Keywords: field experiments, switches and crossings, condition monitoring.

Switches and crossings are the weakest point in the rail infrastructure, with a high rate of failure in comparison with plain lines arising from the amplified dynamic forces that occur as vehicles negotiate through their complex rail geometries. The switch and crossing assets require heavy investment to maintain and renew within the rail infrastructure. Maintenance strategies tend to include a routine inspection followed by corrective maintenance to correct failures or the damage as necessary.

This paper shows a methodology for switches and crossings (S&C) maintenance moving towards a predictive maintenance using condition monitoring. Several sites have been instrumented around the United Kingdom to follow the condition of the assets and different components. The instrumentation method, design of experiments and the results for a single site are shown in the investigation. The various 'design of experiment' parameters and levels are described, with the analysis based on the recorded data at a single site. Strain data and the velocity data are included in the investigation.

Data analysis was completed in two phases, firstly collecting field data, which included strains through the rail, and the velocity of the adjoining structure. This data has given a good understanding of the material degradation over time, and under certain loading conditions. The analysis will continue into the second phase, which will include signal processing to understand the general responses of the structure to the dynamic loading of the wheel-rail contact.

Preliminary instrumentation results of the field experimentation are shown. There are high peaks of strain arising from the vehicle entering the higher magnitude zone of the S&C, shown in the peak to peak and mean strain levels. The mean values of strain are maximum vertically at the switch tips (with the switch blade closed against the stock rail) for the through route.

Investigation of strains and velocities are important to determine the trends and current values that are being seen within the infrastructure under loading. The use of this data will continue to develop through trend analysis. Current limits and tolerances are being recorded for future research to develop triggers for condition monitoring limits and maintenance guidelines.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £110 +P&P)