In this paper a new hybrid glowworm swarm algorithm (SAGSO) for solving structural optimization problems is presented. The structure proposed to be optimized here is a simply-supported concrete I-beam defined by 20 variables. Eight different concrete mixtures are studied, varying the compressive strength grade and compacting system. The solutions are evaluated following the Spanish Code for structural concrete. The algorithm is applied to two objective functions, namely the embedded CO2 emissions and the economic cost of the structure. The ability of glowworm swarm optimization (GSO) to search in the entire solution space is combined with the local search by Simulated Annealing (SA) to obtain better results than using the GSO and SA independently. Finally, the hybrid algorithm can solve structural optimization problems applied to discrete variables. The study showed that large sections with a highly exposed surface area and the use of conventional vibrated concrete (CVC) with the lower strength grade minimize the CO2 emissions.
Hybrid glowworm swarm algorithm; discrete variables; concrete I-beam; CO2 emissions; self-compacting concrete