基于改进能量集中度的S变换与随机森林的电能质量扰动识别
|
摘要
鉴于S变换的窗口函数对不同频带信号的自适应能力差,提出一种新型的改进S变换(Modified S-Transform,MST),该方法通过引入四个辅助参数,优化高斯窗函数尺度因子的自适应能力,使改进S变换的能量集中度最大化,获得了更出色的时频分辨能力。建立了基于扰动信号幅值和相位的特征值评价体系,采用随机森林(Random Forest,RF)算法对包括标准信号和电压暂降、电压暂升、高次谐波、暂态振荡等10种扰动信号共11类电能质量信号分类识别。与已有文献采用的决策树、支持向量机和神经网络分类结果进行了对比分析,仿真试验结果表明,该方法分类准确率高,抗干扰能力强,且在训练样本少、低信噪比(Signal-to-Noise Radio,SNR)条件下分类结果优势明显。
In view of the poor adaptability of the S-transform window function to different frequency band signals,a new modified S-transform( MST) is proposed in this paper. This method optimizes the Gaussian window function scale factor by introducing four auxiliary parameters. Adaptability,it maximizes the energy concentration of the improved S-transformation,and achieves better time-frequency resolution. The eigenvalue evaluation system based on the amplitude and phase of the disturbance signal is established. The random forest( RF) algorithm is used to include 10 kinds of disturbance signals such as standard signal and voltage sag,voltage spurt,higher harmonic and transient oscillation. A total of 11 types of power quality signals are classified and identified. Compared with the decision tree,support vector machine and neural network classification results are used in the existing literature,the simulation results show that the method has high classification accuracy,strong anti-interference ability,and few training samples and low signal-to-noise ratio. Signal-tonoise radio( SNR) conditions have obvious advantages in classification results.
In view of the poor adaptability of the S-transform window function to different frequency band signals,a new modified S-transform( MST) is proposed in this paper. This method optimizes the Gaussian window function scale factor by introducing four auxiliary parameters. Adaptability,it maximizes the energy concentration of the improved S-transformation,and achieves better time-frequency resolution. The eigenvalue evaluation system based on the amplitude and phase of the disturbance signal is established. The random forest( RF) algorithm is used to include 10 kinds of disturbance signals such as standard signal and voltage sag,voltage spurt,higher harmonic and transient oscillation. A total of 11 types of power quality signals are classified and identified. Compared with the decision tree,support vector machine and neural network classification results are used in the existing literature,the simulation results show that the method has high classification accuracy,strong anti-interference ability,and few training samples and low signal-to-noise ratio. Signal-tonoise radio( SNR) conditions have obvious advantages in classification results.
引文
[1]陈华丰,张葛祥.基于决策树和支持向量机的电能质量扰动识别[J].电网技术,2013,37(5):1272-1278.Chen Huafeng,Zhang Gexiang.Power quality disturbance identification using decision tree and support vector machine[J].Power System Technology,2013,37(5):1272-1278.
[2]徐永海,赵燕.基于短时傅里叶变换的电能质量扰动识别与采用奇异值分解的扰动时间定位[J].电网技术,2011,35(8):174-180.Xu Yonghai,Zhao Yan.Identification of power quality disturbance based on short-term fourier transform and disturbance time orientation by singular value decomposition[J].Power System Technology,2011,35(8):174-179.
[3]林涛,樊正伟.利用小波变换及人工神经网络识别电能扰动[J].高电压技术,2007,(7):151-153.Lin Tao,Fan Zhengwei.Application of Wavelet Transform and Artificial Neural Network to Power Disturbance Identification[J].High Voltage Engineering,2007,(7):151-153.
[4]Sundaram P K N R.Analysis and Classification of Power Quality Events using Hilbert Transform and Fuzzy system[J].Sensing,Signal Processing and Security(ICSSS),2017 Third International Conference on.IEEE,2017:269-274.
[5]Stockwell R G M L L R.Localization of the Complex Spectrum The STransform[J].IEEE transactions on signal processing,1996,44(4):998-1001.
[6]易吉良,彭建春,谭会生.基于改进S变换估计电能质量信号的噪声水平[J].电网技术,2009,(20):71-75.Yi Jiliang,Peng Jianchun,Tan Huisheng.Noise Estimation of Power Quality Signal Based on Modified S-Transform[J].Power System Technology,2009,(20):71-75.
[7]尹柏强,何怡刚,朱彦卿.一种广义S变换及模糊SOM网络的电能质量多扰动检测和识别方法[J].中国电机工程学报,2015,35(4):866-872.Yin Baiqiang,He Yigang,Zhu Yanqing.Detection and classification of power quality Multi-disturbances based on generalized S-transform and fuzzy SOM neural network[J].Proceedings of the CSEE,2015,35(4):866-872.
[8]徐志超,杨玲君,李晓明.基于聚类改进S变换与直接支持向量机的电能质量扰动识别[J].电力自动化设备,2015,(7):50-58.Xu Zhichao,Yang Lingjun,Li Xiaoming.Power quality disturbance identification based on clustering-modified S-transform and direct support vector machine[J].Electric Power Automation Equipment,2015,(7):50-58.
[9]Liaw A,M W.Classification and Regression by RandomForest[J].RNews,2002,3(2):18-22.
[10]李钊.基于S变换的特高压直流输电线路快速保护研究[D].济南:山东大学,2016.
[11]Brown R A L M L F.A General Description of Linear Time-Frequency Transforms and Formulation of a Fast,Invertible Transform That Samples the Continuous S-Transform Spectrum Nonredundantly[J].IEEE Transactions on Signal Processing,2010,58(10):281-290.
[12]Moukadem A B Z A D.A new optimized Stockwell transform applied on synthetic and real non-stationary signals[J].2015,46:226-238.
[13]易吉良,彭建春.基于广义S变换的短时电能质量扰动信号分类[J].电网技术,2009,(5):22-27.Yi Jiliang,Peng Jianchun.Classification of Short-Time Power Quality Disturbance Signals Based on Generalized S-Transform[J].Power System Technology,2009,(5):22-27.
[14]张全明,刘会金.最小二乘支持向量机在电能质量扰动分类中的应用[J].中国电机工程学报,2008,28(1):106-110.Zhang Quanming,Liu Huijin.Application of LS-SVM in classification of power quality disturbances[J].Proceeding of the CSEE,2008,28(1):106-110.
[15]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].Transaction of China Electrotechnical Society,2011,26(10):23-30.
[16]方匡南.随机森林组合预测理论及其在金融中的应用[M].厦门大学出版社,2012.
[17]程学新.粒子群优化加权随机森林算法研究[D].郑州:郑州大学,2017.
[18]曹正凤.随机森林算法优化研究[D].北京:首都经济贸易大学,2014.
[19]姚建刚,郭知非,陈锦攀.基于小波和BP神经网络的电能扰动分类新方法[J].电网技术,2012,(5):139-144.Yao Jiangang,Guo Zhifei,Chen Jinpan.A New Approach to Recognize Power Quality Disturbances Based on Wavelet Transform and BP Neural Network[J].Power System Technology,2015,(5):139-144.
[2]徐永海,赵燕.基于短时傅里叶变换的电能质量扰动识别与采用奇异值分解的扰动时间定位[J].电网技术,2011,35(8):174-180.Xu Yonghai,Zhao Yan.Identification of power quality disturbance based on short-term fourier transform and disturbance time orientation by singular value decomposition[J].Power System Technology,2011,35(8):174-179.
[3]林涛,樊正伟.利用小波变换及人工神经网络识别电能扰动[J].高电压技术,2007,(7):151-153.Lin Tao,Fan Zhengwei.Application of Wavelet Transform and Artificial Neural Network to Power Disturbance Identification[J].High Voltage Engineering,2007,(7):151-153.
[4]Sundaram P K N R.Analysis and Classification of Power Quality Events using Hilbert Transform and Fuzzy system[J].Sensing,Signal Processing and Security(ICSSS),2017 Third International Conference on.IEEE,2017:269-274.
[5]Stockwell R G M L L R.Localization of the Complex Spectrum The STransform[J].IEEE transactions on signal processing,1996,44(4):998-1001.
[6]易吉良,彭建春,谭会生.基于改进S变换估计电能质量信号的噪声水平[J].电网技术,2009,(20):71-75.Yi Jiliang,Peng Jianchun,Tan Huisheng.Noise Estimation of Power Quality Signal Based on Modified S-Transform[J].Power System Technology,2009,(20):71-75.
[7]尹柏强,何怡刚,朱彦卿.一种广义S变换及模糊SOM网络的电能质量多扰动检测和识别方法[J].中国电机工程学报,2015,35(4):866-872.Yin Baiqiang,He Yigang,Zhu Yanqing.Detection and classification of power quality Multi-disturbances based on generalized S-transform and fuzzy SOM neural network[J].Proceedings of the CSEE,2015,35(4):866-872.
[8]徐志超,杨玲君,李晓明.基于聚类改进S变换与直接支持向量机的电能质量扰动识别[J].电力自动化设备,2015,(7):50-58.Xu Zhichao,Yang Lingjun,Li Xiaoming.Power quality disturbance identification based on clustering-modified S-transform and direct support vector machine[J].Electric Power Automation Equipment,2015,(7):50-58.
[9]Liaw A,M W.Classification and Regression by RandomForest[J].RNews,2002,3(2):18-22.
[10]李钊.基于S变换的特高压直流输电线路快速保护研究[D].济南:山东大学,2016.
[11]Brown R A L M L F.A General Description of Linear Time-Frequency Transforms and Formulation of a Fast,Invertible Transform That Samples the Continuous S-Transform Spectrum Nonredundantly[J].IEEE Transactions on Signal Processing,2010,58(10):281-290.
[12]Moukadem A B Z A D.A new optimized Stockwell transform applied on synthetic and real non-stationary signals[J].2015,46:226-238.
[13]易吉良,彭建春.基于广义S变换的短时电能质量扰动信号分类[J].电网技术,2009,(5):22-27.Yi Jiliang,Peng Jianchun.Classification of Short-Time Power Quality Disturbance Signals Based on Generalized S-Transform[J].Power System Technology,2009,(5):22-27.
[14]张全明,刘会金.最小二乘支持向量机在电能质量扰动分类中的应用[J].中国电机工程学报,2008,28(1):106-110.Zhang Quanming,Liu Huijin.Application of LS-SVM in classification of power quality disturbances[J].Proceeding of the CSEE,2008,28(1):106-110.
[15]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].Transaction of China Electrotechnical Society,2011,26(10):23-30.
[16]方匡南.随机森林组合预测理论及其在金融中的应用[M].厦门大学出版社,2012.
[17]程学新.粒子群优化加权随机森林算法研究[D].郑州:郑州大学,2017.
[18]曹正凤.随机森林算法优化研究[D].北京:首都经济贸易大学,2014.
[19]姚建刚,郭知非,陈锦攀.基于小波和BP神经网络的电能扰动分类新方法[J].电网技术,2012,(5):139-144.Yao Jiangang,Guo Zhifei,Chen Jinpan.A New Approach to Recognize Power Quality Disturbances Based on Wavelet Transform and BP Neural Network[J].Power System Technology,2015,(5):139-144.