چكيده به لاتين
Structures can be optimized in a variety of ways. In designing structures, choosing the type of materials and their composition, choosing the geometry as well as the appropriate geometric characteristics of the sections are important. In the present dissertation, an attempt has been made to achieve an economically and functionally efficient structural design using the latest construction technologies and advanced optimization methods.
In recent years, the use of composite sections in order to make the most of the desired properties of materials has increased significantly. Concrete-steel composite columns, composite roofs as well as restrained buckling braces are the most common of these sections. In such structures, the effort is to achieve maximum performance of the structure. Fire resistance, corrosion, increased radius of gyration, improved stability, increased ductility and improved hysteretic cycles and many other cases have led researchers, designers and manufacturers to use these sections.
In order to achieve the optimal design, the use of metaheuristc methods has become very popular. The high speed of these algorithms, the lack of locking of the optimization process in the local optimization, the lack of need for gradient specifications and many other advantages have made these algorithms a suitable alternative to classical mathematical optimization methods. In many studies that have been done on the optimal design of structures, steel frame structures have been studied and few studies have been done on the optimization of real scale structures, especially with composite systems. In this dissertation, an attempt has been made to study the latest techniques and algorithms of extraction exploration to achieve the optimal design of composite structures with real scale.
In studying the application of optimization in composite structures, various types of spectral analysis and time history analysis have been used. In this study two methods for selecting appropriate records for performing dynamic analysis are proposed. In the first method, using multi-objective optimization algorithms, suitable records are selected by considering the code regulations and minimizing at the same time the mean distance and dispersion of the selected record set with the target spectrum. In the second method, a framework for selecting the appropriate records using optimization for performing incremental dynamic analysis and estimation of fragility curves is presented. In solving the optimal design examples performed in this study, the proposed record selection methods for conducting dynamic analysis have been used.
In this dissertation, various examples of real large-scale structures have been studied using optimization algorithms. The first problem is the optimal design based on the performance of two samples of two-dimensional frames of eight and twenty stories with composite columns, which has been used to select the best design from the concept of damage margin ratio. In the second and third examples, the optimal design of 10 regular stories and 12 irregular stories buildings has been studied. In the optimization process, linear time history analysis has been used and finally, in the last two examples, the optimal design of two ten- and twenty-story structures with buckling restrained braces has been performed.
