On-line Monitoring System for Spindle Faults on Automobile Reducer Test Device

As a key part of automobile, the automobile reducer must be subjected to a durability test on the test platform before it is put into use, which requires the test device to be highly reliable. In this paper, the spindle fault monitoring technology in durability test is studied by a data-driven method, and an on-line spindle fault monitoring system of rotating machinery based on virtual instrument is established. In this system, the vibration and rotational speed of spindle are taken as the monitoring signals, and the feature parameters which can reflect the operating state of spindle are extracted by using the signal processing technology which combines the time domain analysis, the frequency domain analysis and the wavelet packet analysis, to establish a spindle fault identification model based on fuzzy C-means clustering algorithm, which judges the spindle state by calculating the current fault feature parameters and the membership degree of the known state in the model; finally, a set of on-line monitoring and diagnosis system software for spindle faults on automobile reducer test device is designed with LabVIEW, which is based on virtual instrument technology, as the programming tool, and the theoretical verification and the example verification are carried out in the actual automobile reducer test. The research results show that the system can identify the spindle faults such as impact, friction, looseness and imbalance on line in real time, with the advantages of high speed, high efficiency and high accuracy (98% on average).