Prediction of vertical temperature gradient on concrete box-girder considering different locations in China

Since the bridges are subjected to complex environmental effects for a long period, it is important to have an accurate evaluation of vertical temperature gradient in box-girders for the assessment of the ambient thermal actions on the structure itself. However, the current empirical equations seem to be inadequate to describe temperature gradients of different regions over China. In order to develop a more appropriate prediction equation of concrete box-girders' positive temperature gradient to different regions over China, long-term simulation of continuous temperature variations in concrete box-girders located in different regions have been conducted. The Finite Element Method (FEM) simulation model has been validated by experimental data documented in literature. The numerical results of temperature gradients have been compared with predicted temperature gradients using different empirical equations. Then, a parametric analysis regarding the meteorological factors and location factors has also been performed. Based on the numerical simulations, a prediction equation including the correction coefficients of elevation and latitude has been proposed for the prediction of box-girders' temperature gradient and a comparison between the predicted and simulated temperature gradients has been conducted. The error analysis shows that the average absolute error and the maximum absolute error are less than 1.0 degrees C and 1.5 degrees C, respectively, which indicates a good agreement between the predicted and simulated temperature gradients. Thus, the proposed prediction equation of the box-girder's temperature gradient can provide a more accuracy predicted results than previous empirical equations, and it is applicable for different regions in China.

Automated summary

This study focuses on accurately predicting the temperature gradients in concrete box-girders in different regions of China by conducting long-term simulations and comparing results to experimental data and existing empirical equations. The proposed prediction equation, incorporating correction coefficients for elevation and latitude, shows good agreement with simulated results, with average and maximum absolute errors less than 1.0°C and 1.5°C respectively. This improved prediction equation provides more accurate results than current empirical equations and is applicable to various regions in China.