The art of truss structural optimization lies in finding the most efficient arrangement and dimensions of the truss that provide the necessary strength, stiffness, and stability while minimizing mass and cost. Material properties, geometric constraints, and load effects all play a crucial role in this process. An optimal design can lead to significant savings in material and construction costs. Recently, simultaneous optimization has attracted considerable attention from researchers, who have proposed new optimization algorithms to tackle truss structure optimization. In the thesis, we developed an optimization algorithm to reduce the cost of a truss structure. The base structure is a truss composed of bars made from various materials, cross-sectional dimensions, and shapes. In the first part of the algorithm, the structure consists of only the cheapest bars capable of carrying the applied loads. In the second part, the algorithm repeatedly selects a random node and moves it in the direction of reduced cost. After each move, the cheapest set of bars that can carry the applied loads is again selected. The cost of the structure with the moved node is then calculated. If it is lower than the cost of the previous structure, the new configuration is accepted. The algorithm was developed in Python. Its functionality was demonstrated in two case studies of bridge trusses.
|
