Optimum design of steel tower structures under the effect of stress concentration with particle swarm optimization and firefly algorithm

The demand for steel structures is increasing day by day. Steel structures are more expensive than reinforced concrete structures. For this reason, it is necessary to produce steel structures with less cost. In recent years, optimization techniques based on artificial intelligence have been increasing day by day. Artificial intelligence, which is also used in the solution of engineering problems, provides optimum results by reducing the cost. Material continuity may be impaired due to fatigue or material defects (cavities in the parts, sudden cross-section changes on the element, notches or holes) that may occur in the tension elements after manufacturing. The stresses deviate from their usual value and reach their maximum in locations where the material's continuity is disrupted. This situation is called stress concentration. The stress state can be complex in elements where stress concentration occurs. It may not be possible to solve this complexity with only simple bending or compression, simple torsion and simple bending methods. In this case, the stress concentration factor K is included in the calculations. The aim of this study is to test the effectiveness of two heuristic algorithms developed in different years on the optimization process. For this, particle swarm optimization (PSO) developed in 1995 and firefly algorithm (FFA) developed in 2009 were used among the heuristic optimization algorithms. For the optimization process, 25-element, 72-element and 582-element space truss tower structures and 52-element plane truss tower structures optimized with different algorithm techniques in the literature were used. The results of the optimization process were compared with the studies in which these samples were optimized with different algorithm techniques. Then, considering that the element with the highest stress capacity has a notch, the stress value of this element is multiplied by the stress concentration factor K and the optimization process is done again, and the results are presented in tables. Keywords: Optimization, particle swarm optimization, firefly algorithm, stress concentration, steel tower.