Uso de técnicas de inteligencia artificial para el predimensionamiento óptimo de elementos estructurales metálicos

Abstract

Abstract: The objective of this research work is to develop a mathematical model that allows obtaining a preliminary optimal distribution of metallic structural profiles in a system of unbraced frames (OMF: Ordinary Moment Frame). The model looks for solving the common drawbacks that occur in methodologies of an approximate nature, for which a methodological process is presented in which are identified the respective restrictive variables for the implementation of the optimization algorithm. For this reason, a solution is proposed that can be represented naturally with a method of permutations with repetition defining spaces to search solutions in order to reduce computational expenses; the main objective is to minimize the self-weight of a structural steel frame and consequently the costs in quantities of material. To evaluate its effectiveness, a study was developed where it is concluded that the designs thrown by the tool simultaneously comply with the restrictive variables for the adressed level of design.