Efficiency and Robustness of Three Metaheuristics in the Framework of Structural Optimization (original) (raw)
Related papers
gazi university journal of science, 2018
Optimum solution of an anticipated problem is generally reached through minimizing or maximizing a governing real function while sometimes should satisfy various predefined limitations. Selecting an algorithm as a main optimizer plays a key role on the solution process. In this respect, current study intends to compare the performances of two different common metaheuristic optimization algorithms as integrated particle swarm optimizer (iPSO) and teaching and learning based optimizer (TLBO). The TLBO is two-phase algorithm while the iPSO is a single-phase algorithm. Their capabilities are compared over some benchmark cases including mathematical functions and structural optimization problems. To increase the complexity of the test problems both size and topology specifications of the structural systems are simultaneously taken as the decision variables. Achieved results demonstrate the superiority of the iPSO in comparison with TLBO in both search capability and convergence rate.
Metaheuristics in structural optimization and discussions on harmony search algorithm
Swarm and Evolutionary Computation, 2016
Metaheuristic algorithms have provided efficient tools to engineering designers by which it became possible to determine the optimum solutions of engineering design optimization problems encountered in every day practice. Generally metaheuristics are based on metaphors that are taken from nature or some other processes. Because of their success of providing solutions to complex engineering design optimization problems the recent literature has flourished with a large number of new metaheuristics based on a variety of metaphors. Despite the fact that most of these techniques have numerically proven themselves as reliable and strong tools for solutions of design optimization problems in many different disciplines, some argue against these methods on account of not having mathematical background and making use of irrelevant and odd metaphors. However, so long as these efforts bring about computationally efficient and robust optimum structural tools for designers what type of metaphors they are based on becomes insignificant. After a brief historical review of structural optimization this article opens this issue up for discussion of the readers and attempts to answer some of the criticisms asserted in some recent publications related with the novelty of metaheuristics.
2021
Truss structures are one of the major civil engineering members studied in the optimization research area. In this area, various optimization applications such as topology, size, cost, weight, material usage, etc., can be conducted for different truss structure types. In this scope with the present study, various optimization processes were carried out concerning two different large-scale space trusses to minimize the structural weight. According to this state, three structural models provided via two different truss structures, including 25 bar and 72 bar truss models, were handled for evaluation of six different metaheuristics together with the modification of Levy flight for three of the algorithms using swarm intelligence by considering both constant and variable populations, and different ranges for iterations, too. Additionally, the effects of the Levy flight function and whether it is successful or not in terms of the target of optimization were also investigated by comparing...
Structural Design Optimization Using Particle Swarm Optimization and Its Variants
IOP Conference Series: Earth and Environmental Science, 2020
Structural design optimization has become an extremely challenging and more complex task for most real-world practical applications. A huge number of design variables and complex constraints have contributed to the complexity and nonlinearity of the problems. Mathematical programming and gradient-based search algorithms cannot be used to solve nonlinear problems. Thus, researchers have extensively conducted many experimental studies to address the growing complexity of these problems. Metaheuristic algorithms, which typically use nature as a source inspiration, have been developed over past decades. As one of the widely used algorithms, particle swarm optimization (PSO) has been studied and expanded to deal with many complex problems. Particle swarm optimization and its variants have great accuracy in finding the best solution while maintaining its fast convergence behavior. This study aims to investigate PSO and its variants to solve a set of complex structural optimization problem...
Critical Evaluation of Metaheuristic Algorithms for Weight Minimization of Truss Structures
Frontiers in Built Environment
This study critically compares variants of Genetic Algorithms (GAs), Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), Differential Evolution (DE), and Simulated Annealing (SA) used in truss sizing optimization problems including displacement and stress constraints. The comparison is based on several benchmark problems of varying complexity measured by the number of design variables and the degree of static indeterminacy. Most of these problems have been studied by numerous researchers using a large variety of methods; this allows for absolute rather than relative comparison. Rigorous statistical analysis based on large sample size, as well as monitoring of the success rate throughout the optimization process, reveal and explain the convergence behavior observed for each method. The results indicate that, for the problem at hand, Differential Evolution is the best algorithm in terms of robustness, performance, and scalability.
Application of Metaheuristic Algorithms in Truss Structure Sizing Optimization
IOP Conference Series: Earth and Environmental Science, 2020
Field studies of structural optimization have gained increased attention due to the rapid development of metaheuristic algorithms. One widely known metaheuristic algorithm, Particle Swarm Optimization (PSO), has been extensively used to solve many problems and is reported to have fast convergence behavior and good accuracy. As many problems become more complex, studies have been focused on improving PSO searching capability. This study presents the application of PSO and its variants in optimizing truss structures. The performances of PSO and several PSO variants, namely, linearly decreasing inertia weight PSO (LDW-PSO) and bare bones PSO (BB-PSO), were compared and investigated. All optimization algorithms were tested in 72-bar and 25-bar spatial truss problems. The results indicate that BBPSO was the best algorithm in terms of optimum solution, consistency, and convergence behavior.
A Comparative Study of Three Metaheuristics for Optimum Design of Engineering Structures
A comparative study is carried out on the optimum design of a real-size steel frame by considering three different metaheuristic search techniques. The techniques are selected as the Firefly Algorithm (FFA), Artificial Bee Colony (ABC), and Cuckoo Search (CS) algorithms. Metaheuristic search techniques of optimization are non-deterministic methods and they rely on heuristics in finding the better solutions in the search space. They use random or probabilistic parameters, while they search for the optimum solution, rather than deterministic quantities. The source of random variables may be several depending on the nature and the type of problem. The heuristics behind these innovative techniques is borrowed from the nature or physics. In the design example considered, the design constraints include the displacement limitations, inter-story drift restrictions, strength requirements for beams and beam-columns which are formulated according to provisions of LRFD-AISC (Load and Resistance Factor Design of American Institute of Steel Institution).
Periodica Polytechnica Civil Engineering, 2014
In this paper a metaheuristic algorithm composed of particle swarm, ray optimization, and harmony search (HRPSO) is presented for optimal design of truss structures. This algorithm is based on the particle swarm ray origin making is used to update the positions of the particles, and for enhancing the exploitation of the algorithm the harmony search is utilized. Numerical results demonstrate the efficiency and robustness of the HRPSO method compared to some standard metaheuristic algorithms.
Evolutionary Parallel and Serial Programming Algorithm for Structural Optimization
This paper aims at performing a structural truss mass optimization taking into account frequency constraints. It is well-known that the vibration modes may easily switch due to geometrical structural modifications. This paper investigates the use of a Particle Swarm Optimization (PSO) algorithm as an optimization engine in this type of problems. It is suggested some new implementations on the basic algorithm based on literature reports, such as the parallelization of the code, in order to improve its performance. The algorithms differ in the way of communication between processors (synchronous or asynchronous) and the form of the particle swarm update (immediate or delayed). The proposed models were applied to very common and explored engineering problems of truss optimization. The performance of the proposed algorithm showed a linear behavior for the processing gain curve (speedup) while maintained the ability of finding good solutions.