Comparison of Field Behavior with Results from Numerical Analysis of a Geosynthetic Reinforced Soil Integrated Bridge System Subjected to Thermal Effects

Abstract

When subjected to ambient daily temperature fluctuations, a 109.5 ft-long geosynthetic reinforced soil integrated bridge system (GRS-IBS) was observed to undergo cyclic straining of the superstructure. The upper and lower reaches of the superstructure experienced the highest and lowest strain fluctuation, respectively. These non-uniform strains impose not only axial loading of the superstructure but also bending. Pure axial loading in a horizontal superstructure will cause the footings to slide. However, bending in the superstructure will cause the footings to rotate thereby inducing cyclic fluctuations of the vertical pressure beneath the footing and also lateral pressure behind the end walls. Measured vertical footing pressure closest to the stream experienced the greatest daily pressure fluctuation (≈ 2,500–3,000 psf), while that nearest the end wall experienced the least. The toe pressure fluctuations seem rather large. That these large vertical pressure fluctuations are observed in a tropical climate like Hawaii when no other GRS-IBS in temperate regions has reported the same (or perhaps higher fluctuation) is indeed surprising. The larger these pressures are, the greater the likelihood of inducing cyclic-induced deformations of the GRS abutment. A finite element analysis of the same GRS-IBS was performed by applying an equivalent temperature and gradient to the superstructure over the coldest and hottest periods of a day to see if the field measured values of pressures are reasonable and verifiable, which indeed they were. This methodology is novel in the sense that the effects of axial load and bending of the superstructure are simulated using measured strains rather than measured temperatures.

Statistical Analyses of the Non-Uniform Longitudinal Temperature Distribution in Steel Box Girder Bridge

“…Although this assumption simplifies the analysis of bridges affected by temperature in the two-dimensional plane, it may overestimate the temperature effects at some spanwise locations and underestimate them at others. Moreover, the nonlinear effects caused by temperature effects are neglected by the uniform-distribution assumption [31,32]. Therefore, the temperature effects cannot be accurately evaluated when adopting the above assumption.…”