Reliability-base monitoring and maintenance of urban railway turnout using acoustic emission

Kongpuang, Manwika (2022). Reliability-base monitoring and maintenance of urban railway turnout using acoustic emission. University of Birmingham. Ph.D.

Preview

Kongpuang2022PhD.pdf
Text - Accepted Version
Available under License All rights reserved.
Download (14MB) | Preview

Abstract

Defects such as fatigue cracks and impact damage can develop over time in service urban railway turnouts. In addition, manufacturing defects during the casting or welding process can also arise occasionally, resulting in structural failure at a much earlier stage. Cracks initiating and propagating below the surface are difficult to detect using conventional ultrasonic testing. Even if the application of radiography for the inspection of cast manganese in the field was straightforward, the presence of internal cracks could be easily missed due to the limitations of this technique in detecting relatively small cracks. A more appropriate technique for crack growth monitoring is Acoustic Emission (AE). When a load is applied to a solid structure (e.g., internal pressure or external mechanical means), it begins to deform elastically. Changes in the structure’s stress distribution and storage of elastic strain energy are associated with this elastic deformation. As the load increases further, some permanent microscopic deformation or crack growth may occur, accompanied by a release of stored energy, partly in the form of propagating elastic waves termed ‘Acoustic Emission’. If these emissions are above a certain threshold level, they can be detected and converted to voltage signals by sensitive piezoelectric transducers mounted on the structure’s surface. A typical AE system consists of signal detection, amplification, data acquisition, processing and analysis. Various parameters are used in AE to identify the nature of the source, including waveform, count, duration, amplitude, rise-time, energy, frequency and RMS (Root Mean Square).
An important aspect of AE testing is signal processing. There is a need to separate genuine stress-wave emissions originating from within the material from external signals, such as environmental noise (e.g. rain), mechanical noise (rolling stock noise from the wheels in this case), electric noise, etc. Much of this is achieved by careful electronic filtering of the received AE data. The frequency of the stress waves emitted is generally 30 kHz to 1 MHz. Linear location or triangulation can give positional information and localise the emissions' sources and amount of crack growth.
This study investigates the applicability of quantitatively monitoring the structural degradation of urban railway turnouts using AE. A metallurgical investigation coupled with mechanical testing was carried out. Develop customised AE techniques that can be installed in the field to detect any crack growth or impact damage. During crack growth, stress waves are released in every direction, with some detected by the AE sensors. Since the operational sensor frequency was selected between 100 - 250 kHz, the significance of the scattering effect from the large-grain microstructure of cast manganese steel was much lower than in ultrasonic testing.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):