Rational seismic retrofitting of RC structures based on genetic algorithms

This paper presents a rational strategy developed for optimizing seismic retrofitting of Reinforced Concrete (RC) frames. It is based on combing member- and structure-level techniques in order to achieve optimal design objectives within a multi-level Performance-Based approach. On the one hand, in principle, confinement with composite materials, steel and/or concrete jacketing might be considered as a member-level technique capable to enhance the capacity of under-designed members and, consequently, of the structure as a whole. On the other hand, introducing steel bracing systems or shear walls might be taken into account as structure-level techniques. Generally, member- and structure-level techniques are not employed together in seismic retrofitting or, in the cases in which they are combined, no well-established rules are available for choosing their optimal combination. However, a synergistic use of such techniques by means of well-defined procedures could help designers obtain optimal seismic retrofitting performance. The latest progresses about a procedure developed by the authors for selecting the optimal retrofitting solution among the technically feasible ones, obtained by combining alternative configurations of steel bracing systems and FRP-confinement of critical members, are presented herein. Specifically, the main aspects about formulating a genetic algorithm capable to select the “fittest” retrofitting solution is implemented and summarised. The main assumptions about the representations of “individuals” as part of this genetic algorithm and the main information about the generic operations (i.e. selection, crossover and mutation) are outlined. Finally, the procedure is applied to a 3D frame with the aim to demonstrate its potential.