Flexural cracking-induced acoustic emission peak frequency shift in railway prestressed concrete sleepers. (1st January 2019)

• Record Type: Journal Article
• Authors: Janeliukstis, Rims; Clark, Andrew; Papaelias, Mayorkinos; Kaewunruen, Sakdirat
• Is Part Of: Engineering structures. Volume 178(2019)
• Page Start: 493
• Page End: 505
• Publication Date: 2019-01-01
• Subjects: Acoustic emission -- Concrete sleepers -- Counts -- Cracks -- Peak frequency -- Shift; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2018.10.058
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: Acoustic emissions are recorded from full-scale tests of prestressed concrete sleepers. Non-significant peak frequencies of emissions are filtered using a modified universal threshold. Maximum peak frequencies are assessed using emission counts and distribution centroid. Peak frequency shifts are nearly linear with distance from the main emission source. Abstract: A novel technique for acoustic emission-based condition monitoring of railway prestressed concrete sleepers under flexural loading is established. The evolution of peak frequency of emissions under increasing loads is studied using data from four emission sensors. It is found that the bulk of emitted peak frequencies are clustered in three bands - [150–300] kHz, [300–460] kHz and [500–800] kHz in all cases of full-scale sleepers tested. Only slight variations of band ranges are observed. A correspondence of acoustic emission counts to a particular peak frequency is established. Not all emissions are damage-induced – most of them have relatively small number of counts, suggesting that those are due to random noise. Thus, it is proposed to filter out the non-significant peak frequencies with the least number of counts by applying a universal threshold rule. The largest proportion of emission counts corresponds to mid-span of the sleepers. It is shown that other acoustic emission sources exhibit a nearly linear shift in maximum values of peak frequency with increasing distance from this largest concentration of acoustic emission events. This novel insight into condition-based acoustic emissions is critical to develop a suitable and efficient technique for monitoring safety-critical railway sleepers and bearers located in a discreet and remote area. It will truly enable preventative, predictive and condition-based track maintenance for railway industry, minimizing cost and environmental impacts.