Comment

Maintaining light rail

Dr Adam Bevan, Professor Paul Allen and Professor Jay Jaiswal, of the Institute of Railway Research, take a look at new rail asset management techniques in light-rail systems.

The defining characteristics of light-rail systems – such as embedded and ballasted track formations, small curve radii, steep gradients and frequent station stops – result in a very challenging operating environment. This, coupled with a large variation in route and vehicle characteristics between different networks, an absence of relevant standards and guidance, and limited system availability for maintenance action, means that effective management of the wheel-rail interface becomes a critical topic.

Although degradation to the wheel-rail asset is caused primarily by the contact conditions at the interface, these are strongly influenced by a wider range of parameters, including vehicle and operational characteristics, trackform type and construction, wheel-rail profile shape, metallurgy, and maintenance interventions. Therefore, effective management of these assets must be based on knowledge of the expected primary degradation mechanisms and the rate at which asset life is consumed. This can be achieved through the application of a range of new techniques that combine simulation models, measured condition data and intelligent data analytics.

Rail asset management techniques

Typically, rail will experience different types and rates of degradation across a given network, resulting in a range of expected lifespans. Therefore, it is essential to segregate track sections into ‘sub-assets’ based on an expected rate of degradation and lifespan. This approach not only supports the effective management of each sub-asset, but also helps to identify differences in performance between assets with similar characteristics.

Applying the sub-asset approach helps manage differing maintenance regimes across a network. For example, rails located across different installation types such as plain line, switches and crossings, level crossings, tunnels, and bridges can experience differing degradation mechanisms that may shorten their life. Therefore, effective management of the rail asset can be achieved by discretising a network into sub-assets based on transitional boundaries that depend on installation type. Further segregation can then be applied based on track curvature, allowing the key element of wheel-rail contact characteristics and consequential wear and damage modes to be captured.

The susceptibility of sub-assets to the various rail degradation mechanisms (e.g. wear, rolling contact fatigue, corrugation and plastic deformation) are determined. These susceptibilities can be established through a combination of data-driven and model-based approaches:

• Data-driven approach: Condition and degradation rate of each sub-asset is established from intelligent assessment and trending of available inspection and monitoring data. This provides the opportunity for improved maintenance planning by triggering a maintenance action based on the actual condition of the asset (limit-based). The data can also support the validation of prediction models and helps to evaluate changes in the performance of a sub-asset that would otherwise be difficult to model (e.g. rail corrugation);
• Model-based approach: Dynamic simulation of vehicle/track interaction employing the range of track and vehicle characteristics representative of the network in question. Subsequent prediction of degradation rates of each sub-asset for the relevant damage modes to support the interpretation of measured condition data and analysis of potential mitigation measures (e.g. rail steel grade selection, changes to track alignment and wheel-rail profiles).

To establish the condition of the rail asset, a large volume of condition data is typically acquired by light-rail operators and maintainers. However, recent advances in non-destructive testing technology provides the opportunity to enhance the early detection of rail degradation. From a rail perspective, this has focused on three main areas of interest:

1. Measurement of the level of rail wear is often conducted using manual devices, for example the encoder-arm based MiniProf profile measuring system. However, new techniques such as laser scanners or 3D cameras, mounted on in-service vehicles, can now be deployed to provide more efficient measurement methods;
2. Rail corrugation is a complex mechanism which can result in an increase in noise, vibration and dynamic vehicle-track loading. Detailed assessment of the characteristics of rail corrugation is often undertaken using a specialist corrugation assessment trolley. An alternative system based on the use of wireless microelectromechanical Systems based acceleration measurements provides a potential solution for early identification of track segments that are more susceptible to corrugation and any changes in local severity;
3. An objective assessment of the surface condition of the wheel and rail is generally conducted through manual inspections. However, a requirement for daylight inspection compromises the demand for higher service frequencies, with further measurement challenges presented by pedestrian and road traffic in shared infrastructure areas. This highlights the need for increased adoption of vehicle-mounted techniques, such as linescan cameras, coupled with automated analysis of the acquired images to identify running surface defects, thereby providing a more efficient inspection solution. Image processing and feature extraction algorithms have been developed for automated detection and classification of rail surface damage.

Information on current asset condition and predicted degradation rates is used to define the expected lifespan for each segment, prioritise the maintenance actions (e.g. list of segments to be ground and/or weld restored), and provide a more objective estimation of future rail management budgets. 

An example output for rail life estimation of a track segment is where a combination of inspection and prediction data is used to estimate the life of the rail due to wear. Further detailed assessment of these outputs also helps to identify the key drivers for rail renewal (e.g. vertical, side or keeper wear) and therefore support the future specification of the wheel-rail interface characteristics to optimise rail life. This might include a selection of an alternative rail steel grade to increase resistance to wear, or specifying a rail section with a larger groove width to increase the allowable rail side wear before keeper contact occurs.

Further research work is also being conducted to develop criteria for optimal rail steel grade selection that considers rail degradation mechanisms encountered in various parts of the network, the performance of new and available steel grades, and maintenance practices.

The replacement of embedded rails is disruptive and costly, and therefore in-situ weld restoration of side-worn grooved rails is desirable. However, a number of challenges still exist in relation to the integrity of the weld repair and the pre-heating process. ARR Rail Solutions recently developed a bespoke weld restoration unit that incorporates several unique features to ensure consistency and integrity of the weld deposit on both standard and premium grade rails. Use of novel asset management techniques can be used to support the future planning of weld restoration.

Enjoying RTM? Subscribe here to receive our weekly news updates or click here to receive a copy of the magazine!