Browsing by Author "Üstüner B."

Now showing 1 - 3 of 3
  • No Thumbnail Available
    Item
    Finite element-based analysis of optimally designed steel plane frames
    (Institute for Ionics, 2023) Üstüner B.; Doğan E.
    In this paper, a three-story two-span steel frame and a five-story irregular steel frame models are examined. Hunting Search algorithm (HuS), particle swarm algorithm (PSO), and Aquila algorithm (AO) are used to determine suitable frame sections. The performances of the algorithms are investigated with the welded beam design problem and the three-truss problems. The work consists of two stages as optimization and solution with finite element method. The aim of this study is to determine the frame section within the constraints to have a minimum weight. The models were created using the sections determined because of the optimizations are performed with finite element analysis. In steel frame structures, the provisions of the AISC-LRFD practice code are used. The analysis is done with the ABAQUS CAE program, which is one of the finite element methods. Frames with optimum sections are drawn and analyzed in the program. As a result of the analysis, the displacement and stress values are shown. These all values are within limits. PSO and HuS algorithms give the same results. The worst result is taken from the AO algorithm. © 2023, The Author(s), under exclusive licence to Springer Nature Switzerland AG.
  • No Thumbnail Available
    Item
    Structure Optimization with Metaheuristic Algorithms and Analysis by Finite Element Method
    (Korean Society of Civil Engineers, 2024) Üstüner B.; Doğan E.
    In engineering, design is made by considering functionality, reliability, manufacturability, usability, and total cost. There are a wide variety of methods for design optimization. Metaheuristic methods inspired by nature are one of them. In this study, the Refinement firefly algorithm is proposed as a new method. Grey Wolf, Particle Swarm, and Firefly algorithms are compared with the proposed Refinement Firefly Algorithm. Mathematical benchmark problems are used to examine the performance of algorithms. Also, welded beam, cellular beam, and frame system designs are considered sample problems. These design examples are solved by algorithms and the sections are determined. The sections determined by optimization were analyzed using the ABAQUS CAE program and its reliability was examined. Numerical analysis with the finite element method is very useful as it provides realistic solutions. ABAQUS CAE is used to detect and show deformations in the structure. Finite element solution with ABAQUS solves the problems analytically and it is seen that the sections determined by the optimum design algorithm remain within the limits. The proposed Refinement Firefly algorithm demonstrates superior performance compared to the Firefly algorithm. However, it yields inferior results when compared to the Grey Wolf and Particle Swarm algorithms. © 2024, Korean Society of Civil Engineers.
  • No Thumbnail Available
    Item
    Enhancing Structural Evaluation: Machine Learning Approaches for Inadequate Reinforced Concrete Frames
    (Springer Science and Business Media Deutschland GmbH, 2024) Altıok T.Y.; Üstüner B.; Özyüksel Çiftçioğlu A.; Demir A.
    This paper provides comprehensive analyses of the performance of inadequate reinforced concrete frames from different aspects. First, a one-story, single-span, 1/3 scaled frame is constructed. The ultimate lateral load-bearing capacity of the sample is experimentally determined. Second, the behavior of the sample is obtained by finite element analysis. In the analyses, axial load, rebar diameter, and concrete strength values are taken as variables. Load–displacement behaviors determined from the tests and finite element analyses are compared. Third, machine learning models are developed to estimate the ultimate load-bearing capacity of the frames. Random Forest, ElasticNet, RANSAC, Decision Tree, K-Nearest Neighbors, and Gaussian Naive Bayes are employed to assess the load-bearing capacities of the frames. Coefficient of determination, Root Mean Squared Error, Mean Absolute Error, and Mean Absolute Percentage Error, which are widely recognized performance indicators, are also employed to assess the effectiveness of machine learning methods. The findings reveal that the Random Forest method is the most precise and effective in both regression and classification analyses. It has the highest Coefficient of determination of 87% in predicting the load-bearing capacity of the frame and the highest accuracy of 100% in classifying the frames based on their load-bearing capacity. This is the first attempt to employ machine learning approaches to assess the load-bearing capacity of inadequate reinforced concrete frames. The proposed model can provide a better understanding of inadequate frames and has advantages over other analysis methods with respect to simplicity in application, flexibility, reducing the time and cost associated with the process. © The Author(s), under exclusive licence to Shiraz University 2024.