Proposition of a bearing diagnosis method applied to IAS and vibration signals: The BEAring Frequency Estimation Method. (15th March 2023)

• Record Type: Journal Article
• Authors: Bertoni, R.; André, H.
• Is Part Of: Mechanical systems and signal processing. Volume 187(2023)
• Publication Date: 2023-03-15
• Subjects: Vibration monitoring -- Instantaneous angular speed -- IAS -- Elapse time measurement -- Bearing fault diagnosis -- RDA -- Characteristic frequency estimation -- BEAFEM -- Helicopter transmission; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2022.109891
• Languages: English
• ISSNs: 0888-3270

Abstract

Abstract: Early detection of bearing faults has been a major topic of industrial research for several years now. Although new methods are regularly tested, accelerometry remains the most widely used because of its ease of implementation and its cost. However, the use of accelerometers reaches its limits when the background noise is very high (combustion noise, kinematic frequencies, modal response of the structure), or more prosaically when the environmental conditions do not allow instrumentation near the bearing (temperature, space available). Another disadvantage lies in the need to be as close as possible to the monitored bearing, implying the use of numerous sensors if the drive train includes several bearings. In this paper, we take advantage of the high sensitivity of incremental encoders to kinematic phenomena to detect the early appearance of a bearing fault on a helicopter drive train. First, the deterministic components related to the kinematic chain are subtracted from the signal. Then, a fault detection method which respects the kinematic coherence between the characteristic frequencies of the bearing is proposed: the BEAring Frequency Estimation Method (BEAFEM). This approach seeks to maximize the sum of the spectral components whose frequency exactly corresponds to the characteristic frequency of the bearing, by scanning each potential contact angle. It is shown that the method can be applied indifferently on IAS signals (using CS1 approach) or on vibration signals (using CS2 approach) obtained with accelerometers. Additionally, it is shown that BEAFEM provides convincing results when applied to an encoder signal located further down the kinematic chain. Highlights: Bearing defect monitoring using vibration and IAS on a complex kinematic chain. New post-processing method proposition, based on the characteristic frequency tracking. IAS and vibration-based methods lead to different though promising results. Dual observation of amplitude and characteristic frequency is valuable for prognosis. A bearing fault located three pinions pairs away from the sensor can be detected. Graphical abstract: