Passenger and freight rolling stock traffic density has increased considerably in recent years in the majority of rail networks across the world. At the same time travelling speeds and axle loads have also increased putting considerable pressure to maintenance engineers who are tasked with ensuring reliability and safety of railway network operations. Inspection of rail infrastructure using various types of methods and equipment has traditionally been the cornerstone of evaluating the structural integrity of critical structural components such as rails, crossings, tunnels and bridges. However, as traffic density increases further and the likelihood of round the clock railway operations nears closer to realisation, the opportunities for inspection based on conventional approaches are reduced by a significant margin. It is therefore, important to develop alternatives for the evaluation of the structural integrity of critical railway infrastructure assets before the technical challenges currently faced manifest themselves more imperatively. In this paper, a qualitative comparison of the key advantages and disadvantages of structural health monitoring techniques over conventional inspection methods are presented and analysed from an infrastructure manager perspective. The discussion focuses on how realistic is the replacement of conventional inspection methods with structural health monitoring techniques currently and what steps need to be taken before a change in the status quo can occur.

The fact that critical structural components such as rails and crossings are randomly loaded increases the degree of uncertainty when trying to estimate their remaining service lifetime. Maintenance decisions are predominantly based on the feedback received from inspection engineers coupled with empirical knowledge that has been gained over the years. The use of structural degradation models is too risky due to the uncertainty arising from the variable dynamic loads sustained by the rail track. The use of structural health monitoring techniques offers significant advantages over conventional approaches. First of all, it is non-intrusive and does not interrupt normal rail traffic operations. Secondly, defects can be detected and evaluated in real-time whilst their evolution can be monitored continuously enabling maintenance to be scheduled in advance and at times where the need for rail network availability at the section concerned is at its lowest. This paper analyses the potential risks and benefits of a gradual shift from traditional inspection approaches to advanced structural health monitoring techniques.

Operational efficiency is one of the key performance indicators for all railroad systems. Infrastructure inspection and maintenance engineers are tasked with the responsibility of ensuring the reliability, availability, maintainability and safety of the railroad network. However, as rolling stock traffic density increases throughout the network, inspection and maintenance opportunities become less readily available. Inspection and maintenance activities normally take place at night, when there is little or no train movement to avoid disruption of normal railroad network operation. In addition, conventional inspection methodologies fail to deliver the efficiency required for the optimization of maintenance decisions, particularly with respect to track renewals, due to their defect detection sensitivity and level of resolution limitations. The fact that critical structural components such as rails and crossings (frogs) are randomly loaded increases the degree of uncertainty when trying to estimate their remaining service lifetime. Maintenance decisions are predominantly based on the feedback received from inspection engineers, coupled with empirical knowledge that has been gained over the years. The use of structural degradation models is too risky due to the uncertainty arising from the variable dynamic loads sustained by the rail track. The use of structural health monitoring techniques offers significant advantages over conventional approaches. First of all, it is non-intrusive and does not interrupt normal rail traffic operations. Secondly, defects can be detected and evaluated in real-time whilst their evolution can be monitored continuously, enabling maintenance to be scheduled in advance and at times where the need for rail network availability at the section concerned is at its lowest. This paper analyzes the potential risks and benefits of a gradual shift from traditional inspection approaches to advanced structural health monitoring techniques.

Structural degradation of rails will unavoidably take place with time due to cyclic bending stresses, rolling contact fatigue, impact and environmental degradation. Rail infrastructure managers employ a variety of techniques and equipment to inspect rails. Still tens of rail failures are detected every year on all major rail networks. Inspection of the rail network is normally carried out at night time, when normal traffic has ceased. As the implementation of the 24-h railway moves forward to address the increasing demand for rail transport, conventional inspection processes will become more difficult to implement. Therefore, there is an obvious need to gradually replace outdated inspection methodologies with more efficient remote condition monitoring technology. The remote condition monitoring techniques employed should be able to detect and evaluate defects without causing any reduction in the optimum rail infrastructure availability. Acoustic emission is a passive remote condition monitoring technique which can be employed for the quantitative evaluation of the structural integrity of rails. Acoustic emission sensors can be easily installed on rails in order to monitor the structural degradation rate in real time. Therefore, apart from detecting defects, acoustic emission can be realistically applied to quantify damage. In this study, the authors investigated the performance of acoustic emission in detecting and quantifying damage in rail steel samples subjected to cyclic fatigue loads during experiments carried out under laboratory conditions. Herewith, the key results obtained are presented together with a detailed discussion of the approach employed in filtering noise sources during data acquisition and subsequent signal processing.

The early detection of faults in rolling stock wheels and axle bearings is of paramount importance for rail infrastructure managers as it contributes to the safety of rail operations. In this paper we report on the key results that have arisen from the development and implementation of a novel condition monitoring system based on high-frequency acoustic emission and vibration analysis installed on a train. The novel system makes use of inexpensive and robust acoustic emission sensors and accelerometers, which can be easily installed on the axle bearing box with minimal intervention required. Experimental work carried out under actual conditions at the Long Marston rail track and on the Lisbon – Cas-Cais suburban line has proven that the developed system is capable of detecting wheel and axle bearing-related defects with various levels of severity.

The use of wayside monitoring for the detection of serious faults in rolling stock is of paramount importance for rail infrastructure managers as it contributes to the safety of rail operations. In this paper we report on the key results that have arisen from the development and implementation of a novel integrated wayside condition monitoring system based on high-frequency acoustic emission and vibration analysis which can be interfaced with existing wayside systems such as hot boxes and wheel impact load detectors. The novel system makes use of inexpensive and robust acoustic emission sensors and accelerometers which can be easily installed on the rail track with minimal intervention involved. Experimental work carried out under actual conditions in Long Marston test rail track has proven that the developed system is capable of detecting wheel and axle bearing related defects with various levels of severity.