PSI - Issue 63

Jiří Brožovský et al. / Procedia Structural Integrity 63 (2024) 1– 6

3

(and the real structure) must be elastic with peak stresses being considerably lower than the yield stress of the used material. 3. Model description The model of the studied part of the bridge is shown in Figure 1. The model has 9,398 finite elements and 42,903 unknowns. The isoparametric 8-node finite elements are used. These elements use linear approximation of unknown deformations. The boundary conditions were set up to simulate effects of the rest of the bridge on the element. Their parameters were defined on basis of a global analysis of a beam model of the bridge (a continuous beam model which respected the global geometry of the structure as it is not exactly straight). Material properties were defined as linear elastic, as mentioned above. The steel was considered with the Young's modulus of elasticity for tension or axial compression �� 210 GPa and the concrete with the �� 37.7 GPa. The Poisson rations were used in values 0.3 and 0.2 respectively. 4. Analysis of results The load case studied here was an extreme load located over the center of the detail. Other cases were studied and this one was considered the most influential. In the studied case if was shown that stresses in the studied part are rather slow for and direction but magnitudes higher for the direction (see Figure 2, Figure 3 and Figure 4). The analytical solution which is planned to be used for analysis of crack propagation in the lower flange of the main beam in place expects that the most important normal stress is the � which is in the direction of the main beam axis (labelled “ ” in figures). It was expected that the normal stresses in other directions should have peak values (in areas close to the detail) of about 10% of the stress � .

Fig. 1. Computational model of the structure.