TA625 : Optimization of location,topology and number of bracing spans in steel structures using multi mexta-heuristic baxsed search method and dynamic constraints
Thesis > Central Library of Shahrood University > Civil & Architectural Engineering > PhD > 2021
Authors:
[Author], Vahid reza Kalat Jaari[Supervisor], [Advisor]
Abstarct: The issue of optimization and analysis of structures has been discussed in structural engineering societies for many years. And the interest of engineers and designers has always been on the optimal use of materials such as structural materials and maximum safety. This study aims to optimize the weight of steel frxames (as the objective function). The ultimate goal was to select an optimal arrangement and coordinate the Three types of convergent bracing (multi-story X-bracing, Diagonal-bracing and X-bracing) in low, mid, and high-rise simple steel frxames with specific dimensions and spans. In the end, an optimal weight design will be presented for the frxames under the design constraints (by LRFD method), frequency, drift, and position of braces in frxame height, and the amount of the tensile force of braces. In this research, the frxames have been optimized using the connection between two software OPENSEES and MATLAB baxsed on a multi mexta-heuristic optimization algorithm with discrete variables (a new, mixed-method baxsed on parallel island model with four islands). The findings of this research include the optimal position of braces, sections, and convergence history for the frxames under a multi mexta-heuristic optimization algorithm. The diagrams of convergence history were also provided for the ant colony algorithm in mid-rise frxames with multi-story X-bracing. The results show the superiority of the multi mexta-heuristic search algorithm in the convergence speed and the quality of the optimal response compared to the ant colony optimization algorithm. It should be noted that the weight of the frxames at the onset of optimization with the ant colony algorithm is 51% higher than the algorithm of this research and While the average convergence in multi mexta-heuristic search algorithm is 136 iterations, In all mid-rise frxames studied in this research, the ant colony algorithm has converged with the average 86 iterations more than the multi mexta-heuristic search algorithm. Finally, multi mexta-heuristic search algorithm has designed more optimal mid-rise frxames than ant colony algorithm with an average weight difference of 7.5%. Also, according to the results of optimization with multi mexta-heuristic search algorithm, the optimal weight belongs to the first the multi-story X-bracing frxames (the combination of inverted V- Bracing and V- Bracing) Then for Diagonal-bracing and finally is for the frxames with the X-bracing. More precisely, the average optimum weight of the frxames with multi-story X-bracing was 6.8% less than the average weight of the frxames with the X-bracing and the average weight of the frxames with Diagonal-bracing was 4.2% less than the average weight of the frxames with X-bracing. Regarding the optimum location of braces, the results of the proposed algorithm showed that the braces were located in central and adjacent to each other.
Keywords:
#optimization; multi mexta-heuristic algorithm; simple steel frxame; optimal position and topology of convergent braces; frequency constraint; parallel-island processing Keeping place: Central Library of Shahrood University
Visitor: