PSI - Issue 44

Rebecca Asso et al. / Procedia Structural Integrity 44 (2023) 894–901 Rebecca Asso et al./ Structural Integrity Procedia 00 (2022) 0 0–000

900

7

Figure 7: Finite Element of the bridge

As shown in Table 3, the model is loaded with its self-weight, the dead load of all the non-structural elements, and two components of the thermal load that is acting on the structure, one uniform and one varying along the cross section. National and international standards specify the thermic load in great detail (Eurocode 2004). However, the thermal load, is in this case determined with the obtained data, where the uniform variation of temperature is set as 10.5 °C and the gradient of temperature is calculated with the output of the Dataset 1 from the given time-frame. The temperature gradient, was in fact analyzed using data points from the pre-retrofitted structure to determine the difference between the obtained temperature from the sensors along the box girder height. This parameter is calculated based on the distance of the two instruments (1.90 m) which led to a temperature gradient ( ∇ T) equals to 1.10 °C/m. Thermal distortion was examined in its two components indicated above, rather than through the application of the fifth-degree rule (Rendace 2021).

Table 3: Applied loads: permanent, non-permanent and thermal actions

∇ T

*Permanent Load (g 1 )

Non-Permanent Load (g 2 )

Thermal Action

g 1

g 2,p

g 2,c

g 2,s

g 2,t

g 2,pv

T o

ΔT

[kN/m 3 ]

[kN/m]

[kN/m]

[kN/m]

[kN/m]

[kN/m]

[°C] 15.0

[°C] 10.5

[°C/m]

Value

25.0

22.9

14.0

14.0

14.0

5.6

1.1

*The permanent load is automatically calculated by the software.

1.5. Results and discussion Comparing the results obtained from the data analysis it is possible to highlight the influence of the bearings substitution in the abutments. In fact, the temperature variation that is applied to the FE model provides displacements of 7.2 mm in pile 1 and 14.6 mm in pile 2, although the pre-retrofitted dataset (Dataset 1) is showing a displacement of 23.4 mm in pile 1 and 29.85 mm in pile 2. This first comparison confirms the presence of a non-defined boundary of the bridge before the substitution of the bearings. When post-intervention data is analyzed (Dataset 2) the results obtained from the sensors in both piles show a noticeable reduction, a total displacement of 13.3 mm in pile 1 and 18.8 mm in pile 2. In Figure 8 it is shown a comparison of the findings from the previous mentioned analysis, demonstrating that the displacements caused by the unknown static scheme configuration were minimized for both piles after the retrofitting technique. In terms of post-retrofitting outcomes, MIDAS Civil (MIDAS Information Technology Co. 2022) outputs show a relatively good agreement with respect to post-retrofitting results. However, the results are expected to slightly vary towards the ideal behavior in as the bridge adjusts its response to the new bearings in time.