Parameter-less population pyramid for large-scale tower optimization

•Parameter-less population pyramid algorithm is introduced for the constrained and structural optimization.•The algorithm is examined by solving two large-scale tower design optimization problems.•In order to apply the algorithm to these discrete problems, gray-codes where employed.•The performance of the algorithm is further compared with four well-known algorithms.•The results show improvements in both final solutions and convergence rate. The parameter-less population pyramid (P3) is a recent evolutionary computation algorithm proposed for black box optimization. Shown to be efficient for a variety of benchmark problems, P3 replaces the conventional constant population model with expanding sets of expanding populations. We investigated how this new metaheuristic optimization algorithm would transfer to optimize large-scale tower structure problems involving different constraints: geometric and mechanical. P3 is examined by optimizing two discrete tower design problems, 26-story and 35- story tower structures. The performance of P3 is compared with other well-known evolutionary algorithms for black-box optimization including random restart hill climbing, parameter-less hierarchical Bayesian optimization algorithm, differential evolution, and a modified genetic algorithm. The results show that does P3 not only finds the best final solutions, but it also reaches high quality solutions much faster than the other algorithms This fast optimization is vital for the tedious and large-scale structural engineering problems. Finally, the unique search features used in the P3 and the implications for future studies are discussed.