域的实例迁移学习[J].自动化学报,2013,40(6):1176-1183.
• [15]MENG J,LIN H,LI Y.Knowledge transfer based on feature representation mapping for text classification[J].Expert Systems with Applications,2011,38(8):10562-10567.
• [16]GAO J,FAN W,JIANG J,et al.Knowledge transfer via multiple model local structure mapping[C]//ACM SIGKDDInternational Conference on Knowledge Discovery and Data Mining,Las Vegas,Nevada,USA,August.DBLP,2008:283-291.
• [17]MIHALKOVA L,HUYNH T,MOONEY R J.Mapping and revising Markov logic networks for transfer learning[C]//National Conference on Artificial Intelligence.AAAI Press,2007:608-614.
• [18]GUI L,XU R,LU Q,et al.Negative transfer detection in transductive transfer learning[J].International Journal of Machine Learning&Cybernetics,2018,9(2):185-197.
• [19]HU Q,ZHANG R,ZHOU Y.Transfer learning for shortterm wind speed prediction with deep neural networks[J].Ren ewable Energy,2016(85):83-95.
• [20]GOODFELLOW I,POUGET-ABADIE J,MIRZA M,et al.Generative adversarial nets[C]//Advances in Neural Information Processing Systems,2014:2672-2680.
• [21]崔鸿雁,徐帅,张利锋,等.机器学习中的特征选择方法研究及展望[J].北京邮电大学学报,2018,41(1):1-12.
• [22]SABOUR S,FROSST N,HINTON G,et al.Dynamic Routing Between Capsules[C]//2017 The Thirty-first Annual Conference on Neural Information Processing Systems,Long Beach,CA,USA,Dec 4-9,2017.
• [23]HINTON G,SABOUR S,FROSST N.Matrix Capsules with EM Routing[C]//2018 6th International Conference on Learning Representations,Toronto,Canada,April 30-May3.2018.
• [24]尚宇炜,马钊,彭晨阳,等.内嵌专业知识和经验的机器学习方法探索(一):引导学习的提出与理论基础[J].中国电机工程学报,2017,37(19):5560-5571.
• [25]尚宇炜,马钊,彭晨阳,等.内嵌专业知识和经验的机器学习方法探索(二):引导学习的提出与理论基础[J].中国电机工程学报,2017,37(20):5852-5861.
• [26]杨培宏,刘文颖.基于改进Prony算法在线辨识电力系统低频振荡模式[J].低压电器,2008(5):47-51.
• [27]宋墩文,温渤婴,杨学涛,等.基于多信息源的大电网低频振荡预警及防控决策系统[J].电力系统保护与控制,2016(21):54-60.
• [28]王学健.基于智能计算的电力系统低频振荡模式辨识的研究[D].广州:华南理工大学,2017.
• [29]王颖凯.基于深度学习算法的电力系统低频振荡模式识别[D].广州:华南理工大学,2017.
• [30]YU X H,GAO F,et al.Deep Learning Based Transient Stability Assessment for Grid-Connected Inverter[C]//2018IEEE International Power Electronics and Application Conference and Exposition,Shenzhen,China,November 4-7,2018.
• [31]GUO Z,TANG X W,et al.Research on Controlling Method of Frequent Monitoring Signal in Regional Distribution Network Based on AI Algorithm[C]//2018 China International Conference on Electricity Distribution.Tianjin,China,17-19 Sep.2018.
• [32]张婧昕.考虑分布式发电与需求响应的电网优化调度与配网规划研究[D].杭州:浙江大学,2018.
• [33]韩传家,张孝顺,余涛,等.风险调度中引入知识迁移的细菌觅食强化学习优化算法[J].电力系统自动化,2017(8):69-77.
• [34]GUO ZONG,TANG X W,et al.Smart Grid Optimization by Deep Reinforcement Learning over Discrete and Continuous Action Space[C]//2018 China International Conference on Electricity Distribution.Tianjin,China,17-19Sep.2018.
• [35]PENG X S,WANG B,et al.A Deep Learning Approach for Wind Power Prediction Based on Stacked Denoising Auto Encoders Optimized by Bat Algorithm[C]//2018 China International Conference on Electricity Distribution,Tianjin,China,17-19 Sep.2018.
• [36]QU X Y,KANG X N,et al.Short-term prediction of wind power based on deep Long Short-Term Memory[C]//2018China International Conference on Electricity Distribution,Tianjin,China,October 25-28,2016.
• [37]TAO Y B,CHEN H K,et al.Wind power prediction and pattern feature based on deep learning method[C]//2014IEEE PES Asia-Pacific Power and Energy Engineering Conference(APPEEC),Hong Kong,China,7-10 December,2014.
• [38]刘世昌.多尺度分析与数据互迁移相结合的短期电力负荷预测方法[D].长沙:湖南大学,2017.
• [39]董浩.深度学习算法在电力系统短期负荷预测中的应用[J].电气时代,2017(2):82-84.
• [40]黄宇腾,韩翊,赖尚栋.深度神经网络在配电网公变短期负荷预测中的应用研究[J].浙江电力,2018,37(5):1-6.
• [41]单成龙.基于深度学习的电力负荷预测[D].湘潭:湘潭大学,2017.
• [42]许祖锋.基于模型驱动的变电站智能告警应用软件研究与设计[D].北京:华北电力大学,2017.
• [43]唐雅洁.基于云服务的智能电网调度监控平台与辅助决策[D].杭州:浙江大学,2011.
• [44]李宏伟.应用于智能变电站的智能视觉系统[J].电力自动化设备,2012,32(8):141-147.
• [45]赵继生.基于卷积神经网络的变电站监控图像识别方法研究[D].北京:华北电力大学,2016.
• [46]林磊,钱平,董毅.基于深度学习的变电站环境下行人检测方法研究[J].浙江电力,2018,37(7):68-73.
• [47]赵令令.基于超列的变电设备红外与可见光图像配准研究[D].北京:华北电力大学,2017.
• [48]陈旭.基于深度学习的变电设备图像特征提取[D].南京:南京邮电大学,2018.
• [49]吕宁.基于深度学习的输电线路防外力破坏监测系统研究[D].哈尔滨:哈尔滨工程大学,2018.
• [50]李程启,林颖,秦佳峰.基于深度学习的输电线路危险源智能监控系统[J].南通大学学报,2018,17(1):10-14.
• [51]黄良,王佳丽,赵立进.面向文本非结构化数据的输变电系统故障诊断方法[J].电力科学与技术学报,2017,32(3):154-161.
• [52]尚宇炜,郭剑波,吴文传,等.电力脑初探:一种多模态自适应学习系统[J].中国电机工程学报,2018,38(11):3133-3143.