基于决策树的智能变电站运维专家系统规则提取方法
|
摘要
针对智能变电站运维专家系统存在规则制定严重依赖人类专家与规则扩充困难的问题,提出一种基于决策树算法的智能变电站专家系统规则提取方法。该方法利用决策树的归纳学习能力,在变电站运行规程实例中通过自动学习可提取得到专家系统的运维规则。专家系统利用提取出的规则,可对任意智能站进行合理推导得到完整运行规程,也可将规则用于智能变电站在线辅助运维。最后,基于运行规程实例,将提取出的规则应用于某500 kV实际智能变电站,仿真结果验证了提取过程的高效、以及规则的通用性、准确性、全面性。
In order to solve the problem that the rules setting are heavily dependent on human experts and limitation of rules expansion for the operation and maintenance of intelligent substation, a decision extraction method based on decision tree algorithm is proposed for intelligent substation expert system. This method utilizes the inductive learning ability of decision tree, and the operation and maintenance rules of the expert system can then be extracted through automatic learning in the example of substation operation procedure. These rules could also be utilized in online decision making system to help substation staff monitor the operation and maintenance of substations. At the end, a 500 kV smart substation is included as an example to present the process and verify versatility, accuracy and comprehensiveness of the proposed method.
In order to solve the problem that the rules setting are heavily dependent on human experts and limitation of rules expansion for the operation and maintenance of intelligent substation, a decision extraction method based on decision tree algorithm is proposed for intelligent substation expert system. This method utilizes the inductive learning ability of decision tree, and the operation and maintenance rules of the expert system can then be extracted through automatic learning in the example of substation operation procedure. These rules could also be utilized in online decision making system to help substation staff monitor the operation and maintenance of substations. At the end, a 500 kV smart substation is included as an example to present the process and verify versatility, accuracy and comprehensiveness of the proposed method.
引文
[1]笃峻,叶翔,葛立青,等.智能变电站继电保护在线运维系统关键技术的研究及实现[J].电力自动化设备,2016,36(7):163-175.DU Jun,YE Xiang,GE Li-qing,et al.Key technologies of online maintenance system for relay protections in smart substation and its implementation[J].Electric Power Automation Equipment,2016,36(7):163-175.
[2]刘伟,李江林,杨恢宏,等.智能变电站智能告警与辅助决策的实现[J].电力系统保护与控制,2011,39(15):146-150.LIU Wei,LI jiang-lin,YANG Hui-hong,et al.Implementation of intelligent alarm and AMD system in the smart substation[J].Power System Protection and Control,2011,39(15):146-150.
[3]王玉磊,应黎明,陆海洋,等.基于效能—成本的智能变电站二次设备运维策略优化[J].电力自动化设备,2016,36(6):182-188.WANG Yu-lei,YING Li-ming,LU Hai-yang,et al.Operation maintenance strategy based on efficiency-cost sensitivity analysis for secondary equipment of intelligent substation[J].Electric Power Automation Equipment,2016,36(6):182-188.
[4]王保民.智能变电站故障信息综合分析决策系统的研究与应用[D].保定:华北电力大学,2012.
[5]孙宏斌,王康,张伯明,等.采用线性决策树的暂态稳定规则提取[J].中国电机工程学报,2011,31(34):61-67.SUN Hong-bin,WANG Kang,ZHANG Bo-ming,et al.Rule extraction in transient stability study using linear decision trees[J].Proceedings of the CSEE,2011,31(34):61-67.
[6]王同文,管霖.基于模式发现的电力系统稳定评估和规则提取[J].中国电机工程学报,2007,27(19):25-31.WANG Tong-wen,GUAN Lin.Power system stability assessment and rule extraction based on pattern discovery[J].Proceeding of the CSEE,2007,27(19):25-31.
[7]Leszek Rutkowski,Maciej Jaworski,Lena Pietruczuk,et al.The CART decision tree for mining data streams[J].Information Sciences,2014,266:1-15.
[8]Suarez A,Lutsko J F.Globally optimal fuzzy decision trees for classification and regression[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1999,21(12):1 297-1 311.
[9]Motohide Umano,Hirotaka Okamoto,Itsuo Hatono,et al.Fuzzy decision tree by fuzzy ID3 algorithm and its ap-plication to diagnosis system[C].IEEE World Con-gress on Computational Intelligence,Orlando,America,1994.
[10]赖邦传,陈晓红.一种基于决策树的多属性分类方法[J].计算机工程,2005,31(5):88-89,226.LAI Bang-chuan,CHEN Xiao-hong.A method for multi-attribute classifying based on decision tree[J].Computer Engineering,2005,31(5):88-89,226.
[2]刘伟,李江林,杨恢宏,等.智能变电站智能告警与辅助决策的实现[J].电力系统保护与控制,2011,39(15):146-150.LIU Wei,LI jiang-lin,YANG Hui-hong,et al.Implementation of intelligent alarm and AMD system in the smart substation[J].Power System Protection and Control,2011,39(15):146-150.
[3]王玉磊,应黎明,陆海洋,等.基于效能—成本的智能变电站二次设备运维策略优化[J].电力自动化设备,2016,36(6):182-188.WANG Yu-lei,YING Li-ming,LU Hai-yang,et al.Operation maintenance strategy based on efficiency-cost sensitivity analysis for secondary equipment of intelligent substation[J].Electric Power Automation Equipment,2016,36(6):182-188.
[4]王保民.智能变电站故障信息综合分析决策系统的研究与应用[D].保定:华北电力大学,2012.
[5]孙宏斌,王康,张伯明,等.采用线性决策树的暂态稳定规则提取[J].中国电机工程学报,2011,31(34):61-67.SUN Hong-bin,WANG Kang,ZHANG Bo-ming,et al.Rule extraction in transient stability study using linear decision trees[J].Proceedings of the CSEE,2011,31(34):61-67.
[6]王同文,管霖.基于模式发现的电力系统稳定评估和规则提取[J].中国电机工程学报,2007,27(19):25-31.WANG Tong-wen,GUAN Lin.Power system stability assessment and rule extraction based on pattern discovery[J].Proceeding of the CSEE,2007,27(19):25-31.
[7]Leszek Rutkowski,Maciej Jaworski,Lena Pietruczuk,et al.The CART decision tree for mining data streams[J].Information Sciences,2014,266:1-15.
[8]Suarez A,Lutsko J F.Globally optimal fuzzy decision trees for classification and regression[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1999,21(12):1 297-1 311.
[9]Motohide Umano,Hirotaka Okamoto,Itsuo Hatono,et al.Fuzzy decision tree by fuzzy ID3 algorithm and its ap-plication to diagnosis system[C].IEEE World Con-gress on Computational Intelligence,Orlando,America,1994.
[10]赖邦传,陈晓红.一种基于决策树的多属性分类方法[J].计算机工程,2005,31(5):88-89,226.LAI Bang-chuan,CHEN Xiao-hong.A method for multi-attribute classifying based on decision tree[J].Computer Engineering,2005,31(5):88-89,226.