Showing 27 results for Optimization.
P. Hosseini, A. Kaveh, S. R. Hoseini Vaez,
Volume 12, Issue 4 (8-2022)
Abstract
The existence of uncertainties in engineering problems makes it essential to consider these effects at all times. Robust design optimization allows a design to be made less sensitive to uncertain input parameters. Actually, robust design optimization reduces the sensitivity of the objective function and the variations in design performance when uncertainty exists. In this study, two space trusses were optimized based on the modulus of elasticity, yield stress, and cross-sectional uncertainties in order to increase the response robustness and decrease the weight. The displacement of one node has been used as the criterion for Robust Design Optimization (RDO) of these two structures. Two trusses with 72 members and 582 members are considered, which are famous trusses in the field of structural optimization. Also, the EVPS meta-heuristic algorithm was employed which is an enhanced version of the VPS algorithm based on the single degrees of freedom of a system with viscous damping.
V. Nzarpour, S. Gholizadeh,
Volume 13, Issue 1 (1-2023)
Abstract
Design optimization of cable-stayed bridges is a challenging optimization problem because a large number of variables is usually involved in the optimization process. For these structures the design variables are cross-sectional areas of the cables. In this study, an efficient metaheuristic algorithm namely, momentum search algorithm (MSA) is used to optimize the design of cable-stayed bridges. The MSA is inspired by the Physics and its superiority over many metaheuristics has been demonstrated in tackling several standard benchmark test functions. In the current work, the performance of MSA is compared with that of two other metaheuristics and it is shown that the MSA is an efficient algorithm to tackle the optimization problem of cable-stayed bridges.
M. Ghorbanzadeh, P. Homami, M. Shahrouzi,
Volume 13, Issue 1 (1-2023)
Abstract
The real-world applications addressing the nonlinear functions of multiple variables could be implicitly assessed through structural reliability analysis. This study establishes an efficient algorithm for resolving highly nonlinear structural reliability problems. To this end, first a numerical nonlinear optimization algorithm with a new simple filter is defined to locate and estimate the most probable point in the standard normal space and the subsequent reliability index with a fast convergence rate. The problem is solved by using a modified trust-region sequential quadratic programming approach that evaluates step direction and tunes step size through a linearized procedure. Then, the probability expectation method is implemented to eliminate the linearization error. The new applications of the proposed method could overcome high nonlinearity of the limit state function and improve the accuracy of the final result, in good agreement with the Monte Carlo sampling results. The proposed algorithm robustness is comparatively shown in various numerical benchmark examples via well-established classes of the first-order reliability methods. The results demonstrate the successive performance of the proposed method in capturing an accurate reliability index with higher convergence rate and competitive effectiveness compared with the other first-order methods.
M. Ramezani, M. R. Mohammadizadeh, S. Shojaee,
Volume 13, Issue 2 (4-2023)
Abstract
In recent years, there has been a lot of interest in the development and deployment of control methods that use different components of the building to mitigate the seismic response of the structure. Meanwhile, the building facade, as a non-structural component, can be a suitable alternative in affecting the structure's behavior because of its role as an envelope of the building with a significant weight. Among the modular cladding systems, the Double Skin Facade (DSF) can be considered a passive system due to the distance of the exterior layer from the main structure and sufficient continuity and rigidity. In this study, DSF systems are used as Peripheral Mass Dampers (PMDs) that control structural movements by dissipating energy during strong motions. The PMD system provides a building with several inherent dampers without the need for extra mass. To show the reliability and efficiency of the proposed approach, the PMD model is investigated and compared with results available in uncontrolled and Tuned Mass Damper (TMD) models. The PMD model is examined in three structural frames with 10, 20, and 30 stories with the extreme Mass Ratios (MRs) of 5% to 20%. The Particle Swarm Optimization (PSO) is performed on damper parameters of PMD and TMD systems to minimize structural responses. The results demonstrate that an optimal PMD system with multiple inherent mass dampers outperforms a single TMD system.
F. Damghani , S. M. Tavakkoli,
Volume 13, Issue 2 (4-2023)
Abstract
An efficient method is proposed by using time domain responses and topology optimization to identify the location and severity of damages in two-dimensional structures under plane stress assumption. Damage is assumed in the form of material density reduction in the finite element model of the structure. The time domain responses utilized here, are the nodal accelerations measured at certain points of the structure. The responses are obtained by the Newmark method and contaminated with uniformly random noise in order to simulate real conditions. Damage indicators are extracted from the time domain responses by using Singular Value Decomposition (SVD). The problem of damage detection is presented as a topology optimization problem and the Solid Isotropic Material with Penalization (SIMP) method is used for appropriate damage modeling. The objective function is formed based on the difference of singular values of the Hankel matrix for responses of real structure and the analytical model. In order to evaluate the correctness of the proposed method, some numerical examples are examined. The results indicate efficiency of the proposed method in structural damage detection and its parameters such as resampling length in SVD, penalty factor in the SIMP method and number and location of sensors are effective parameters for improving the results.
P. Hosseini, A. Kaveh, A. Naghian,
Volume 13, Issue 3 (7-2023)
Abstract
Cement, water, fine aggregates, and coarse aggregates are combined to produce concrete, which is the most common substance after water and has a distinctly compressive strength, the most important quality indicator. Hardened concrete's compressive strength is one of its most important properties. The compressive strength of concrete allows us to determine a wide range of concrete properties based on this characteristic, including tensile strength, shear strength, specific weight, durability, erosion resistance, sulfate resistance, and others. Increasing concrete's compressive strength solely by modifying aggregate characteristics and without affecting water and cement content is a challenge in the direction of concrete production. Artificial neural networks (ANNs) can be used to reduce laboratory work and predict concrete's compressive strength. Metaheuristic algorithms can be applied to ANN in an efficient and targeted manner, since they are intelligent systems capable of solving a wide range of problems. This study proposes new samples using the Taguchi method and tests them in the laboratory. Following the training of an ANN with the obtained results, the highest compressive strength is calculated using the EVPS and SA-EVPS algorithms.
A. H. Karimi, A. Bazrafshan Moghaddam,
Volume 14, Issue 1 (1-2024)
Abstract
Most industrial-practical projects deal with nonlinearity phenomena. Therefore, it is vital to implement a nonlinear method to analyze their behavior. The Finite Element Method (FEM) is one of the most powerful and popular numerical methods for either linear or nonlinear analysis. Although this method is absolutely robust, it suffers from some drawbacks. One of them is convergency issues, especially in large deformation problems. Prevalent iterative methods such as the Newton-Raphson algorithm and its various modified versions cannot converge in certain problems including some cases such as snap-back or through-back. There are some appropriate methods to overcome this issue such as the arc-length method. However, these methods are difficult to implement. In this paper, a computational framework is presented based on meta-heuristic algorithms to improve nonlinear finite element analysis, especially in large deformation problems. The proposed method is verified via different benchmark problems solved by commercial software. Finally, the robustness of the proposed algorithm is discussed compared to the classic methods.
