Efficient calculation and monitoring of temperature actions on supertall structures

Abstract

Numerical heat transfer analysis and field monitoring have been developed to investigate the effects of varying temperature on supertall buildings. The conventional heat transfer analysis studies one or several components of a structure each time, causing the calculated temperature of the entire structure inaccurate. Moreover, the finite element (FE) model used for calculating temperature distribution cannot be directly used for computing the temperature-induced responses of the structure, which requires considerable manual inputs of the temperature load. This paper presents an automatic and efficient FE approach to calculating the temperature distribution and the associated responses of an entire structure using field meteorological monitoring data. The position of the sun relative to the structure can be determined by introducing a new radiation calendar timing system. A virtual sun is then created to determine the irradiation and shade elements of the structural model, from which the solar radiation intensity on the surfaces of all elements can be calculated at any particular time on any particular day. Consequently, the dynamic thermal boundary conditions of the FE model are formulated automatically. This enables the heat transfer analysis of the entire structure to be conducted and the temperature distribution of the entire structure to be calculated in real time. The calculated temperature distribution is transferred to the temperature load in the same FE model, and the temperature-induced stress and displacement responses of the structure can be obtained. The method is applied to the 335 m tall Wuhan Yangtze River Navigation Centre. A 3D solid FE model of this structure during the construction stage is established. Varying wind speed and air temperature along the height of the structure are taken into account from the SHM system. The calculated temperature distribution and temperature-induced stress of the structure are in good agreement with the field monitoring data. The proposed technique offers an effective and efficient real-time monitoring of the temperature actions on large-scale structures.