Modelling and calibration of infill frame-buildings, using ambient vibration tests and genetic algorithms

Abstract

The presence of infill walls modifies the global structural response of frame buildings subjected to seismic loads. In evaluation studies, the stiffness of the structural elements and the mass of the building do not have great scattering; consequently, the variable to be analysed is the stiffness of the masonry. This investigation presents a process to determine a more realistic calibration of an existing building model using structural software and genetic algorithm programs made in Python 3 where the properties of the masonry were iterated to obtain modal periods similar to those obtained from ambient vibration tests. This programming can be freely downloaded and can be used in any optimisation case. The results highlighted the influence that non-structural masonry has over the overall behaviour of a building. It is also demonstrated that the use of a simplified macro model, with one spring that depicts the nonlinear performance of masonry, is an excellent option to address the limitations of commonly used software, as well as reducing processing time.