基于XGBoost算法的电力系统暂态稳定评估
|
摘要
针对暂态稳定评估问题的特点,在改进极限梯度提升(XGBoost)算法的基础上进行暂态稳定评估。根据电网物理特点,定义能够反映电力系统稳态运行状态的特征集;研究XGBoost算法用于暂态稳定评估的过程:针对暂态稳定预测中2类错误严重程度不同的特点,定义包含注意力系数的对数损失函数,使得模型对不稳定样本的误预测情况减少;使用Logistic函数将模型输出概率化,用于衡量XGBoost模型输出的可靠程度,预防部分误预测;给出针对任意系统随机产生样本集的方法。IEEE 39节点系统仿真结果表明,XGBoost算法在准确率上均高于其他几类常用机器学习算法,优化后的损失函数降低了不稳定样本错误分类的可能性,使该算法的召回率较优于其他方法,且概率化输出的形式有助于评估模型输出的可靠程度,降低了误预测的概率。
The TSA(Transient Stability Assessment) is carried out based on the improved XGBoost(eXtreme Gradient Boosting) algorithm according to the characteristics of TSA problem. The feature set that can reflect the stable operation state of power system is defined based on the physical characteristics of power grid. The TSA process with XGBoost is researched. Since the severities of two mistakes in TSA process are different,a logarithmic loss function containing the attention coefficient is defined to reduce the mistaken prediction situations of the model to unstable samples. The Logistic function is introduced to transform the model output into a probabilistic form,which is used to measure the reliability of model output to prevent part of mistaken prediction. A method to generate sample sets for any power system is given. The simulative results of IEEE 39-bus power system show that the accuracy of XGBoost is higher than that of other commonly used machine learning methods,the misclassification probability of unstable samples is reduced by the loss function after optimization,which makes the recall rate of the proposed algorithm better than other methods,and the probabilistic output format is helpful for evaluating the reliability of model output and reduces the probability of mistaken prediction.
The TSA(Transient Stability Assessment) is carried out based on the improved XGBoost(eXtreme Gradient Boosting) algorithm according to the characteristics of TSA problem. The feature set that can reflect the stable operation state of power system is defined based on the physical characteristics of power grid. The TSA process with XGBoost is researched. Since the severities of two mistakes in TSA process are different,a logarithmic loss function containing the attention coefficient is defined to reduce the mistaken prediction situations of the model to unstable samples. The Logistic function is introduced to transform the model output into a probabilistic form,which is used to measure the reliability of model output to prevent part of mistaken prediction. A method to generate sample sets for any power system is given. The simulative results of IEEE 39-bus power system show that the accuracy of XGBoost is higher than that of other commonly used machine learning methods,the misclassification probability of unstable samples is reduced by the loss function after optimization,which makes the recall rate of the proposed algorithm better than other methods,and the probabilistic output format is helpful for evaluating the reliability of model output and reduces the probability of mistaken prediction.
引文
[1]MAKAROV Y V,RESHETOV V I,STROEV A,et al. Blackout prevention in the United States,Europe,and Russia[J]. Proceedings of the IEEE,2005,93(11):1942-1955.
[2]王同文,管霖,张尧.人工智能技术在电网稳定评估中的应用综述[J].电网技术,2009,33(12):60-65.WANG Tongwen,GUAN Lin,ZHANG Yao. A survey on application of artificial intelligence technology in power system stability assessment[J]. Power System Technology,2009,33(12):60-65.
[3]鞠平,周孝信,陈维江,等.“智能电网+”研究综述[J].电力自动化设备,2018,38(5):2-11.JU Ping,ZHOU Xiaoxin,CHEN Weijiang,et al.“Smart Grid Plus”research overview[J]. Electric Power Automation Equipment,2018,38(5):2-11.
[4]段青,赵建国,马艳.基于稀疏贝叶斯学习的电力系统暂态稳定评估[J].电力自动化设备,2009,29(9):36-40.DUAN Qing,ZHAO Jianguo,MA Yan. Power systems transient stability assessment based on sparse Bayesian learning[J]. Electric Power Automation Equipment,2009,29(9):36-40.
[5]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.
[6]陈振,肖先勇,李长松,等.基于代价敏感极端学习机的电力系统暂态稳定评估方法[J].电力自动化设备,2016,36(2):118-123.CHEN Zhen,XIAO Xianyong,LI Changsong,et al. Power system transient stability assessment based on cost-sensitive extreme learning machine[J]. Electric Power Automation Equipment,2016,36(2):118-123.
[7]MOULIN L S,ALVESD S A P,EL-SHARKAWA M A,et al. Support vector machines for transient stability analysis of large-scale power systems[J]. IEEE Transactions on Power Systems,2004,19(2):818-825.
[8]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.
[9]XU Y,DONG Z Y,ZHAO J H,et al. A reliable intelligent system for real-time dynamic security assessment of power systems[J].IEEE Transactions on Power Systems,2012,27(3):1253-1263.
[10]DAI Y,CHEN L,ZHANG W,et al. Multi-support vector machine power system transient stability assessment based on relief algorithm[C]∥Power and Energy Engineering Conference. Brisbane,Australia:[s.n.],2016:1-5.
[11]胡伟,郑乐,闵勇,等.基于深度学习的电力系统故障后暂态稳定评估研究[J].电网技术,2017,41(10):3140-3146.HU Wei,ZHENG Le,MIN Yong,et al. Research on power system transient stability assessment based on deep learning of big data technique[J]. Power System Technology,2017,41(10):3140-3146.
[12]管霖,何楚瑶,曾毅豪,等.基于模式匹配的智能稳定评估方法[J].电力自动化设备,2016,36(11):107-111.GUAN Lin,HE Chuyao,ZENG Yihao,et al. Intelligent stability assessment based on pattern recognition[J]. Electric Power Automation Equipment,2016,36(11):107-111.
[13]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.
[14]CHEN T,GUESTRIN C. XGBoost:a scalable tree boosting system[C]∥Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco,USA:[s.n.],2016:85-97.
[15]XU Bing,ZHU Nan. Awesome XGBoost[EB/OL].[2017-09-20].https:∥github. com/dmlc/xgboost/tree/master/demo#machine-learning-challenge-winning-solutions.
[16]王锡凡.现代电力系统分析[M].北京:科学出版社,2003:292-295.
[17]BREIMAN L I,FRIEDMAN J H,OLSHEN R A,et al. Classification And Regression Trees(CART)[J]. Encyclopedia of Ecology,1984,40(3):582-588.
[18]ZIMMERMANRD,MURILLO-SANCHEZ C E,THOMAS R J. Matpower:steady-state operations,planning,and analysis tools for power systems research and education[J]. IEEE Transactions on Power Systems,2011,26(1):12-19.
[19]VANFRETTI L,MILANO F. Application of the PSAT,an open source software,for educational and research purposes[C]∥Power Engineering Society General Meeting. Tampa,USA:[s.n.],2007:24-28.
[20] BROWNE M W. Cross-validation methods[M]. Salt Lake City,USA:Academic Press,Inc,2000:1-49.
[21] PEDREGOSA F,GRAMFORT A,MICHEL V,et al. Scikit-learn:ma-chine learning in Python[J]. Journal of Machine Learning Research,2016,12(10):2825-2830.
[2]王同文,管霖,张尧.人工智能技术在电网稳定评估中的应用综述[J].电网技术,2009,33(12):60-65.WANG Tongwen,GUAN Lin,ZHANG Yao. A survey on application of artificial intelligence technology in power system stability assessment[J]. Power System Technology,2009,33(12):60-65.
[3]鞠平,周孝信,陈维江,等.“智能电网+”研究综述[J].电力自动化设备,2018,38(5):2-11.JU Ping,ZHOU Xiaoxin,CHEN Weijiang,et al.“Smart Grid Plus”research overview[J]. Electric Power Automation Equipment,2018,38(5):2-11.
[4]段青,赵建国,马艳.基于稀疏贝叶斯学习的电力系统暂态稳定评估[J].电力自动化设备,2009,29(9):36-40.DUAN Qing,ZHAO Jianguo,MA Yan. Power systems transient stability assessment based on sparse Bayesian learning[J]. Electric Power Automation Equipment,2009,29(9):36-40.
[5]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.
[6]陈振,肖先勇,李长松,等.基于代价敏感极端学习机的电力系统暂态稳定评估方法[J].电力自动化设备,2016,36(2):118-123.CHEN Zhen,XIAO Xianyong,LI Changsong,et al. Power system transient stability assessment based on cost-sensitive extreme learning machine[J]. Electric Power Automation Equipment,2016,36(2):118-123.
[7]MOULIN L S,ALVESD S A P,EL-SHARKAWA M A,et al. Support vector machines for transient stability analysis of large-scale power systems[J]. IEEE Transactions on Power Systems,2004,19(2):818-825.
[8]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.
[9]XU Y,DONG Z Y,ZHAO J H,et al. A reliable intelligent system for real-time dynamic security assessment of power systems[J].IEEE Transactions on Power Systems,2012,27(3):1253-1263.
[10]DAI Y,CHEN L,ZHANG W,et al. Multi-support vector machine power system transient stability assessment based on relief algorithm[C]∥Power and Energy Engineering Conference. Brisbane,Australia:[s.n.],2016:1-5.
[11]胡伟,郑乐,闵勇,等.基于深度学习的电力系统故障后暂态稳定评估研究[J].电网技术,2017,41(10):3140-3146.HU Wei,ZHENG Le,MIN Yong,et al. Research on power system transient stability assessment based on deep learning of big data technique[J]. Power System Technology,2017,41(10):3140-3146.
[12]管霖,何楚瑶,曾毅豪,等.基于模式匹配的智能稳定评估方法[J].电力自动化设备,2016,36(11):107-111.GUAN Lin,HE Chuyao,ZENG Yihao,et al. Intelligent stability assessment based on pattern recognition[J]. Electric Power Automation Equipment,2016,36(11):107-111.
[13]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.
[14]CHEN T,GUESTRIN C. XGBoost:a scalable tree boosting system[C]∥Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco,USA:[s.n.],2016:85-97.
[15]XU Bing,ZHU Nan. Awesome XGBoost[EB/OL].[2017-09-20].https:∥github. com/dmlc/xgboost/tree/master/demo#machine-learning-challenge-winning-solutions.
[16]王锡凡.现代电力系统分析[M].北京:科学出版社,2003:292-295.
[17]BREIMAN L I,FRIEDMAN J H,OLSHEN R A,et al. Classification And Regression Trees(CART)[J]. Encyclopedia of Ecology,1984,40(3):582-588.
[18]ZIMMERMANRD,MURILLO-SANCHEZ C E,THOMAS R J. Matpower:steady-state operations,planning,and analysis tools for power systems research and education[J]. IEEE Transactions on Power Systems,2011,26(1):12-19.
[19]VANFRETTI L,MILANO F. Application of the PSAT,an open source software,for educational and research purposes[C]∥Power Engineering Society General Meeting. Tampa,USA:[s.n.],2007:24-28.
[20] BROWNE M W. Cross-validation methods[M]. Salt Lake City,USA:Academic Press,Inc,2000:1-49.
[21] PEDREGOSA F,GRAMFORT A,MICHEL V,et al. Scikit-learn:ma-chine learning in Python[J]. Journal of Machine Learning Research,2016,12(10):2825-2830.