Indirect assessment of railway track vertical stiffness

Abstract

Track stiffness and track deflection measurements are fundamental to understand the track behaviour and the causes that lead to its underperformance, and ultimately its failure (Berggren, 2009; Puzavac et al., 2012). Although in recent years several techniques and methods have emerged to evaluate and quantify the track support modulus, most of them are difficult to implement (Kerr, 2000) and only perform well at limited speeds, around 60 km/h (INNOTRACK, 2008). Therefore, such procedures may cause disruptions and disturbances in the normal traffic and involve high costs for the railway managers (Berggren et al., 2014). Most of these techniques require the deduction of the axle loads during train passage - a procedure that may lead to divergent interpretations. A recent work by Le Pen et al. (2016) proposes a new method that analyses the measurements in the frequency domain and does not require the measurement of the axle loads of the trains. Because the number of reported applications of this method is still very limited, though some promising results have already been obtained, it is important to keep improving its understanding. Therefore, to obtain more insight into the limitations and strengths of this approach, the authors have applied the abovementioned method to a set of track measurements obtained in earlier experimental works. This work focuses on the conclusions drawn during the application of the method to transition zones in railway tracks. In previous studies, this method seems to have only been applied to railway tracks on low stiffness soils and at sites without abrupt differences in vertical stiffness. The novelty of this study lies on its application to a wide range of track support conditions, including Under Sleeper Pads (thus different vertical stiffness levels), and different types of trains such as high-speed services, regular passenger trains and freight trains.