Investigation of concrete box girder positive temperature gradient patterns considering different climatic regions

Due to the inevitable exposure of bridges to complex ambient environment, the effects of temperature variation on the structure itself should be considered. Accurate evaluation of the cross-sectional temperature distribution is a significant concern in the bridge design phase. Temperature gradient patterns have been provided by current bridges codes with the consideration of the thermal effects on the structures. However, the current codes are insufficient to describe the temperature gradients of different climatic regions in China. This paper aims to develop a more efficient positive temperature gradient pattern considering different climatic regions within a 100-year return period. Long-term numerical simulations of two-year continuous thermal actions regarding seven different regions have been conducted to investigate the thermal effects on concrete box girders. Then, the comparison of the simulated results and current bridge design specifications illustrates the deficiencies of the current specifications. Furthermore, a 2-step method with curve fitting and extreme values analysis (EVA) has been utilized to determine the temperature gradient patterns. The curve fitting shows that the temperature gradient patterns of different regions can be described by a piecewise function with parabola. And, EVA shows that all the temperature gradients fit well with the Weilbull distribution. The representative values of different regions have also been obtained through the EVA. Finally, a positive temperature gradient pattern with the consideration of both the locations and design period has been proposed in this paper. The proposed temperature gradient pattern shows a better applicability in different regions comparing with the current Chinese codes. It hopes that the present study could contribute to further application of bridge temperature gradients and the investigation of other infrastructural areas in respect of the temperature gradients.