Strut-and-tie modelling of reinforced concrete beams using genetic algorithms optimization

This paper presents a method for automatically producing optimal strut-and-tie models for the design of reinforced concrete beams. The optimal model is generated by means of an optimization problem solved by using genetic algorithms. The basic idea developed here is that from an initial random generation of possible configurations of the strut-and-tie model for the beam subjected to study, new populations of possible configurations may be generated in an iterative way by using genetic operators until reaching an optimum solution for the studied problem which corresponds to the strut-and-tie configuration which minimizes the total strain energy. In the optimal configuration, compressive struts are not enforced to be parallel, which allows representing more consistently the physical reality of the flow of forces. Furthermore, the method is more simple and easier to apply than the methods based on the concepts of evolutionary structural optimization.