Study on Prediction Methods for the Fault State of Rotating Machinery Based on Dynamic Grey Model and Metabolism Grey Model

Rotors and bearings are the key parts of rotating machinery. Mechanical faults will occur easily when rotors and bearings are running for a long time in the condition of high speed and full load. In this paper, first the dynamic grey model and metabolism grey model (MGM) are respectively used to predict the trend of the vibration amplitude of rotors and bearings, and the prediction results are compared. Then based on the root mean square value of the vibration amplitude of rotors and bearings, a back propagation network prediction model of fault feature information is established, which can predict the fault of rotors and bearings in advance. Experiments show that the dynamic grey model can predict both the rising and comprehensive growth trends of the vibration signal amplitude of rotors and bearings. However, the prediction error will increase with an increase of vibration amplitude. Experiments also indicate that the accuracy of prediction based on the MGM is higher than that of dynamic grey model.