VMD Method And Its Application To Rotaing Machinery Diagnosis

Rotating machinery is a key power plant machinery and equipment,once the failure of rotating machinery,machinery and equipment directly affect the operation of the situation,and even will bring tremendous economic losses and casualties.Therefore,to carry out the diagnosis of rotating machinery fault and trend research is of great significance.The key of gear fault diagnosis lies in extracting fault characteristic information,which needs to select the appropriate signal decomposition method.At present,the commonly used signal decomposition methods include empirical mode decomposition(EMD)and local mean decomposition(LMD)Etc.However,there are still many difficult problems to be solved in these methods.Therefore,it is particularly important to introduce new methods of signal decomposition into the research of gear fault diagnosis.This paper studies the application of the Variational mode decomposition(VMD)method to the fault diagnosis and trend analysis of rotating machinery.By analyzing the defects of the VMD method and improving the robustness of the proposed method,the robust source separation method is combined with the Robust Generalized Morphological Component Analysis(RGMCA)method to realize the blind source separation of the hybrid faults in rotating machinery.Compared with the Variable Predictive Model Based Class Discriminate(VPMCD)and other methods to achieve quantitative diagnosis of root deformity under variable speed conditions.This paper mainly completes two aspects of research: the theoretical research of VMD method and the application of VMD method in the fault diagnosis of rotating machinery.The main research contents are as follows:(1)Aimed at the problem that the number of modes in the VMD method is pre-set,(GD-IVMD)is proposed based on the gradient descent method;Aimed at the problem that the number of modal modal k and the penalty parameter ? in VMD method,FA-PMA-VMD is proposed based on based on firefly algorithm and Principle Mode Analysis;The improved VMD method is simulated and compared with the EEMD method Contrast,reflects the method in a certain range of superiority and practicality.The simulation results show that the VMD method is superior to the EMD method in a certain range and the defects of the VMD method are analyzed.(2)Aiming at the underdetermined problem to be solved during the blind source separation of vibration signals,the RGMCA method and the IVMD method are Combined.the IVMD method is used to decompose the si gnal and combine the decomposed signal with the original signal to solve the problem of undecidability;RGMCA method is used to separate the new mixed signal by blind source separation;The IVMD-RGMCA method is used to decompose the simulated mixed failure signal to verify the effectiveness of this method in simulating the mixed failure of rotating machinery.The experimental results show that the method is also practical in the real signal.(3)Based on the analysis of the characteristics of dynamic response under gear failure,the eigenvalues of the crack are sensitive to the change of the speed of rotation.The FA-PMD-VMD method is applied to the experimental signal to demonstrate that the eigenvalues calculated after FA-PMA-VMD pretreatment show the sensitivity and followability of the gear crack fault degree of the actual gear signal.The VPMCD pattern recognition method,Combined with FA-PMD-VMD,applied to gear fault diagnosis with different crack degree.The experimental analysis verified that this method can quantitatively diagnose the gear crack fault and has certain practicality.