The Relationship Between Shaft And Bearing Vibration Of Rotation Machinery Based On Full Vector Spectrum

With the rapid development of modern industrial technology and rotating machinery increasing degree of automation, the normal operation conditions of device also will be increased. Therefore, for the large rotating machinery, especially for the critical equipment in real-time and accurate monitoring, as soon as possible to troubleshoot the source, thus ensuring the equipment normal, stable and highly efficient operation is necessary. At present, the eddy current sensor is used for measuring the relative vibration of the journal compared with the bearing, and the velocity, acceleration sensor is used for measuring the absolute vibration of the bearing vibration. However, due to the impact of the eddy whirling of rotor, the signal of shaft vibration and bearing vibration from the single channel can't overall reflect the real operation status of the equipment. Moreover at the most conditions, the different channels of the same section express the different vibration information. So these lead to much difficulty for the fault identification. Therefore, the signal of shaft vibration and bearing vibration is necessary to be treated with the information fusion.And in order to improve the reliability and accuracy of the rotating machinery condition monitoring and fault identification, the study between shaft vibration and bearing vibration based on the full vector spectrum is very useful.Aiming at these problems, the dynamic models of the rotor-bearing-foundation system with non-failure, pedestal looseness and rub-impact were established. The signal of shaft vibration and bearing vibration was obtained by the numerical integration method. The two channels signal of shaft vibration and bearing vibration was treated with the information fusion by full vector spectrum. And the relationship between the synthetic tracks of the two signals was studied. At last, the simulation results were verified by experiments.The results show that the trend of main vector of shaft vibration which is changed under the speed variation is close to that of the main vector of bearing vibration in case of the rigid rotor system with non-failure. For the system with pedestal looseness, the trend of main vector of the two signals with the speed variation is similar. For the rub-impact fault, the trend of main vector of shaft vibration arid bearing vibration changed under the speed variation is similar only for the rigid rotor system. Furthermore, the fault feature of bearing vibration is clearer than that of shaft vibration for the pedestal looseness fault and the rub-impact fault, especially for the early failure. But the noise level of bearing vibration is higher than that of shaft vibration.Therefore, the two channels signal of shaft vibration and bearing vibration treated by full vector spectrum can improve the accuracy and reliability of rotating machinery condition monitoring and fault diagnosis. So the combination of shaft vibration monitoring and bearing vibration monitoring is beneficial to realize the evaluation of the rotating machine's running state.