Rolling bearing fault diagnosis techniques - autocorrelation and cepstrum analyses

Abstract

The faulty vibration signals generally represent a combination of source and transmission path effects. For example, internal forces in vehicle gearbox which are the source of vibration act on a structure whose properties may be described by a frequency response function between the point of application and the point of measurements. A conventional frequency analysis of a spectrum will have transmission path effects affecting the true source signature and also cannot pinpoint the defects accurately where the problem is associated with more than one sideband and harmonic in case of vehicle gearbox. In the present paper, the transmission path effects are additive and can be separated in autocorrelation and cepstrum analyses, also it gives an accurate detection of periodic structure in a spectrum associated with many harmonics and sidebands as a single component for each family of sidebands without any difficulty in interpreting the sideband structure like in spectrum. The test stand is equipped with two dynamometers; the input dynamometer serves as internal combustion engine, the output dynamometer introduces the load on the flange of output joint shaft. An artificial fault is introduced in vehicle gearbox bearing: an orthogonally placed groove on the inner race with the initial width of 0.6 mm approximately. The results show the effectiveness of proposed analyses in diagnosis and detection of the rolling bearing condition.