基于不完全S变换与梯度提升树的电能质量复合扰动识别
|
摘要
针对电能质量复合扰动中特征选择困难和分类准确率不高的问题,提出基于不完全S变换和梯度提升树的特征选择和分类器构建方法。首先通过选取特定频率的不完全S变换得到扰动的时频矩阵。再从时频矩阵中提取53种原始特征量,并基于梯度提升树对各个特征的重要性进行度量,选取重要特征。最后根据选取的特征集训练和构建梯度提升树,得到扰动分类器。仿真实验表明,对于包括8种复合扰动在内的共17种扰动类型,该方法的分类准确率高于CART决策树、随机森林(RF)、多层感知机(MLP)等现有方法。不同噪声条件下的分类结果表明,该方法具有良好的抗噪性能和算法鲁棒性,展现出良好的应用前景。
In order to solve the problem of feature selection and classification accuracy in complex power quality disturbances recognition, a method of feature selection and classifier construction based on incomplete S-transform and Gradient Boosting Decision Tree(GBDT) is proposed. Firstly, the time-frequency matrixes of the disturbances are obtained by the incomplete S-transform at the specific frequencies. Then the original feature set of 53 features is extracted from the time-frequency matrix, and the importance level of each feature is measured by GBDT to select optimal features. Finally, the disturbance classifier is trained and constructed by GBDT based on the selected feature set. The simulation results show that the classification accuracy of the proposed method is higher than the CART, the Random Forest(RF), the Multilayer Perceptron(MLP), and so on, for a total of 17 disturbance classes including 8 classes of combined disturbances. Moreover, the classification results in different noise conditions indicate that the method has good anti-noise performance and algorithm robustness, which shows good prospects for application.
In order to solve the problem of feature selection and classification accuracy in complex power quality disturbances recognition, a method of feature selection and classifier construction based on incomplete S-transform and Gradient Boosting Decision Tree(GBDT) is proposed. Firstly, the time-frequency matrixes of the disturbances are obtained by the incomplete S-transform at the specific frequencies. Then the original feature set of 53 features is extracted from the time-frequency matrix, and the importance level of each feature is measured by GBDT to select optimal features. Finally, the disturbance classifier is trained and constructed by GBDT based on the selected feature set. The simulation results show that the classification accuracy of the proposed method is higher than the CART, the Random Forest(RF), the Multilayer Perceptron(MLP), and so on, for a total of 17 disturbance classes including 8 classes of combined disturbances. Moreover, the classification results in different noise conditions indicate that the method has good anti-noise performance and algorithm robustness, which shows good prospects for application.
引文
[1]LI J,TENG Z,TANG Q,et al.Detection and classification of power quality disturbances using double resolution S-transform and DAG-SVMs[J].IEEETransactions on Instrumentation and Measurement,2016,65(10):2302-2312.
[2]KHOKHAR S,ZIN A A B M,MOKHTAR A S B,et al.Acomprehensive overview on signal processing and artificial intelligence techniques applications in classification of power quality disturbances[J].Renewable and Sustainable Energy Reviews,2015,51:1650-1663.
[3]黄建明,瞿合祚,李晓明.基于短时傅里叶变换及其谱峭度的电能质量混合扰动分类[J].电网技术,2016,40(10):3184-3191.HUANG Jianming,QU Hezuo,LI Xiaoming.Classification for hybrid power quality disturbance based on STFT and its spectral kurtosis[J].Power System Technology,2016,40(10):3184-3191.
[4]任子晖,刘昊岳,徐进霞.基于小波变换和改进Prony方法的电能质量扰动分析[J].电力系统保护与控制,2016,44(9):122-128.REN Zihui,LIU Haoyue,XU Jinxia.Power quality disturbance analysis based on wavelet transform and improved Prony method[J].Power System Technology,2016,44(9):122-128.
[5]田振果,傅成华,吴浩,等.基于HHT的电能质量扰动定位与分类[J].电力系统保护与控制,2015,43(16):36-42.TIAN Zhenguo,FU Chenghua,WU Hao,et al.Power quality disturbance for location and classification based on HHT[J].Power System Protection and Control,2015,43(16):36-42.
[6]郭俊文,李开成,何顺帆,等.基于改进不完全S变换与决策树的实时电能质量扰动分类[J].电力系统保护与控制,2013,41(22):103-110.GUO Junwen,LI Kaicheng,HE Shunfan,et al.A real time power quality disturbance classification based on improved incomplete S-transform and decision tree[J].Power System Protection and Control,2013,41(22):103-110.
[7]BISWAL M,DASH P K.Detection and characterization of multiple power quality disturbances with a fast S-transform and decision tree based classifier[J].Digital Signal Processing,2013,23(4):1071-1083.
[8]KUMAR R,SINGH B,SHAHANI D T,et al.Recognition of power-quality disturbances using S-transform-based ANN classifier and rule-based decision tree[J].IEEETransactions on Industry Applications,2015,51(2):1249-1258.
[9]XUE B,ZHANG M,BROWNE W N,et al.A survey on evolutionary computation approaches to feature selection[J].IEEE Transactions on Evolutionary Computation,2016,20(4):606-626.
[10]AHILA R,SADASIVAM V,MANIMALA K.An integrated PSO for parameter determination and feature selection of ELM and its application in classification of power system disturbances[J].Applied Soft Computing,2015,32:23-37.
[11]SINGH U,SINGH S N.Optimal feature selection via NSGA-II for power quality disturbances classification[J].IEEE Transactions on Industrial Informatics,2017,14(7):2994-3002.
[12]黄南天,彭华,蔡国伟,等.电能质量复合扰动特征选择与最优决策树构建[J].中国电机工程学报,2017,37(3):776-786.HUANG Nantian,PENG Hua,CAI Guowei,et al.Feature selection and optimal decision tree construction of complex power quality disturbances[J].Proceedings of the CSEE,2017,37(3):776-786.
[13]刘志刚,张巧革,张杨.电能质量复合扰动分类的研究进展[J].电力系统保护与控制,2013,41(13):146-153.LIU Zhigang,ZHANG Qiaoge,ZHANG Yang.Review of power quality mixed disturbances identification[J].Power System Protection and Control,2013,41(13):146-153.
[14]NAIK C A,KUNDU P.Power quality disturbance classification employing S-transform and three-module artificial neural network[J].International Transactions on Electrical Energy Systems,2014,24(9):1301-1322.
[15]何巨龙,王根平,刘丹,等.基于提升小波和改进BP神经网络的配电网系统电能质量扰动定位与识别[J].电力系统保护与控制,2017,45(10):69-76.HE Julong,WANG Genping,LIU Dan,et al.Localization and identification of power quality disturbance in distribution network system based on lifting wavelet and improved BP neural network[J].Power System Protection and Control,2017,45(10):69-76.
[16]陈晓静,李开成,肖剑,等.一种实时电能质量扰动分类方法[J].电工技术学报,2017,32(3):45-55.CHEN Xiaojing,LI Kaicheng,XIAO Jian,et al.A method of real-time power quality disturbance classification[J].Transactions of China Electrotechnical Society,2017,32(3):45-55.
[17]黄南天,徐殿国,刘晓胜.基于S变换与SVM的电能质量复合扰动识别[J].电工技术学报,2011,26(10):23-30.HUANG Nantian,XU Dianguo,LIU Xiaosheng.Identification of power quality complex disturbances based on S-transform and SVM[J].Transactions of China Electrotechnical Society,2011,26(10):23-30.
[18]张巧革,刘志刚,朱玲,等.基于多标签Rank-WSVM的复合电能质量扰动分类[J].中国电机工程学报,2013,33(28):114-120.ZHANG Qiaoge,LIU Zhigang,ZHU Ling,et al.Recognition of multiple power quality disturbances using multi-label wavelet support vector machine[J].Proceedings of the CSEE,2013,33(28):114-120.
[19]王凌云,李开成,肖厦颖,等.基于稀疏分解的复合电能质量扰动分类[J].电测与仪表,2018,55(1):14-20.WANG Lingyun,LI Kaicheng,XIAO Xiaying,et al.Classification for multiple power quality disturbances based on sparse decomposition[J].Electrical Measurement&Instrumentation 2018,55(1):14-20.
[20]KE G,MENG Q,FINLEY T,et al.LightGBM:a highly efficient gradient boosting decision tree[C]//Advances in Neural Information Processing Systems,2017:3149-3157.
[21]CHEN T,GUESTRIN C.Xgboost:A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDDInternational Conference on Knowledge Discovery and Data Mining,ACM,2016:785-794.
[22]PANIGRAHI B K,PANDI V R.Optimal feature selection for classification of power quality disturbances using wavelet packet-based fuzzy k-nearest neighbour algorithm[J].IET Generation,Transmission&Distribution,2009,3(3):296-306.
[23]JAMALI S,FARSA A R,GHAFFARZADEH N.Identification of optimal features for fast and accurate classification of power quality disturbances[J].Measurement,2018,116:565-574.
[24]瞿合祚,刘恒,李晓明,等.基于多标签随机森林的电能质量复合扰动分类方法[J].电力系统保护与控制,2017,45(11):1-7.QU Hezuo,LIU Heng,LI Xiaoming,et al.Recognition of multiple power quality disturbances using multi-label random forest[J].Power System Protection and Control,2017,45(11):1-7.
[25]FERREIRA D D,DE SEIXAS J M,CERQUEIRA A S.Amethod based on independent component analysis for single and multiple power quality disturbance classification[J].Electric Power Systems Research,2015,119:425-431.
[2]KHOKHAR S,ZIN A A B M,MOKHTAR A S B,et al.Acomprehensive overview on signal processing and artificial intelligence techniques applications in classification of power quality disturbances[J].Renewable and Sustainable Energy Reviews,2015,51:1650-1663.
[3]黄建明,瞿合祚,李晓明.基于短时傅里叶变换及其谱峭度的电能质量混合扰动分类[J].电网技术,2016,40(10):3184-3191.HUANG Jianming,QU Hezuo,LI Xiaoming.Classification for hybrid power quality disturbance based on STFT and its spectral kurtosis[J].Power System Technology,2016,40(10):3184-3191.
[4]任子晖,刘昊岳,徐进霞.基于小波变换和改进Prony方法的电能质量扰动分析[J].电力系统保护与控制,2016,44(9):122-128.REN Zihui,LIU Haoyue,XU Jinxia.Power quality disturbance analysis based on wavelet transform and improved Prony method[J].Power System Technology,2016,44(9):122-128.
[5]田振果,傅成华,吴浩,等.基于HHT的电能质量扰动定位与分类[J].电力系统保护与控制,2015,43(16):36-42.TIAN Zhenguo,FU Chenghua,WU Hao,et al.Power quality disturbance for location and classification based on HHT[J].Power System Protection and Control,2015,43(16):36-42.
[6]郭俊文,李开成,何顺帆,等.基于改进不完全S变换与决策树的实时电能质量扰动分类[J].电力系统保护与控制,2013,41(22):103-110.GUO Junwen,LI Kaicheng,HE Shunfan,et al.A real time power quality disturbance classification based on improved incomplete S-transform and decision tree[J].Power System Protection and Control,2013,41(22):103-110.
[7]BISWAL M,DASH P K.Detection and characterization of multiple power quality disturbances with a fast S-transform and decision tree based classifier[J].Digital Signal Processing,2013,23(4):1071-1083.
[8]KUMAR R,SINGH B,SHAHANI D T,et al.Recognition of power-quality disturbances using S-transform-based ANN classifier and rule-based decision tree[J].IEEETransactions on Industry Applications,2015,51(2):1249-1258.
[9]XUE B,ZHANG M,BROWNE W N,et al.A survey on evolutionary computation approaches to feature selection[J].IEEE Transactions on Evolutionary Computation,2016,20(4):606-626.
[10]AHILA R,SADASIVAM V,MANIMALA K.An integrated PSO for parameter determination and feature selection of ELM and its application in classification of power system disturbances[J].Applied Soft Computing,2015,32:23-37.
[11]SINGH U,SINGH S N.Optimal feature selection via NSGA-II for power quality disturbances classification[J].IEEE Transactions on Industrial Informatics,2017,14(7):2994-3002.
[12]黄南天,彭华,蔡国伟,等.电能质量复合扰动特征选择与最优决策树构建[J].中国电机工程学报,2017,37(3):776-786.HUANG Nantian,PENG Hua,CAI Guowei,et al.Feature selection and optimal decision tree construction of complex power quality disturbances[J].Proceedings of the CSEE,2017,37(3):776-786.
[13]刘志刚,张巧革,张杨.电能质量复合扰动分类的研究进展[J].电力系统保护与控制,2013,41(13):146-153.LIU Zhigang,ZHANG Qiaoge,ZHANG Yang.Review of power quality mixed disturbances identification[J].Power System Protection and Control,2013,41(13):146-153.
[14]NAIK C A,KUNDU P.Power quality disturbance classification employing S-transform and three-module artificial neural network[J].International Transactions on Electrical Energy Systems,2014,24(9):1301-1322.
[15]何巨龙,王根平,刘丹,等.基于提升小波和改进BP神经网络的配电网系统电能质量扰动定位与识别[J].电力系统保护与控制,2017,45(10):69-76.HE Julong,WANG Genping,LIU Dan,et al.Localization and identification of power quality disturbance in distribution network system based on lifting wavelet and improved BP neural network[J].Power System Protection and Control,2017,45(10):69-76.
[16]陈晓静,李开成,肖剑,等.一种实时电能质量扰动分类方法[J].电工技术学报,2017,32(3):45-55.CHEN Xiaojing,LI Kaicheng,XIAO Jian,et al.A method of real-time power quality disturbance classification[J].Transactions of China Electrotechnical Society,2017,32(3):45-55.
[17]黄南天,徐殿国,刘晓胜.基于S变换与SVM的电能质量复合扰动识别[J].电工技术学报,2011,26(10):23-30.HUANG Nantian,XU Dianguo,LIU Xiaosheng.Identification of power quality complex disturbances based on S-transform and SVM[J].Transactions of China Electrotechnical Society,2011,26(10):23-30.
[18]张巧革,刘志刚,朱玲,等.基于多标签Rank-WSVM的复合电能质量扰动分类[J].中国电机工程学报,2013,33(28):114-120.ZHANG Qiaoge,LIU Zhigang,ZHU Ling,et al.Recognition of multiple power quality disturbances using multi-label wavelet support vector machine[J].Proceedings of the CSEE,2013,33(28):114-120.
[19]王凌云,李开成,肖厦颖,等.基于稀疏分解的复合电能质量扰动分类[J].电测与仪表,2018,55(1):14-20.WANG Lingyun,LI Kaicheng,XIAO Xiaying,et al.Classification for multiple power quality disturbances based on sparse decomposition[J].Electrical Measurement&Instrumentation 2018,55(1):14-20.
[20]KE G,MENG Q,FINLEY T,et al.LightGBM:a highly efficient gradient boosting decision tree[C]//Advances in Neural Information Processing Systems,2017:3149-3157.
[21]CHEN T,GUESTRIN C.Xgboost:A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDDInternational Conference on Knowledge Discovery and Data Mining,ACM,2016:785-794.
[22]PANIGRAHI B K,PANDI V R.Optimal feature selection for classification of power quality disturbances using wavelet packet-based fuzzy k-nearest neighbour algorithm[J].IET Generation,Transmission&Distribution,2009,3(3):296-306.
[23]JAMALI S,FARSA A R,GHAFFARZADEH N.Identification of optimal features for fast and accurate classification of power quality disturbances[J].Measurement,2018,116:565-574.
[24]瞿合祚,刘恒,李晓明,等.基于多标签随机森林的电能质量复合扰动分类方法[J].电力系统保护与控制,2017,45(11):1-7.QU Hezuo,LIU Heng,LI Xiaoming,et al.Recognition of multiple power quality disturbances using multi-label random forest[J].Power System Protection and Control,2017,45(11):1-7.
[25]FERREIRA D D,DE SEIXAS J M,CERQUEIRA A S.Amethod based on independent component analysis for single and multiple power quality disturbance classification[J].Electric Power Systems Research,2015,119:425-431.