Research On Rotor Crack Recognition Based On Vibration Characteristics

Rotating machinery is one of the most important mechanical equipment in the modern society,it has been widely used in various engineering industries.Rotor is the most core components in rotating machinery,Whether the rotor’s normal directly decides the whole rotating machinery can run properly.The crack fault is one of the most common failure forms of rotating machinery,therefore it on the rotor for crack identification in order to ensure safe and efficient operation has important social significance and economic value.Based on the theory of fracture mechanics and the finite element method,this paper studies the research on crack recognition.In the process of crack identification of the rotating shaft,the cantilever crack axis and the rotor crack axis are experimentally studied.The influence of crack on the vibration parameters of the rotating shaft is studied experimentally,and the natural frequency of the cracked axis in the horizontal direction and the vertical direction is measured.By analyzing the natural frequency values in the two directions of the cantilever shafts with different crack conditions,it is concluded that the natural frequency of the crack axis is asymmetric due to the crack.Based on the cantilever axis of the crack,the modal parameters of the cantilever shaft with different crack are simulated by parametric modeling,and all the crack parameters and modal parameters are taken as samples to form the sample database.Then,the neural network is designed and designed in MATLAB,and the natural frequency of the horizontal and vertical directions of the crack axis is selected as the input of the neural network.By selecting the neural network from the sample database and correcting the neural network with the experimental data,the cantilever shaft crack recognition method based on the vibration characteristic is obtained.Compared with the contour method,the average height of the crack is 1.76%,which is 2.61%higher on the recognition accuracy of the crack location,and the advantage of using the asymmetric vibration characteristic parameters to crack identification is fully explained.Finally,based on the research of cantilever shaft crack recognition,a complete rotor crack identification method is obtained by modeling the rotor system model and adjusting the parameters of the neural network.When the rotor system model is established accurately,the bearing stiffness is simulated by the equivalent stiffness spring element,and the finite element model of the rotor crack axis is corrected by the experimental data,and the accurate rotor crack system model is obtained finally.In the training of neural network,the neural network can be applied to the research of rotor crack recognition by adjusting the parameters such as the number of hidden nodes,weights and learning speed of the neural network.In this paper,based on the asymmetric vibration characteristic parameters can achieve the cantilever shaft and rotor system crack recognition,and has a high accuracy,and can change the finite element model and adjust the neural network can be applied to a variety of crack recognition research.