Bearing Fault Diagnosis Based On Convolution Filtering

The operation status of the key equipments in machinery industry,transportation,energy,and petrochemical industry has a close relationship with the normal operation and healthy development of the national economy.The key components of the equipment,the bearing running in such complex and severe conditions as heavy load,elevated temperature,would inevitably experience performance degradation and failure.To diagnose the failures of the components,guarantee the safety operation of the equipments and avoid catastrophic accidents,it is of great significance to extract the fault features of the key components.With the aim of fault feature extraction of rolling bearing,two methods of signal feature extraction based on convolution filtering are studied theoretically and experimentally.The theoretical research and application research are studied in depth,respectively.The failure types and the characteristics of the fault vibration signals of the bearing,which has localized fault,were analyzed respectively.The sparsity nature of the vibration signal is revealed.Through the simulation signal analysis,the problems of the minimum entropy deconvolution in the signal feature detection is truly demonstrated.The key issues of the signal convolution filtering are also revealed in this thesis.All the works provides the theoretical foundation of the research in this thesis.Solving the simplicity and instability of sparse criterion in minimum entropy deconvolution method,the deconvolution algorithm based on the generalized P operator sparse criteria was proposed.Firstly,deriving the optimal deconvolution expression based on the generalized P operator.Then,the bearing faulty feature extraction method was constructed by combining the generalized sparse criterion and the normalized frequency energy ratio.The analysis results of the simulated signals revealed that the proposed method can well extract the bearing fault feature under the strong background noise and steadily filter the noise component under the different signal to noise ratio.Finally,the effectiveness of the proposed method for fault feature extraction is verified through the applications in fault rolling bearing and outperforms some existed methods for bearing fault diagnosis,respectively.Furthermore,the minimum entropy deconvolution method have poor capability of fault feature extraction in the poor sparsity of signal under strong background noise.A novel weak bearing fault diagnosis method are proposed by incorporating a l0 norm regularized solution based on approximate hyperbolic tangent function into the iterative de-convolving procedure of MED.Due to the introduction of l0 norm regularized solution,the proposed method can obtain a sparser representation of the faulty signal and increases robustness to inverse filter length compared to the MED.Both simulation and the experimental results demonstrate that the proposed method can accurately identify weak bearing defects and outperforms some existed methods for bearing fault diagnosis.In this thesis,through the research on the transient feature extraction based on the convolution filtering,it is confirmed that both the proposed methods are effective in detecting the fault feature of machinery components,which has theoretical and practical value for fault extraction and diagnosis of rotating machinery.