基于改进CGAN的电力系统暂态稳定评估样本增强方法
|
摘要
基于数据驱动的暂态稳定评估方法已成为电网安全领域研究的重点,由于实际电力系统中暂态失稳情况极少,给通过数据挖掘方法判断失稳情况带来了极大困难。针对这个问题,提出了一种用于暂态稳定评估中失稳样本合成的数据增强方法,对条件生成对抗神经网络(CGAN)训练方法的适应性进行改进以提高其学习稳定性,在离线训练时利用改进CGAN交替训练生成器和判别器,学习电力系统暂态数据的分布特性,然后采用极限学习机(ELM)分类器筛选出改进CGAN所生成的多组样本中G-mean值最高的生成样本,将其中失稳样本对原始失稳样本进行增强,最后用增强后的原始样本训练分类器,实现在线暂态稳定评估。仿真结果表明,所提出的样本数据增强方法通过改进CGAN实现对原始数据分布特征的有效学习,进而提升暂态稳定评估的正确率,具有抗噪声干扰性强、对高维数据鲁棒性好的优点,能够有效平衡电力系统失稳数据。
Data-driven transient stability assessment method has become the focus of research in the field of power network security.However,transient unstable situation in the actual power system is very rare,which brings great difficulties to the data acquisition method for judging the instability.This paper proposes a data augment method for the synthesis of unstable samples in the transient stability assessment.It enhances the adaptability of training methods for conditional generative adversarial network(CGAN)to improve their learning stability and uses the improved CGAN training generators and discriminators during offline training to learn the distribution characteristics of raw data.Then,the extreme learning machine(ELM)classifier is used to filter out the generated samples with the highest G-mean value among the multiple sets of samples generated by the improved CGAN.The unstable samples are used to augment the original unstable samples,and the augmented original samples are used to train the classifier to achieve online transient stability assessment.The simulation results show that the proposed method can effectively learn the distribution characteristics of the original data by the improved CGANs.The method has the advantages of strong anti-noise interference and good robustness to high-dimensional data,and it can effectively balance the unstable data of power system.
Data-driven transient stability assessment method has become the focus of research in the field of power network security.However,transient unstable situation in the actual power system is very rare,which brings great difficulties to the data acquisition method for judging the instability.This paper proposes a data augment method for the synthesis of unstable samples in the transient stability assessment.It enhances the adaptability of training methods for conditional generative adversarial network(CGAN)to improve their learning stability and uses the improved CGAN training generators and discriminators during offline training to learn the distribution characteristics of raw data.Then,the extreme learning machine(ELM)classifier is used to filter out the generated samples with the highest G-mean value among the multiple sets of samples generated by the improved CGAN.The unstable samples are used to augment the original unstable samples,and the augmented original samples are used to train the classifier to achieve online transient stability assessment.The simulation results show that the proposed method can effectively learn the distribution characteristics of the original data by the improved CGANs.The method has the advantages of strong anti-noise interference and good robustness to high-dimensional data,and it can effectively balance the unstable data of power system.
引文
[1]KUNDUR P,PASERBA J,AJJARAPU V,et al.Definition and classification of power system stability[J].IEEETransactions on Power Systems,2004,19(3):1387-1401.
[2]ZHANG C,LI Y,YU Z,et al.A weighted random forest approach to improve predictive performance for power system transient stability assessment[C]//Power and Energy Engineering Conference(APPEEC),April 15-17,2016,Suzhou,China:1259-1263.
[3]KUNDUR P,BALU N J,LAUBY M G.Power system stability and control[M].New York:McGraw-hill,1994:848-859.
[4]XUE Y,VAN CUTSEM T,RIBBENS-PAVELLA M.Asimple direct method for fast transient stability assessment of large power systems[J].IEEE Transactions on Power Systems,1988,3(2):400-412.
[5]王锡凡.现代电力系统分析[M].北京:科学出版社,2003:354-362.WANG Xifan.Analysis on modern power system[M].Beijing:Science Press,2003:354-362.
[6]ZHANG Yingchen,MARKHAM P,XIA Tao,et al.Wide-area frequency monitoring network(FNET)architecture and applications[J].IEEE Transactions on Smart Grid,2010,1(2):159-167.
[7]DE LA R J,CENTENO V,THORP J S,et al.Synchronized phasor measurement applications in power systems[J].IEEETransactions on Smart Grid,2010,1(1):20-27.
[8]于之虹,黄彦浩,鲁广明,等.基于时间序列关联分析的稳定运行规则提取方法[J].中国电机工程学报,2015,35(3):519-526.YU Zhihong,HUANG Yanhao,LU Guangming,et al.A time series associative classification method for the operation rule extracting of transient stability[J].Proceedings of the CSEE,2015,35(3):519-526.
[9]黄天罡,薛禹胜,陈国平,等.暂态稳定算例的高效筛除[J].电力系统自动化,2018,42(8):83-91.DOI:10.7500/AEPS20180302011.HUANG Tiangang,XUE Yusheng,CHEN Guoping,et al.An efficient stable case screening algorithm for transient stability assessment[J].Automation of Electric Power Systems,2018,42(8):83-91.DOI:10.7500/AEPS20180302011.
[10]朱乔木,陈金富,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.
[11]TANG Y,LI F,WANG Q,et al.Hybrid method for power system transient stability prediction based on two-stage computing resources[J].IET Generation,Transmission&Distribution,2018,12(8):1697-1703.
[12]GUO T,MILANOVIC J V.Probabilistic framework for assessing the accuracy of data mining tool for online prediction of transient stability[J].IEEE Transactions on Power Systems,2014,29(1):377-385.
[13]TSO S K,GU X P,ZENG Q Y,et al.An ANN-based multilevel classification approach using decomposed input space for transient stability assessment[J].Electric Power Systems Research,1998,46(3):259-266.
[14]尹雪燕,闫炯程,刘玉田,等.基于深度学习的暂态稳定评估与严重度分级[J].电力自动化设备,2018,38(5):64-69.YIN Xueyan,YAN Jiongcheng,LIU Yutian,et al.Deep learning based transient stability assessment and severity grading[J].Electric Power Automation Equipment,2018,38(5):64-69.
[15]叶圣永,王晓茹,刘志刚,等.基于支持向量机增量学习的电力系统暂态稳定评估[J].电力系统自动化,2011,35(11):15-19.YE Shengyong,WANG Xiaoru,LIU Zhigang,et al.Power system transient stability assessment based on support vector machine incremental learning method[J].Automation of Electric Power Systems,2011,35(11):15-19.
[16]WANG B,FANG B,WANG Y,et al.Power system transient stability assessment based on big data and the core vector machine[J].IEEE Transactions on Smart Grid,2016,7(5):2561-2570.
[17]ZHANG Rui,XU Yan,DONG Zhaoyang,et al.Postdisturbance transient stability assessment of power systems by a self-adaptive intelligent system[J].IET Generation Transmission&Distribution,2015,9(3):296-305.
[18]JAMES J Q,HILL D J,LAM A S,et al.Intelligent timeadaptive transient stability assessment system[J].IEEETransactions on Power Systems,2018,33(1):1049-1058.
[19]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.
[20]李倩倩,刘胥影.多类类别不平衡学习算法:EasyEnsemble.M[J].模式识别与人工智能,2014,27(2):187-192.LI Qianqian,LIU Xuying.EasyEnsemble.M for multiclass imbalance problem[J].Pattern Recognition and Artificial Intelligence,2014,27(2):187-192.
[21]FAN W,STOLFO S J,ZHANG J,et al.AdaCost:misclassification cost-sensitive boosting[C]//Sixteenth International Conference on Machine Learning,June 27-30,1999,Bled,Slovenia:97-105.
[22]ZHU L,LU C,DONG Z Y,et al.Imbalance learning machine-based power system short-term voltage stability assessment[J].IEEE Transactions on Industrial Informatics,2017,13(5):2533-2543.
[23]BATISTA G E,PRATI R C,MONARD M C.A study of the behavior of several methods for balancing machine learning training data[J].ACM Sigkdd Explorations Newsletter,2004,6(1):20-29.
[24]CHAWLA N V,BOWYER K W,HALL L O,et al.SMOTE:synthetic minority over-sampling technique[J].Journal of Artificial Intelligence Research,2002,16:321-357.
[25]CHEN Y,WANG Y,KIRSCHEN D,et al.Model-free renewable scenario generation using generative adversarial networks[J].IEEE Transactions on Power Systems,2018,33(3):3265-3275.
[26]ARJOVSKY M,CHINTALA S,BOTTOU L.Wasserstein generative adversarial networks[C]//International Conference on Machine Learning,August 6-11,2017,Sydney,Australia:214-223.
[27]GOODFELLOW I,POUGET-ABADIE J,MIRZA M,et al.Generative adversarial nets[C]//Advances in Neural Information Processing Systems,December 8-13,2014,Montreal,Canada:2672-2680.
[28]MIRZA M,OSINDERO S.Conditional generative adversarial nets[R/OL].[2014-11-04].https://arxiv.org/abs/1411.1784.
[29]NAIR V,HINTON G E.Rectified linear units improve restricted Boltzmann machines[C]//Proceedings of the 27th International Conference on Machine Learning,June 21-24,2010,Haifa,Israel:807-814.
[30]DAUPHIN Y,DE VRIES H,BENGIO Y.Equilibrated adaptive learning rates for non-convex optimization[C]//Advances in Neural Information Processing Systems,December 7-12,2015,Montreal,Canada:1504-1512.
[31]HUANG G B,ZHU Q Y,SIEW C K.Extreme learning machine:theory and applications[J].Neurocomputing,2006,70(1/2/3):489-501.
[32]XU Y,DONG Z Y,MENG K,et al.Real-time transient stability assessment model using extreme learning machine[J].IET Generation Transmission&Distribution,2011,5(3):314-322.
[33]ZHANG Y,XU Y,DONG Z Y,et al.Intelligent early warning of power system dynamic insecurity risk:toward optimal accuracy-earliness tradeoff[J].IEEE Transactions on Industrial Informatics,2017,13(5):2544-2554.
[34]ABADI M,BARHAM P,CHEN J,et al.TensorFlow:a system for large-scale machine learning[C]//OSDI,November 2-4,2016,Savannah,USA:265-283.
[35]CHOLLET F.Keras:Theano-based deep learning library[EB/OL].[2018-05-09].https://github.com/keras-team/keras.
[36]PAI M A.Energy function analysis for power system stability[M].New York:Springer Science&Business Media,1989:223-227.
[37]SIMENS.PSS/E[EB/OL].[2018-05-09].https://www.siemens.com/global/en/home/products/energy/services/transmission-distribution-smart-grid/consulting-and-planning/pss-software/pss-e.html.
[38]CHEN T,HE T,BENESTY M.XGBoost[EB/OL].[2018-05-09].http://xgboost.readthedocs.io/en/latest/.
[39]HE M,VITTAL V,ZHANG J.Online dynamic security assessment with missing PMU measurements:a data mining approach[J].IEEE Transactions on Power Systems,2013,28(2):1969-1977.
[40]BLAGUS R,LUSA L.SMOTE for high-dimensional classimbalanced data[J].BMC Bioinformatics,2013,14(1):106.
[2]ZHANG C,LI Y,YU Z,et al.A weighted random forest approach to improve predictive performance for power system transient stability assessment[C]//Power and Energy Engineering Conference(APPEEC),April 15-17,2016,Suzhou,China:1259-1263.
[3]KUNDUR P,BALU N J,LAUBY M G.Power system stability and control[M].New York:McGraw-hill,1994:848-859.
[4]XUE Y,VAN CUTSEM T,RIBBENS-PAVELLA M.Asimple direct method for fast transient stability assessment of large power systems[J].IEEE Transactions on Power Systems,1988,3(2):400-412.
[5]王锡凡.现代电力系统分析[M].北京:科学出版社,2003:354-362.WANG Xifan.Analysis on modern power system[M].Beijing:Science Press,2003:354-362.
[6]ZHANG Yingchen,MARKHAM P,XIA Tao,et al.Wide-area frequency monitoring network(FNET)architecture and applications[J].IEEE Transactions on Smart Grid,2010,1(2):159-167.
[7]DE LA R J,CENTENO V,THORP J S,et al.Synchronized phasor measurement applications in power systems[J].IEEETransactions on Smart Grid,2010,1(1):20-27.
[8]于之虹,黄彦浩,鲁广明,等.基于时间序列关联分析的稳定运行规则提取方法[J].中国电机工程学报,2015,35(3):519-526.YU Zhihong,HUANG Yanhao,LU Guangming,et al.A time series associative classification method for the operation rule extracting of transient stability[J].Proceedings of the CSEE,2015,35(3):519-526.
[9]黄天罡,薛禹胜,陈国平,等.暂态稳定算例的高效筛除[J].电力系统自动化,2018,42(8):83-91.DOI:10.7500/AEPS20180302011.HUANG Tiangang,XUE Yusheng,CHEN Guoping,et al.An efficient stable case screening algorithm for transient stability assessment[J].Automation of Electric Power Systems,2018,42(8):83-91.DOI:10.7500/AEPS20180302011.
[10]朱乔木,陈金富,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.
[11]TANG Y,LI F,WANG Q,et al.Hybrid method for power system transient stability prediction based on two-stage computing resources[J].IET Generation,Transmission&Distribution,2018,12(8):1697-1703.
[12]GUO T,MILANOVIC J V.Probabilistic framework for assessing the accuracy of data mining tool for online prediction of transient stability[J].IEEE Transactions on Power Systems,2014,29(1):377-385.
[13]TSO S K,GU X P,ZENG Q Y,et al.An ANN-based multilevel classification approach using decomposed input space for transient stability assessment[J].Electric Power Systems Research,1998,46(3):259-266.
[14]尹雪燕,闫炯程,刘玉田,等.基于深度学习的暂态稳定评估与严重度分级[J].电力自动化设备,2018,38(5):64-69.YIN Xueyan,YAN Jiongcheng,LIU Yutian,et al.Deep learning based transient stability assessment and severity grading[J].Electric Power Automation Equipment,2018,38(5):64-69.
[15]叶圣永,王晓茹,刘志刚,等.基于支持向量机增量学习的电力系统暂态稳定评估[J].电力系统自动化,2011,35(11):15-19.YE Shengyong,WANG Xiaoru,LIU Zhigang,et al.Power system transient stability assessment based on support vector machine incremental learning method[J].Automation of Electric Power Systems,2011,35(11):15-19.
[16]WANG B,FANG B,WANG Y,et al.Power system transient stability assessment based on big data and the core vector machine[J].IEEE Transactions on Smart Grid,2016,7(5):2561-2570.
[17]ZHANG Rui,XU Yan,DONG Zhaoyang,et al.Postdisturbance transient stability assessment of power systems by a self-adaptive intelligent system[J].IET Generation Transmission&Distribution,2015,9(3):296-305.
[18]JAMES J Q,HILL D J,LAM A S,et al.Intelligent timeadaptive transient stability assessment system[J].IEEETransactions on Power Systems,2018,33(1):1049-1058.
[19]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.
[20]李倩倩,刘胥影.多类类别不平衡学习算法:EasyEnsemble.M[J].模式识别与人工智能,2014,27(2):187-192.LI Qianqian,LIU Xuying.EasyEnsemble.M for multiclass imbalance problem[J].Pattern Recognition and Artificial Intelligence,2014,27(2):187-192.
[21]FAN W,STOLFO S J,ZHANG J,et al.AdaCost:misclassification cost-sensitive boosting[C]//Sixteenth International Conference on Machine Learning,June 27-30,1999,Bled,Slovenia:97-105.
[22]ZHU L,LU C,DONG Z Y,et al.Imbalance learning machine-based power system short-term voltage stability assessment[J].IEEE Transactions on Industrial Informatics,2017,13(5):2533-2543.
[23]BATISTA G E,PRATI R C,MONARD M C.A study of the behavior of several methods for balancing machine learning training data[J].ACM Sigkdd Explorations Newsletter,2004,6(1):20-29.
[24]CHAWLA N V,BOWYER K W,HALL L O,et al.SMOTE:synthetic minority over-sampling technique[J].Journal of Artificial Intelligence Research,2002,16:321-357.
[25]CHEN Y,WANG Y,KIRSCHEN D,et al.Model-free renewable scenario generation using generative adversarial networks[J].IEEE Transactions on Power Systems,2018,33(3):3265-3275.
[26]ARJOVSKY M,CHINTALA S,BOTTOU L.Wasserstein generative adversarial networks[C]//International Conference on Machine Learning,August 6-11,2017,Sydney,Australia:214-223.
[27]GOODFELLOW I,POUGET-ABADIE J,MIRZA M,et al.Generative adversarial nets[C]//Advances in Neural Information Processing Systems,December 8-13,2014,Montreal,Canada:2672-2680.
[28]MIRZA M,OSINDERO S.Conditional generative adversarial nets[R/OL].[2014-11-04].https://arxiv.org/abs/1411.1784.
[29]NAIR V,HINTON G E.Rectified linear units improve restricted Boltzmann machines[C]//Proceedings of the 27th International Conference on Machine Learning,June 21-24,2010,Haifa,Israel:807-814.
[30]DAUPHIN Y,DE VRIES H,BENGIO Y.Equilibrated adaptive learning rates for non-convex optimization[C]//Advances in Neural Information Processing Systems,December 7-12,2015,Montreal,Canada:1504-1512.
[31]HUANG G B,ZHU Q Y,SIEW C K.Extreme learning machine:theory and applications[J].Neurocomputing,2006,70(1/2/3):489-501.
[32]XU Y,DONG Z Y,MENG K,et al.Real-time transient stability assessment model using extreme learning machine[J].IET Generation Transmission&Distribution,2011,5(3):314-322.
[33]ZHANG Y,XU Y,DONG Z Y,et al.Intelligent early warning of power system dynamic insecurity risk:toward optimal accuracy-earliness tradeoff[J].IEEE Transactions on Industrial Informatics,2017,13(5):2544-2554.
[34]ABADI M,BARHAM P,CHEN J,et al.TensorFlow:a system for large-scale machine learning[C]//OSDI,November 2-4,2016,Savannah,USA:265-283.
[35]CHOLLET F.Keras:Theano-based deep learning library[EB/OL].[2018-05-09].https://github.com/keras-team/keras.
[36]PAI M A.Energy function analysis for power system stability[M].New York:Springer Science&Business Media,1989:223-227.
[37]SIMENS.PSS/E[EB/OL].[2018-05-09].https://www.siemens.com/global/en/home/products/energy/services/transmission-distribution-smart-grid/consulting-and-planning/pss-software/pss-e.html.
[38]CHEN T,HE T,BENESTY M.XGBoost[EB/OL].[2018-05-09].http://xgboost.readthedocs.io/en/latest/.
[39]HE M,VITTAL V,ZHANG J.Online dynamic security assessment with missing PMU measurements:a data mining approach[J].IEEE Transactions on Power Systems,2013,28(2):1969-1977.
[40]BLAGUS R,LUSA L.SMOTE for high-dimensional classimbalanced data[J].BMC Bioinformatics,2013,14(1):106.