With the development of multi-signal monitoring technology,the research on multiple signal analysis and processing has become a hot subject.Mechanical equipment often works under variable working conditions,and the ac...With the development of multi-signal monitoring technology,the research on multiple signal analysis and processing has become a hot subject.Mechanical equipment often works under variable working conditions,and the acquired vibration signals are often non-stationary and nonlinear,which are difficult to be processed by traditional analysis methods.In order to solve the noise reduction problem of multiple signals under variable speed,a COT-DCS method combining the Computed Order Tracking(COT)based on Chirplet Path Pursuit(CPP)and Distributed Compressed Sensing(DCS)is proposed.Firstly,the instantaneous frequency(IF)is extracted by CPP,and the speed is obtained by fitting.Then,the speed is used for equal angle sampling of time-domain signals,and angle-domain signals are obtained by COT without a tachometer to eliminate the nonstationarity,and the angledomain signals are compressed and reconstructed by DCS to achieve noise reduction of multiple signals.The accuracy of the CPP method is verified by simulated,experimental signals and compared with some existing IF extraction methods.The COT method also shows good signal stabilization ability through simulation and experiment.Finally,combined with the comparative test of the other two algorithms and four noise reduction effect indicators,the COT-DCS based on the CPP method combines the advantages of the two algorithms and has better noise reduction effect and stability.It is shown that this method is an effective multi-signal noise reduction method.展开更多
为解决变转速工况滚动轴承微弱故障特征难以提取的问题,提出了时时(time-time,TT)变换结合计算阶比跟踪(computed order tracking,COT)的滚动轴承时变微弱故障特征提取方法。首先对变转速状态的轴承微弱故障信号进行时时变换,得到反映...为解决变转速工况滚动轴承微弱故障特征难以提取的问题,提出了时时(time-time,TT)变换结合计算阶比跟踪(computed order tracking,COT)的滚动轴承时变微弱故障特征提取方法。首先对变转速状态的轴承微弱故障信号进行时时变换,得到反映故障信号二维时时特征的TT变换矩阵。为消除TT变换矩阵的冗余性,提出了基于峭度准则的奇异值分解(singular value decomposition,SVD)方法对TT变换矩阵降噪。然后对降噪后的TT变换矩阵实施TT反变换,获取滚动轴承时变故障特征增强信号。最后对增强信号进行COT分析得到其包络阶比谱,从而提取滚动轴承的故障特征阶次。对仿真信号和实验测试信号进行分析验证,均实现了滚动轴承变转速工况故障类型的精确识别,分析效果优于包络阶比谱方法,证明了该方法的有效性。展开更多
基金the financial support of this work by the National Natural Science Foundation of Hebei Province China under Grant E2020208052.
文摘With the development of multi-signal monitoring technology,the research on multiple signal analysis and processing has become a hot subject.Mechanical equipment often works under variable working conditions,and the acquired vibration signals are often non-stationary and nonlinear,which are difficult to be processed by traditional analysis methods.In order to solve the noise reduction problem of multiple signals under variable speed,a COT-DCS method combining the Computed Order Tracking(COT)based on Chirplet Path Pursuit(CPP)and Distributed Compressed Sensing(DCS)is proposed.Firstly,the instantaneous frequency(IF)is extracted by CPP,and the speed is obtained by fitting.Then,the speed is used for equal angle sampling of time-domain signals,and angle-domain signals are obtained by COT without a tachometer to eliminate the nonstationarity,and the angledomain signals are compressed and reconstructed by DCS to achieve noise reduction of multiple signals.The accuracy of the CPP method is verified by simulated,experimental signals and compared with some existing IF extraction methods.The COT method also shows good signal stabilization ability through simulation and experiment.Finally,combined with the comparative test of the other two algorithms and four noise reduction effect indicators,the COT-DCS based on the CPP method combines the advantages of the two algorithms and has better noise reduction effect and stability.It is shown that this method is an effective multi-signal noise reduction method.
文摘为解决变转速工况滚动轴承微弱故障特征难以提取的问题,提出了时时(time-time,TT)变换结合计算阶比跟踪(computed order tracking,COT)的滚动轴承时变微弱故障特征提取方法。首先对变转速状态的轴承微弱故障信号进行时时变换,得到反映故障信号二维时时特征的TT变换矩阵。为消除TT变换矩阵的冗余性,提出了基于峭度准则的奇异值分解(singular value decomposition,SVD)方法对TT变换矩阵降噪。然后对降噪后的TT变换矩阵实施TT反变换,获取滚动轴承时变故障特征增强信号。最后对增强信号进行COT分析得到其包络阶比谱,从而提取滚动轴承的故障特征阶次。对仿真信号和实验测试信号进行分析验证,均实现了滚动轴承变转速工况故障类型的精确识别,分析效果优于包络阶比谱方法,证明了该方法的有效性。
