Abstract Summary
For railway bridges on high-speed lines, dynamic analyses are usually required to ensure that vibrations from passing trains do not exceed certain limits, e.g. regarding the vertical deck acceleration. Using over-simplified models may lead to inaccurate results that sometimes show theoretical exceedance for bridges that are not susceptible to dynamic loading. This may cause unnecessary changes in the design of new bridges or speed restrictions on existing bridges. This paper presents an efficient 2D beam model of a continuous single-track concrete slab bridge with span lengths of 13 + 17 + 13 m. At the end supports, the deck has a 1.5 m overhang with an integral structure in interaction with the adjacent embankment. The columns are rigidly connected to the bridge deck. All supports are built on slab foundations, and one of the supports sits on a stratum of sandy silt that will affect the dynamic behaviour. Experimental testing has been performed using a hydraulic exciter with a controlled frequency and a known input. The resulting frequency response functions (FRF) clearly show the first three vertical bending modes. Testing with increased load amplitude shows a trend of decreased natural frequency and increased damping, especially for the first mode. The response from passing trains was also recorded. The geometry of the 2D model is based on design drawings and geotechnical documents. The boundary conditions are updated based on the experimental FRFs. The model is then used to simulate the response from passing trains and compared with the experimental data. Good agreement is generally found between the 2D model and the experimental data, both regarding the FRFs and passing trains. The model further shows the importance of accounting for soil-structure interaction in dynamic analysis, otherwise excessive vibrations may occur.
