EEMD样本熵的扬声器异常音分类方法
|
摘要
为了更准确的对扬声器异常音进行分类,提出一种基于集合经验模态分解(EEMD)与样本熵的扬声器异常音特征提取方法并利用极限梯度提升(XGBoost)算法进行分类。在基频陷波预处理后,对信号进行EEMD,结合相关性分析选取固有模态函数(IMF)分量并计算其样本熵构成特征向量。实验结果表明,针对扬声器异常音分类问题,在小样本的情况下,扬声器声响应信号经基频陷波预处理后,XGBoost算法使用EEMD与样本熵的特征提取方法取得了95. 33%的分类准确率,高于小波包变换和样本熵特征提取方法所取得的准确率,验证了特征提取及分类方法的有效性。
To classify the abnormal sound of loudspeaker more accurately,an algorithm was proposed in this paper,in which ensemble empirical mode decomposition( EEMD) and sample entropy were used for feature extraction and extreme gradient boosting( XGBoost)was used for classification. After preprocessing of pitch notching,the loudspeaker response signal was decomposed using EEMD. The intrinsic mode function( IMF) components were selected with correlation analysis and their sample entropy values were calculated to structure the feature vectors. Focused on the classification for loudspeaker abnormal sound with small sample condition,the experiment results have shown that,after the preprocessing of pitch notching for loudspeaker response signals,XGBoost algorithm with EEMD and sample entropy feature extraction method can accurately classify abnormal sound of loudspeaker,which is more accurate than XGBoost algorithm using feature vectors of wavelet packets and sample entropy,moreover,it achieves 95. 33% classification accuracy.
To classify the abnormal sound of loudspeaker more accurately,an algorithm was proposed in this paper,in which ensemble empirical mode decomposition( EEMD) and sample entropy were used for feature extraction and extreme gradient boosting( XGBoost)was used for classification. After preprocessing of pitch notching,the loudspeaker response signal was decomposed using EEMD. The intrinsic mode function( IMF) components were selected with correlation analysis and their sample entropy values were calculated to structure the feature vectors. Focused on the classification for loudspeaker abnormal sound with small sample condition,the experiment results have shown that,after the preprocessing of pitch notching for loudspeaker response signals,XGBoost algorithm with EEMD and sample entropy feature extraction method can accurately classify abnormal sound of loudspeaker,which is more accurate than XGBoost algorithm using feature vectors of wavelet packets and sample entropy,moreover,it achieves 95. 33% classification accuracy.
引文
[1]DELGADO M,ORTEGA J A,RUIZ G,et al.Loudspeaker rub fault detection by means of a new nonstationary procedure test[C].Audio Engineering Society Convention,2010:1-7.
[2]TEMME S,BRUNET P,QARABAQ P.Measurement of harmonic distortion audibility using a simplified psychoacoustic model-updated[C].Audio Engineering Society Conference:51st International Conference:Loudspeakers and Headphones,2013:1-8.
[3]李宏斌,徐楚林,温周斌.基于短时傅里叶变换的异常音检测方法[J].声学技术,2014,33(2):145-149.LI H B,XU CH L,WEN ZH B.A STFT based method for detecting Rub&Buzz defects of loudspeaker and its application research[J].Technical Acoustics,2014,33(2):145-149.
[4]HUANG N E.Hilbert-Huang Transform and Its Applications[M].2nd ed.Singapore:World Scientific,2014.
[5]崔冰波,陈熙源.光纤陀螺温度漂移的多尺度集成建模研究[J].仪器仪表学报,2015,36(2):304-309.CUI B B,CHEN X Y.Study on multi-scale ensemble modeling of FOG temperature drift[J].Chinese Journal of Scientific Instrument,2015,36(2):304-309.
[6]WU Z,HUANG N E.Ensemble empirical mode decomposition:A noise-assisted data analysis method[J].Advances in Adaptive Data Analysis,2009,1(1):1-41.
[7]陈法法,杨勇,马婧华,等.信息熵与优化LS-SVM的轴承性能退化模糊粒化预测[J].仪器仪表学报,2016,37(4):779-787.CHEN F F,YANG Y,MA J H,et al.Fuzzy granulation prediction for bearing performance degradation based on information entropy and optimized LS-SVM[J].Chinese Journal of Scientific Instrument,2016,37(4):779-787.
[8]陈科,单姗,郑红梅.具有自适应性的实时睡眠信号处理算法研究[J].电子测量与仪器学报,2016,30(10):1497-1505.CHEN K,SHAN SH,ZHENG H M.Research on an adaptive algorithm for real-time sleep signal processing[J].Journal of Electronic Measurement and Instrumentation,2016,30(10):1497-1505.
[9]田书.基于EEMD和能量法的谐振接地配电网故障选线研究[J].电子测量技术,2016,39(9):176-181.TIAN SH.Study on fault line selection in resonant earthed distribution network based on EEMD and energy[J].Electronic Measurement Technology,2016,39(9):176-181.
[10]向丹,葛爽.基于EMD样本熵-LLTSA的故障特征提取方法[J].航空动力学报,2014,29(7):1535-1542.XIANG D,GE SH.Method of fault feature extraction based on EMD sample entropy and LLTSA[J].Journal of Aerospace Power,2014,29(7):1535-1542.
[11]ALCARAZ R,RIETA J J.A review on sample entropy applications for the non-invasive analysis of atrial fibrillation electrocardiograms[J].Biomedical Signal Processing and Control,2010,5(1):1-14.
[12]吴正平,魏欢,杨翔宇,等.无线脑电睡眠分期系统设计及算法研究[J].电子测量与仪器学报,2016,30(12):1861-1868.WU ZH P,WEI H,YANG X Y,et al.System design and algorithm research on wireless EEG signal sleep stage[J].Journal of Electronic Measurement and Instrumentation,2016,30(12):1861-1868.
[13]CHEN T,GUESTRIN C.Xgboost:A scalable tree boosting system[C].Proceedings of the 22nd ACMSigkdd International Conference on Knowledge Discovery and Data Mining,2016:785-794.
[14]LARSON J,DELLASAIA G,KEELE JR D B.A tutorial on the audibility of loudspeaker distortion at bass frequencies[C].Audio Engineering Society Convention,2017:1-10.
[15]李献斌,王跃科,杨俊.级联陷波器抗窄带干扰设计与实现[J].电子测量与仪器学报,2013,27(6):492-497.LI X B,WANG Y K,YANG J.Design and implementation of cascade IIR notch filters for narrowband interferences suppression[J].Journal of Electronic Measurement and Instrumentation,2013,27(6):492-497.
[16]TEMME S,DOBOS V.Evaluation of audio test methods and measurements for end-of-line automotive loudspeaker quality control[C].Audio Engineering Society Convention,2017:1-10.
[17]孙萍萍.基于小波包理论的激波信号去噪研究[J].国外电子测量技术,2016,35(7):38-42.SUN P P.Denoising of shock signal based on wavelet packet theory[J].Foreign Electronic Measurement Technology,2016,35(7):38-42.
[2]TEMME S,BRUNET P,QARABAQ P.Measurement of harmonic distortion audibility using a simplified psychoacoustic model-updated[C].Audio Engineering Society Conference:51st International Conference:Loudspeakers and Headphones,2013:1-8.
[3]李宏斌,徐楚林,温周斌.基于短时傅里叶变换的异常音检测方法[J].声学技术,2014,33(2):145-149.LI H B,XU CH L,WEN ZH B.A STFT based method for detecting Rub&Buzz defects of loudspeaker and its application research[J].Technical Acoustics,2014,33(2):145-149.
[4]HUANG N E.Hilbert-Huang Transform and Its Applications[M].2nd ed.Singapore:World Scientific,2014.
[5]崔冰波,陈熙源.光纤陀螺温度漂移的多尺度集成建模研究[J].仪器仪表学报,2015,36(2):304-309.CUI B B,CHEN X Y.Study on multi-scale ensemble modeling of FOG temperature drift[J].Chinese Journal of Scientific Instrument,2015,36(2):304-309.
[6]WU Z,HUANG N E.Ensemble empirical mode decomposition:A noise-assisted data analysis method[J].Advances in Adaptive Data Analysis,2009,1(1):1-41.
[7]陈法法,杨勇,马婧华,等.信息熵与优化LS-SVM的轴承性能退化模糊粒化预测[J].仪器仪表学报,2016,37(4):779-787.CHEN F F,YANG Y,MA J H,et al.Fuzzy granulation prediction for bearing performance degradation based on information entropy and optimized LS-SVM[J].Chinese Journal of Scientific Instrument,2016,37(4):779-787.
[8]陈科,单姗,郑红梅.具有自适应性的实时睡眠信号处理算法研究[J].电子测量与仪器学报,2016,30(10):1497-1505.CHEN K,SHAN SH,ZHENG H M.Research on an adaptive algorithm for real-time sleep signal processing[J].Journal of Electronic Measurement and Instrumentation,2016,30(10):1497-1505.
[9]田书.基于EEMD和能量法的谐振接地配电网故障选线研究[J].电子测量技术,2016,39(9):176-181.TIAN SH.Study on fault line selection in resonant earthed distribution network based on EEMD and energy[J].Electronic Measurement Technology,2016,39(9):176-181.
[10]向丹,葛爽.基于EMD样本熵-LLTSA的故障特征提取方法[J].航空动力学报,2014,29(7):1535-1542.XIANG D,GE SH.Method of fault feature extraction based on EMD sample entropy and LLTSA[J].Journal of Aerospace Power,2014,29(7):1535-1542.
[11]ALCARAZ R,RIETA J J.A review on sample entropy applications for the non-invasive analysis of atrial fibrillation electrocardiograms[J].Biomedical Signal Processing and Control,2010,5(1):1-14.
[12]吴正平,魏欢,杨翔宇,等.无线脑电睡眠分期系统设计及算法研究[J].电子测量与仪器学报,2016,30(12):1861-1868.WU ZH P,WEI H,YANG X Y,et al.System design and algorithm research on wireless EEG signal sleep stage[J].Journal of Electronic Measurement and Instrumentation,2016,30(12):1861-1868.
[13]CHEN T,GUESTRIN C.Xgboost:A scalable tree boosting system[C].Proceedings of the 22nd ACMSigkdd International Conference on Knowledge Discovery and Data Mining,2016:785-794.
[14]LARSON J,DELLASAIA G,KEELE JR D B.A tutorial on the audibility of loudspeaker distortion at bass frequencies[C].Audio Engineering Society Convention,2017:1-10.
[15]李献斌,王跃科,杨俊.级联陷波器抗窄带干扰设计与实现[J].电子测量与仪器学报,2013,27(6):492-497.LI X B,WANG Y K,YANG J.Design and implementation of cascade IIR notch filters for narrowband interferences suppression[J].Journal of Electronic Measurement and Instrumentation,2013,27(6):492-497.
[16]TEMME S,DOBOS V.Evaluation of audio test methods and measurements for end-of-line automotive loudspeaker quality control[C].Audio Engineering Society Convention,2017:1-10.
[17]孙萍萍.基于小波包理论的激波信号去噪研究[J].国外电子测量技术,2016,35(7):38-42.SUN P P.Denoising of shock signal based on wavelet packet theory[J].Foreign Electronic Measurement Technology,2016,35(7):38-42.