PSI - Issue 5

Pavel Ryjáček et al. / Procedia Structural Integrity 5 (2017) 1051 – 1056 Ryjacek/ Structural Integrity Procedia 00 (2017) 000 – 000

1054

4

a

b

Fig. 4. Results of the parametric study: (a) The acceleration of the arch; (b) The acceleration of the main girder.

Based on the dynamic analysis of the bridge during the design process, the improvements of the bridge were proposed and done by the designer. First improvement was focused on the modification of the arch bracing in order to decrease the lateral vibrations, the final form served for the above mentioned study. Also, the application of the amphibolite ballast to increase the bridge weight and application of the ballast in the box girders was done. The last modification (not included in the parametric study above) was the connections of the hangers to each other in the intersections by rubber damping ties. Based on the recommendations and findings in the dynamic analysis, the extensive testing programme was proposed for the bridge verifications. The tests consisted of:  Static load tests with the heavy rail crane GEK80  Dynamic load test consisting of the passing of the bridge with heavy locomotives and coal coaches, of total weight of 974 t,  Modal analysis, based on the vibrations from the passing trains,  Brake and acceleration tests, focused on the verification of the proper function of the MW steering bar device However, this paper is focused only on the dynamic tests results. They were mainly focused on measurement of strains in the bridge, the rail, the measurement of acceleration on the girders, arches and tension bars and evaluation of the basic natural frequencies and mode shapes. The high number of various sensors was used for the dynamic load tests, such as:  acceleration sensors in the vertical and lateral direction on both arches (9x) and girders (9x), in three sections along the bridge,  horizontal acceleration on the selected 7 hangers (7x),  vertical displacement in the midspan of the girders and in the bearings location (2x),  strain gauges on the hangers (7x), main girder in the midspan (4x),  strain gauges on the rail, that were used for the brake and acceleration tests. The comparison between the measured and analyzed mode shapes and natural frequencies is given in the table 2 and Fig. 5. The correlation is quite good. However, it should be noted, that in order to have such a good results, several modeling assumptions were used. First of all, the mass of the ballast and the ballasted bed was measured during the construction of the bridge. Also, even for the heavy trains, just after the train passes the bridge the level of the vibrations is decreasing. During the time period that was used for the evaluation of the natural frequencies the bearings are in the original position and does not move, see Fig. 5. It means that for the numerical model friction is so high, 3. Dynamic load test results and comparison