| The rotor is a frequently used component in life and industry,and it is also a key link affecting the mechanical running performance.If the rotor is suddenly damaged during operation,it may cause mechanical shutdown or even complete smashing,so it is very important to prevent and diagnose the fault of the rotor.However,crack is one of the main faults of the rotor,so it is very important to diagnose the crack fault of the rotor.This paper will study the analysis and diagnosis of the crack fault of Jeffcott rotor.The dynamic model of the cracked rotor is studied,and the dynamic model of the cracked rotor is established.The dynamic unbalanced vibration characteristics of the rotor after the crack is generated are analyzed.first all,the rotor is modeled by the open and closed breathing crack model,and then the model analysis is carried out to obtain the characteristics of the crack rotor rotation signal.The vibration signal analysis method of the cracked rotor is studied,and the vibration signal of the cracked rotor is prepared by HHT method.and then,the calculation and analysis process of the HHT method is presented.The HHT method includes two steps of EMD decomposition and Hilbert transform.The EMD decomposition decomposes the signal into several IMF components,and the Hilbert transform solves the IMF component to obtain the Instantaneous frequency of each component.,then analyze the signal.And the simulation of HHT analysis method is carried out.The data acquisition experiment of the cracked rotor was carried out,and the experimental data was analyzed.First,the rotors without cracks,3 mm cracks,and 4 mm cracks were subjected to rotation experiments,and their data were collected at 1/3 and 1/2 critical rotational speeds.HHT calculations and marginal spectra were obtained from the experimentally acquired data.The components of the main IMF components after rotor EMD decomposition of different crack states were analyzed,and their HHT spectra and marginal spectra were compared under different crack conditions.The analysis gives characteristics that can distinguish the crack state.Finally,the diagnosis of rotor crack failure is studied.Two kinds of diagnosis methods for rotor crack fault are designed.The first is to decompose the cracked rotor signal EMD,and the obtained IMF components form a characteristic matrix,solve the singular value of the matrix,and solve the singular value entropy,called EMD singular value entropy,and use it as the crack fault diagnosis eigenvalue of the rotor.The second is to perform HHT analysis on the signal of the cracked rotor to obtain the HHT spectrum.According to the energy distribution of the time spectrum,the time-frequency entropy is calculated,which is called HHT time-frequency entropy.and is used as the crack fault diagnosis characteristic value of the rotor,and the crack rotor is diagnosed by the HHT spectrum.Two kinds of diagnostic methods are verified by simulation and experiment. |