Optimization of reinforced concrete plane frames using a hybridization of genetic algorithms and the Nelder-Mead algorithm

Abstract

This paper presents an algorithm using CSi Application Programming Interface (CSi API), which allows the automation of the  structural optimization process for plane frames, linking powerful computational tools such as SAP2000v16 and Matlab. In addition, aspects usually ignored in modelling and structural analysis for structural optimization are included, such as the influence of  cracking of concrete elements in their stiffness and the inclusion of a second order analysis. On the other hand, due to the complexity of the objective function, it was necessary to use metaheuristic optimization methods: a hybridization between genetic
algorithms and the Nelder-Mead algorithm, instead of the classical methods used in previous studies, obtaining satisfactory results. The results are expressed in the form of design recommendations and they are compared with those obtained in previous investigations.