An inverse analysis method to identify maximum overfire temperature based on an improved ant colony algorithm

Abstract

There are many causes of building fire incidents that can manifest in many forms. Overfire temperature is one of the most important factors affecting structural residual deformation, and it is the basic parameter for postfire structural safety evaluations. To study the overfire temperature and the temperature field distribution of a structure under fire, an overfire temperature inversion method based on an improved ant colony algorithm has been proposed, where the entire structure is discretized into multiple zones and the zonal maximum overfire temperature is assumed to be the load action. The finite element software ANSYS is used for developing a numerical model of the structure, and a positive analysis is applied to the mechanical response of the structure under fire. As a result, the residual displacements of each zone are obtained, the minimum differences between the numerical and measured displacements are considered as the target function, the overfire temperature is assumed to be the design variable, the improved ant colony algorithm is utilised to inversely analyse the structure at the overfire temperature, the highest overfire temperature is determined, and the temperature field distribution of the structure under fire is obtained. Furthermore, a scale model fire test of the space grid structure is performed, and the fire temperature and node displacements are recorded. The inverse method is applied to the fire test model to verify its feasibility and effectiveness. The results suggest that the ant colony algorithm has good applicability to structure overfire temperature inverse analysis.