امين ميرزايي

The main objective of structural engineering is to ensure the safety of structures while also considering economic factors. Reliability-based design optimization involves striking a balance between the cost and safety of structures, in order to achieve maximum safety with minimum cost. This thesis utilizes multi-objective meta-heuristic algorithms in the first step to solve these types of problems. In the second step, the reliability of the designs obtained in the first step is eva‎luated, taking into account the presence of uncertainties. Using this method, a range of optimal solutions with varying levels of reliability can be obtained in a single run. By utilizing this approach, designers can select an optimal solution from this set based on the specific conditions and objectives of each problem. The first step of this thesis involves achieving optimal designs of structures using multi-objective algorithms such as the Gray Wolf, Dragonfly, Grasshopper, and Ant Lion algorithms. In the second step, the reliability of structures is eva‎luated using the Latin hypercube sampling method. To demonstrate the effectiveness and practicality of this proposed method, the research carried out three structural examples, including a spatial 582-bar tower truss, a 520-bar double-layer grid, and a 3-bay 15-story frame.

طراحي بهينه , بهينه‌سازي چند هدفه , الگوريتم‌هاي فراابتكاري چند هدفه , قابليت اعتماد سازه , نمونه‌برداري ابر مكعب لاتين

Optimal design , Multi-objective optimization , Multi-objective meta-heuristic algorithms , Reliability of the structure , Latin hypercube sampling

Amin Mirzayi

Dr. Majid Ilchi Ghazaan