基于随机共振和随机森林的轴承故障诊断
|
摘要
本文研究了基于随机共振和随机森林的轴承故障诊断方法。首先将原始的轴承故障数据分为训练集和测试集,然后将数据集通过随机共振处理以增强弱故障特征,之后将增强后的故障数据输入随机森林中进行模型训练,通过调整合适的模型参数,计算出最终正确率。结果表明,与传统的机器学习相比,本文提出的方法具有更高的识别准确率,可为智能印刷机的设计及故障诊断提供理论指导和技术支持。
In this study,the bearing fault diagnosis method based on stochastic resonance and random forest was mainly researched.Firstly,the original bearing fault data was divided into training set and test set,then the data set was subjected to stochastic resonance processing to enhance the weak fault feature.The enhanced fault data was input into the random forest for model training,and the appropriate model parameters were adjusted to calculate the final correct rate.This method proposed has higher recognition accuracy than traditional machine learning.This research provides theoretical guidance and technical support for the design and fault diagnosis of printing machine.
In this study,the bearing fault diagnosis method based on stochastic resonance and random forest was mainly researched.Firstly,the original bearing fault data was divided into training set and test set,then the data set was subjected to stochastic resonance processing to enhance the weak fault feature.The enhanced fault data was input into the random forest for model training,and the appropriate model parameters were adjusted to calculate the final correct rate.This method proposed has higher recognition accuracy than traditional machine learning.This research provides theoretical guidance and technical support for the design and fault diagnosis of printing machine.
引文
[1]黄辉.基于SOPC的滚动轴承故障诊断系统研究和开发[D].太原:中北大学,2010.HUANG Hui.Research and Development of Rolling Bearing Fault Diagnosis System Based on SOPC[D].Taiyuan:North University of China,2010.
[2]武吉梅,张嘉玲,王砚,等.印刷包装设备常用深沟球轴承参数对性能的影响[J].包装工程,2017(7):1-5.WU Ji-mei,ZHANG Jia-ling,WANG Yan,et al.Effect of Deep Groove Ball Bearing Parameters on Performance of Printing and Packaging Equipment[J].Packaging Engineering,2017(7):1-5.
[3]黄媛,蔡吉飞,陈朋飞.卷筒纸输纸装置设计研究[J].中国印刷与包装研究,2010,2(5):46-49.HUANG Yuan,CAI Ji-fei,CHEN Peng-fei.Research on Design of Web Feeding Device[J].China Printing and Packaging Research,2010,2(5):46-49.
[4]蒲子玺.基于VMD的滚动轴承故障诊断研究[D].兰州:兰州交通大学,2017.PU Zi-xi.Research on Fault Diagnosis of Rolling Bearing Based on VMD[D].Lanzhou:Lanzhou Jiaotong University,2017.
[5]刘峰,杨晓萍,刘晓黎,等.基于神经网络的水轮发电机组振动故障诊断专家系统的研究[J].西安理工大学学报,2003,19(4):372-376.LIU Feng,YANG Xiao-ping,LIU Xiao-li,et al.Research on the Expert System of Vibration Fault Diagnosis of Hydroelectric Generating Set Based on Neural Network[J].Journal of Xi'an University of Technology,2003,19(4):372-376.
[6]周文涛.基于时频分析的信号检测与识别方法研究[D].成都:电子科技大学,2012.ZHOU Wen-tao.Research on Signal Detection and Recognition Based on Time-Frequency Analysis[D].Chengdu:University of Electronic Science and Technology of China,2012.
[7]王志武.强噪声背景下机械故障微弱信号特征提取方法研究[D].太原:中北大学,2014.WANG Zhi-wu.Research on feature Extraction Method of Weak Signal of Mechanical Fault under S t r o n g N o i s e B a c k g r o u n d[D].Taiyuan:N o r t h University of China,2014.
[8]刘进.基于非线性随机共振的弱信号检测理论研究[D].西安:西安电子科技大学,2015.LIU Jin.Theoretical Study of Weak Signal Detection Based on Nonlinear Stochastic Resonance[D].Xi'an:Xidian University,2015.
[9]张嘉玲.基于变分模态分解和随机共振的轴承故障自适应诊断研究[D].西安:西安理工大学,2018.ZHANG Jia-ling.Adaptive Fault Diagnosis of Bearing Fault Based on Variational Mode Decomposition and Stochastic Resonance[D].Xi'an:Xi'an University of Technology,2018.
[10]章鸣嬛,陈瑛,沈瑛,等.人工神经网络和支持向量机性能比较及其在DMD疾病识别中的应用[J].上海理工大学学报,2016,38(4):346-351.ZHANG Ming-huan,CHEN Ying,SHEN Ying,et al.Performance Comparison of Artificial Neural Network and Support Vector Machine and Its Application in DMD Disease Identification[J].Journal of University of Shanghai for Science and Technology,2016,38(4):346-351.
[11]魏洒洒.基于随机森林的SVM训练数据选择研究[D].西安:西安电子科技大学,2017.WEI Sa-sa.Research on SVM Training Data Selection Based on Random Forest[D].Xi'an:Xidian University,2017.
[12]DONG L J,LI X B,PENG K.Prediction of Rock-burst Classification using Random Forest[J].Transactions of Nonferrous Metals Society of China,2013,23(2):472-477.
[13]冷永刚,王太勇,郭焱,等.双稳随机共振参数特性的研究[J].物理学报,2007,56(1):30-35.LENG Yong-gang,WANG Tai-yong,GUO Yan,et al.Study on the Characteristics of Bistable Stochastic Resonance Parameters[J].Acta Physica Sinica,2007,56(1):30-35.
[14]胡兵兵,何正嘉,陈雪峰,等.基于自适应多尺度随机共振的机械故障诊断方法研究[J/OL].(2015-12-26)[2019-01].http://www.paper.edu.cn/releasepaper/content/4670265.HU Bing-bing,HE Zheng-jia,CHEN Xue-feng,et al.Research on Mechanical Fault Diagnosis Based on Adaptive Multi-Scale Stochastic Resonance[J/OL].(2015-12-26)[2019-01].http://www.paper.edu.cn/releasepaper/content/4670265.
[15]ZHANG L Y,LI C,WU D J,et al.Stochastic Resonance in a Single-Mode Laser Driven by Quadratic Colored Pump Noise:Effects of Biased Amplitude Modulation Signal[J].Communications in Theoretical Physics,2009,52(7):143-148.
[16]DEBASHIS Kushary.Bootstrap Methods and Their Application[J].Technometrics,2000,42(2):216-217.
[17]王海燕,侯琳娜.Bootstrap重采样Bayesian方法在白车身制造尺寸不合格率估计中的应用[J].工业工程,2012,15(6):15-19.WA N G H a i-y a n,H O U L i n-n a.A p p l i c a t i o n o f B o o t s t r a p R e s a m p l i n g B a y e s i a n M e t h o d i n Estimating the Unqualified Rate of Body-in-White Manufacturing[J].Industrial Engineering,2012,15(6):15-19.
[18]方匡南,吴见彬,朱建平,等.随机森林方法研究综述[J].统计与信息论坛,2011,26(3):32-38.FANG Kuang-nan,WU Jian-bin,ZHU Jian-ping,et al.Review of Random Forest Methods Research[J].Journal of Statistics and Information,2011,26(3):32-38.
[19]LOPARO K A.Bearing Data Center,Ase Western Reserveuniversity[EB/OL].http://www.eecs.case.edu/laboratory/bearing,2013.
[2]武吉梅,张嘉玲,王砚,等.印刷包装设备常用深沟球轴承参数对性能的影响[J].包装工程,2017(7):1-5.WU Ji-mei,ZHANG Jia-ling,WANG Yan,et al.Effect of Deep Groove Ball Bearing Parameters on Performance of Printing and Packaging Equipment[J].Packaging Engineering,2017(7):1-5.
[3]黄媛,蔡吉飞,陈朋飞.卷筒纸输纸装置设计研究[J].中国印刷与包装研究,2010,2(5):46-49.HUANG Yuan,CAI Ji-fei,CHEN Peng-fei.Research on Design of Web Feeding Device[J].China Printing and Packaging Research,2010,2(5):46-49.
[4]蒲子玺.基于VMD的滚动轴承故障诊断研究[D].兰州:兰州交通大学,2017.PU Zi-xi.Research on Fault Diagnosis of Rolling Bearing Based on VMD[D].Lanzhou:Lanzhou Jiaotong University,2017.
[5]刘峰,杨晓萍,刘晓黎,等.基于神经网络的水轮发电机组振动故障诊断专家系统的研究[J].西安理工大学学报,2003,19(4):372-376.LIU Feng,YANG Xiao-ping,LIU Xiao-li,et al.Research on the Expert System of Vibration Fault Diagnosis of Hydroelectric Generating Set Based on Neural Network[J].Journal of Xi'an University of Technology,2003,19(4):372-376.
[6]周文涛.基于时频分析的信号检测与识别方法研究[D].成都:电子科技大学,2012.ZHOU Wen-tao.Research on Signal Detection and Recognition Based on Time-Frequency Analysis[D].Chengdu:University of Electronic Science and Technology of China,2012.
[7]王志武.强噪声背景下机械故障微弱信号特征提取方法研究[D].太原:中北大学,2014.WANG Zhi-wu.Research on feature Extraction Method of Weak Signal of Mechanical Fault under S t r o n g N o i s e B a c k g r o u n d[D].Taiyuan:N o r t h University of China,2014.
[8]刘进.基于非线性随机共振的弱信号检测理论研究[D].西安:西安电子科技大学,2015.LIU Jin.Theoretical Study of Weak Signal Detection Based on Nonlinear Stochastic Resonance[D].Xi'an:Xidian University,2015.
[9]张嘉玲.基于变分模态分解和随机共振的轴承故障自适应诊断研究[D].西安:西安理工大学,2018.ZHANG Jia-ling.Adaptive Fault Diagnosis of Bearing Fault Based on Variational Mode Decomposition and Stochastic Resonance[D].Xi'an:Xi'an University of Technology,2018.
[10]章鸣嬛,陈瑛,沈瑛,等.人工神经网络和支持向量机性能比较及其在DMD疾病识别中的应用[J].上海理工大学学报,2016,38(4):346-351.ZHANG Ming-huan,CHEN Ying,SHEN Ying,et al.Performance Comparison of Artificial Neural Network and Support Vector Machine and Its Application in DMD Disease Identification[J].Journal of University of Shanghai for Science and Technology,2016,38(4):346-351.
[11]魏洒洒.基于随机森林的SVM训练数据选择研究[D].西安:西安电子科技大学,2017.WEI Sa-sa.Research on SVM Training Data Selection Based on Random Forest[D].Xi'an:Xidian University,2017.
[12]DONG L J,LI X B,PENG K.Prediction of Rock-burst Classification using Random Forest[J].Transactions of Nonferrous Metals Society of China,2013,23(2):472-477.
[13]冷永刚,王太勇,郭焱,等.双稳随机共振参数特性的研究[J].物理学报,2007,56(1):30-35.LENG Yong-gang,WANG Tai-yong,GUO Yan,et al.Study on the Characteristics of Bistable Stochastic Resonance Parameters[J].Acta Physica Sinica,2007,56(1):30-35.
[14]胡兵兵,何正嘉,陈雪峰,等.基于自适应多尺度随机共振的机械故障诊断方法研究[J/OL].(2015-12-26)[2019-01].http://www.paper.edu.cn/releasepaper/content/4670265.HU Bing-bing,HE Zheng-jia,CHEN Xue-feng,et al.Research on Mechanical Fault Diagnosis Based on Adaptive Multi-Scale Stochastic Resonance[J/OL].(2015-12-26)[2019-01].http://www.paper.edu.cn/releasepaper/content/4670265.
[15]ZHANG L Y,LI C,WU D J,et al.Stochastic Resonance in a Single-Mode Laser Driven by Quadratic Colored Pump Noise:Effects of Biased Amplitude Modulation Signal[J].Communications in Theoretical Physics,2009,52(7):143-148.
[16]DEBASHIS Kushary.Bootstrap Methods and Their Application[J].Technometrics,2000,42(2):216-217.
[17]王海燕,侯琳娜.Bootstrap重采样Bayesian方法在白车身制造尺寸不合格率估计中的应用[J].工业工程,2012,15(6):15-19.WA N G H a i-y a n,H O U L i n-n a.A p p l i c a t i o n o f B o o t s t r a p R e s a m p l i n g B a y e s i a n M e t h o d i n Estimating the Unqualified Rate of Body-in-White Manufacturing[J].Industrial Engineering,2012,15(6):15-19.
[18]方匡南,吴见彬,朱建平,等.随机森林方法研究综述[J].统计与信息论坛,2011,26(3):32-38.FANG Kuang-nan,WU Jian-bin,ZHU Jian-ping,et al.Review of Random Forest Methods Research[J].Journal of Statistics and Information,2011,26(3):32-38.
[19]LOPARO K A.Bearing Data Center,Ase Western Reserveuniversity[EB/OL].http://www.eecs.case.edu/laboratory/bearing,2013.