Switches are an essential, safety-critical part of the railway infrastructure. Compared to open tracks, their complex geometry leads to increased dynamic loading on the track superstructure from passing trains, resulting in high maintenance costs. To increase efficiency, condition monitoring methods specific to railway switches are required. A common approach to track superstructure monitoring is to measure the acceleration caused by vehicle track interaction. Local interruptions in the wheel–rail contact, caused for example by local defects or track discontinuities, appear in the data as transient impact events. In this paper, such transient events are investigated in an experimental setup of a railway switch with track-side acceleration sensors, using frequency and waveform analysis. The aim is to understand if and how the origins of these impact events can be distinguished in the data of this experiment, and what the implications for condition monitoring of local track discontinuities and defects with wayside acceleration sensors are in practice. For the same experimental configuration, individual impact events are shown to be reproducible in waveform and frequency content. Nevertheless, with this track-side sensor setup, the different types of track discontinuities and defects (squats, joints, crossing) could not be clearly distinguished using characteristic frequencies or waveforms. Other factors, such as the location of impact event origin relative to the sensor, are shown to have a much stronger influence. The experimental data suggest that filtering the data to narrow frequency bands around certain natural track frequencies could be beneficial for impact event detection in practice, but differentiating between individual impact event origins requires broadband signals. A multi-sensor setup with time-synchronized acceleration sensors distributed over the switch is recommended.

Switches and crossings (S&Cs) are also known as turnouts or railway points. They are important assets in railway infrastructures and a defect in such a critical asset might lead to a long delay for the railway network and decrease the quality of service. A squat is a common rail head defect for S&Cs and needs to be detected and monitored as early as possible to avoid costly emergent maintenance activities and enhance both the reliability and availability of the railway system. Squats on the switchblade could even potentially cause the blade to break and cause a derailment. This study presented a method to collect and process vibration data at the point machine with accelerometers on three axes to extract useful features. The two most important features, the number of peaks and the total power, were found. Three different unsupervised machine learning algorithms were applied to cluster the data. The results showed that the presented method could provide promising features. The k-means and the agglomerative hierarchical clustering methods are suitable for this data set. The density-based spatial clustering of applications with noise (DBSCAN) encounters some challenges.

Railway switches and crossings (S&C) are among the most important high-value components in a railway network and a failure of such an asset could result in severe network disturbance. Therefore, potential defects need to be detected at an early stage to prevent traffic-disturbing downtime or even severe accidents. A squat is a common defect of S&Cs that has to be monitored and repaired to reduce such risks. In this study, a testbed including a full-scale S&C and a bogie wagon was developed. Vibrations were measured for different squat sizes by an accelerometer mounted at the point machine. A method of processing the vibration data and the speed data is proposed to investigate the possibility of detecting and quantifying the severity of a squat. One key technology used is wavelet denoising. The study shows that it is possible to monitor the development of the squat size on the rail up to around 13 m from the point machine. The relationships between the normalised peak-to-peak amplitude of the vibration signal and the squat depth were also estimated.

Eddy current (EC) testing is the most commonly used method to inspect near-surface cracks in railheads. Monitoring surface defects periodically is important to assess the track quality for serving daily operations. Nevertheless, despite being used in many countries, this method has limitations when characterizing cracks under the rail surface. Theoretically, EC testing is unreliable for the inspection of many cracks situated too close to each other in a concentrated location. This study has aimed to prove these limitations. EC signals from inspected cracks were compared with real crack profile parameters, i.e. depth and area, which were delivered by slicing the inspected cracked spots into 0.65 mm-thick pieces. The results show that the EC signal responses to the parameters of area and depth may lead to misleading measurements of the near-surface crack depth in the railhead. For instance, a shallower crack with a larger area can generate a higher EC signal response than a deeper crack with a smaller area. Another important conclusion is that the EC testing in this experiment could not be used to measure densely located cracks, which are those near-surface cracks which are typically found in a rail track. 

The friction forces between the wheel and rail depend on a number of variables including the third body layer at the wheel–rail interface, the wheel and rail profiles, and the train dynamics. The third body layer significantly influences the damage mechanisms at the wheel-rail interface, especially wear, rolling contact fatigue (RCF), corrugations and other surface defects that then require maintenance. The introduction of additional constituents at the wheel–rail interface in the form of an additive with anti-wear and anti-crack properties can reduce the wear and RCF. In general, such an additive also reduces the friction. However, it is important to avoid the friction coefficient between the wheel tread and the top of the rail falling below 0.3 because the result would be wheel slip and long braking distances. Measuring friction coefficients accurately is still a challenge, as most existing tribometers are unable to replicate the wheel-rail contact conditions, specifically the contact pressure and sliding speed. The present study used a newly designed handheld tribometer that is able to match the typical contact pressure. Results obtained with the handheld tribometer have been compared with values extracted from the traction-force measurement system of a locomotive. The tribometer field measurements have shown that by using a top-of-rail friction modifier (TOR-FM), both the wear and the friction coefficients can be reduced, but also that heavy TOR-FM films may cause unacceptably low friction. Comparing the results of field and laboratory tests confirms that weather and realistic third bodies present on the track have a significant effect on friction and wear. © IMechE 2021.

The switch and crossing (S&C) is one of the most important parts of the railway infrastructure network due to its significant influence on traffic delays and maintenance costs. Two central questions were investigated in this paper: (I) the first question is related to the feasibility of exploring the vibration data for wear size estimation of railway S&C and (II) the second one is how to take advantage of the Artificial Intelligence (AI)-based framework to design an effective early-warning system at early stage of S&C wear development. The aim of the study was to predict the amount of wear in the entire S&C, using medium-range accelerometer sensors. Vibration data were collected, processed, and used for developing accurate data-driven models. Within this study, AI-based methods and signal-processing techniques were applied and tested in a full-scale S&C test rig at Lulea University of Technology to investigate the effectiveness of the proposed method. A real-scale railway wagon bogie was used to study different relevant types of wear on the switchblades, support rail, middle rail, and crossing part. All the sensors were housed inside the point machine as an optimal location for protection of the data acquisition system from harsh weather conditions such as ice and snow and from the ballast. The vibration data resulting from the measurements were used to feed two different deep-learning architectures, to make it possible to achieve an acceptable correlation between the measured vibration data and the actual amount of wear. The first model is based on the ResNet architecture where the input data are converted to spectrograms. The second model was based on a long short-term memory (LSTM) architecture. The proposed model was tested in terms of its accuracy in wear severity classification. The results show that this machine learning method accurately estimates the amount of wear in different locations in the S&C.

This study is motivated by the need for a new advanced vibration-based bearing monitoring approach. The ARX-Laguerre model (autoregressive with exogenous) and genetic algorithms (GAs) use collected vibration data to estimate a bearing’s remaining useful life (RUL). The concept is based on the actual running conditions of the bearing combined with a new linear ARX-Laguerre representation. The proposed model exploits the vibration and force measurements to reconstruct the Laguerre filter outputs; the dimensionality reduction of the model is subject to an optimal choice of Laguerre poles which is performed using GAs. The paper explains the test rig, data collection, approach, and results. So far and compared to classic methods, the proposed model is effective in tracking the evolution of the bearing’s health state and accurately estimates the bearing’s RUL. As long as the collected data are relevant to the real health state of the bearing, it is possible to estimate the bearing’s lifetime under different operating conditions.

The application of top-of-rail friction modifiers (TOR-FMs) is claimed by their manufacturers as a well-established technique for minimising the damages in the wheel–rail interface. There are various methods for applying friction modifiers at the wheel–rail interface, among which stationary wayside systems are recommended by TOR-FM manufacturers when a distance of a few kilometres is to be covered. An on-board system is recommended when an area of many kilometres has to be covered and focus is more on particular trains. Trafikverket in Sweden is considering the implementation of the TOR-FM technology on the iron ore line. Directly implementing such technology can be inappropriate and expensive, because the life cycle cost of a TOR-FM system has never been assessed for the conditions of the iron ore line. In the present study, the life cycle cost is calculated for wayside and on-board application systems, by taking inputs from the research performed on iron ore line. The present research has taken the iron ore line as a case study, but the results will be applicable to other infrastructure with similar conditions. The results have shown that the wayside equipment is economically unfeasible for the iron ore line. In this case, the life cycle cost increases by 4% when the friction modifier is applied on all curves with a radius smaller than 550 m and by 19% when the friction modifier is applied on all curves with a radius smaller than 850 m. The on-board system used in this study is shown to be economically feasible, as it has a significantly lower operation and maintenance cost than the wayside equipment. The reduction in the maintenance (grinding and rail replacement) cost when the cost of the friction modifier application is added is 27% when the friction modifier is applied on curves with a radius smaller than 550 m and 23% when the friction modifier is applied on curves with a radius smaller than 850 m.

In this study, near-surface cracks in a railhead are inspected thoroughly using phased array ultrasonic testing (PAUT). This research finds an alternative technique to inspect for near-surface cracks because the conventional non-destructive testing method for rail inspection lacks the capacity to inspect the near-surface crack profile. This study shows that PAUT can determine not only the crack depth but also the near-surface crack profile, so that the inspector can estimate the stage of crack growth and how the crack propagates. This information is valuable to the rail maintainer as one of the considerations for deciding the thickness of metal to remove when grinding the rail. In this study, after the measurement, the inspected region of the cracked railhead is sliced into thin pieces so that crack network information can be extracted. A 3D image reconstruction of the surface cracks based on the crack marks from all of the sliced rail pieces is performed. This image is then used as a reference to confirm the PAUT results. The results show that PAUT can clearly deliver crack profile estimation and provide an accurate estimation of a 3.51 mm crack-tip depth with an absolute error range of 8%-18%. The results also suggest that PAUT is a potential method for installation in a measurement train for near-surface crack inspection.

The aim of this study has been to develop a two-level multi-objective genetic algorithm (MOGA) to optimize risk-based LCC analysis to find the optimal maintenance replacement time for road tunnel ventilation fans. Level 1 uses a MOGA based on a financial risk model to provide different risk percentages, while level 2 uses a MOGA based on an LCC model to estimate the optimal fan replacement time. Our method is compared with the approach of using a risk-based LCC model. The results are promising, showing that the risk-based LCC offers the possibility of significantly reducing the maintenance costs of the ventilation system by optimising the replacement schedule by considering the risk costs. The risk-based LCC can be used with repairable components, making it applicable, useful and implementable within Swedish Transport Administration (Trafikverket). In this study, MOGA operators have selected the cost of maintenance and risk data through the previous levels using different ways to provide different possible solutions. A drawback of the MOGA based on a risk-based LCC model with regard to its estimation is that a late replacement period over 20-year period might increase the maintenance cost. Therefore, the MOGA does not provide a good solution for a risk-based LCC.