European Journal of Mechanics A/Solids 24 (2005) 293–303 Detection of rolling element bearing defects by adaptive filtering Imed Khemili, Mnaouar Chouchane Laboratoire de génie mécanique, École nationale d’ingénieurs de Monastir, avenue Ibn El Jazzar, Monastir 5019, Tunisie Received 8 January 2004; accepted 5 October 2004 Available online 19 December 2004 Abstract Most of the difficulties encountered when detecting initial bearing defects are due to the presence of noise in the measured signals. In the present paper, we show the effectiveness of using Adaptive Noise Cancellation techniques, in enhancing the vibration signal which is corrupted with noise in the low frequency spectrum. Two algorithms have been used in the paper, the Adaptive Noise Cancellation (ANC) and the Adaptive Self-Tuning filter (AST). The two algorithms have been first of all tested on a computer simulated signals; then, applied on measured signals on a bearing fault simulator. It has been found that adaptive filtering increases the signal to noise ratio for both simulated and measured signals.  2004 Elsevier SAS. All rights reserved. Keywords: Bearing fault detection; Vibration monitoring; Adaptive filtering; ANC; AST 1. Introduction Rolling element bearing faults are among the main causes of breakdown of rotating machines. Condition based maintenance techniques such as vibration analysis, oil and temperature monitoring are usually used for the detection and monitoring of bearing faults. Vibration analysis is undoubtedly the most developed technique for the monitoring and the diagnosis of bearing defects in rotating machines. The detection of the bearing defects using vibration spectral analysis is based on the identification and monitoring of the spectral components generated by the defects. The frequencies of vibration due to defects on inner race, outer race, rolling elements or cage of a rolling element bearing can be computed using a well established formulas. Early detection of theses spectral components may be difficult when the fault size is small or the noise due to other sources of vibration in the machine is high. Techniques which enhance the signal in these situations are therefore highly desirable for early fault detection. This paper concentrates on the application of adaptive filtering techniques for the detection of rolling element bearing faults. Adaptive noise elimination methods, such as adaptive noise cancellation: ANC (Widrow et al., 1975) and adaptive self- Tuning: AST (Shiroishi et al., 1997), constitute a very essential tool in the field of noise filtering and are used in various fields such as mechanics, medicine, telecommunication, etc. They have been applied in elimination of the periodic interferences in the speech signals (Kaunitz, 1972), in cancellation of the periodic interferences in the electrocardiography (Widrow et al., 1975) and in elimination of the interferences in the Hertzian waves (Widrow et al., 1967; Griffiths, 1969). In the field of detecting bearings defects, adaptive filtering has been used by Chaturvedi and Thomas (1981, 1982), the technique was applied on signals measured from a bearing test rig and presented a significant increase in the signal-to-noise ratio. The detection of bearing defects using the ANC algorithm has been also studied by Tan (1985) to extract the components of the signal due to the bearing defects E-mail address: Imed.khemili@isetso.rnu.tn (I. Khemili). 0997-7538/$ – see front matter  2004 Elsevier SAS. All rights reserved. doi:10.1016/j.euromechsol.2004.10.003