Assessment of the track condition using the Gray Relational Analysis method

Abstract

The article concerns the developed methodology for assessing the technical condition of a tramway track. Thanks to the data collected from multiple tram journeys equipped with an on-board vibration recording system, it was possible to create profiles of crossings through track sections in different technical condition. In order to identify the track condition, an algorithm based on the gray-scale modeling was proposed, and a similarity comparison between the obtained track profiles. A new measure of similarity has been proposed that has not been used so far in gray-scale modeling. The obtained results confirm the applicability of the proposed methodology

Multifractal Analysis of the Common Crossing Track-Side Measurements

“…A common crossing fault prediction with the track-side inertial measurements using non-linear signal processing techniques and regression methods is proposed in the study [23]. An assessment of the track condition from on-board acceleration measurements using the grey relational analysis method for is presented in [24]. The method determines the track condition using a similarity comparison between the obtained track profiles.…”

Monitoring of the technical condition of tracks based on machine learning

“…Our analyses covered data collected over a period of more than two months of operation of a low-floor tram in regular passenger service. Information used in the assessment of the technical condition was based on the root mean square of vibration acceleration (marked in Figure 1 and subsequent figures as: A1), determined for 1 s window over 0 to 100 Hz frequency band, registered on the vehicle body over the first bogie (as a ride comfort measure), 16 as well as on the peak value of vibration acceleration measured on the left side of bogie frame (marked as A2), which is a normative measure of running safety. Furthermore, the general information about rail type and vehicle speed over a given section were taken into account.…”

“…However, in urban transportation systems, track condition monitoring becomes more complicated due to a considerable spread of measurement results, which had been mentioned in literature. 16 This spread on the same track section is due to vehicle velocity, its acceleration (resulting from changeable traffic conditions), number of passengers (vehicle load) or the technical condition of the wheelset. Suggested methods of dealing with this issue can be found in literature.…”

“…Suggested methods of dealing with this issue can be found in literature. 16–18 Moreover, measured values can be strongly influenced by the type of track settlement and not just its technical condition. The type of subgrade is of particular importance, as its susceptibility varies much more than in railway lines.…”

Study of transitional phenomena in rail vehicle dynamics in relation to the reliability and operational state of the continuous welded rail track in terms of rail joints

“…On the other hand, papers [8,15,16] describe the results of experimental and numerical analyses of the impact of the level of the track geometry irregularity values on the dynamic interaction of forces in the wheel-rail contact area for trains and trams. Similarly, on the basis of experimental studies of the tram dynamic interaction with the track, a novel method of assessing the track condition based on the Gray Relational Analysis method was proposed in paper [26]. In the case of monitoring track components using inertial measurements during the passage of a train, the work [18] used the results of measurements from the train axle box to monitor the condition of welded joints.…”