Research on optimal solutions and algorithm stability analyses in RC continuous beam problems


ÖZTÜRK H. T.

Structures, cilt.62, 2024 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 62
  • Basım Tarihi: 2024
  • Doi Numarası: 10.1016/j.istruc.2024.106239
  • Dergi Adı: Structures
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Anahtar Kelimeler: Continuous beam problems, Feasible solution, Metaheuristic search, Optimization, Reinforced concrete, Stability analysis
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

The fundamental objective of engineering is to design a structural system at minimum cost. Moreover, it is crucial that the algorithms used to achieve this objective be stable and fast. In a comparative study of optimization algorithms, the simulation environments should be standardized, sufficient competitors should be used, and algorithms should be examined in terms of stability and computational complexity. The goal of this research is to achieve the optimum design of reinforced concrete continuous beams at the lowest cost, which has been the topic of numerous studies, and to thoroughly evaluate the performance of selected metaheuristic algorithms for this problem. To achieve this goal, a benchmark suite of problems with various numbers of spans was created, and a two-phase research process was followed. The simulation environment was created in a fair and standardized manner. In the first phase, the ability of the twenty-five metaheuristic algorithms to determine optimal designs was examined, feasible solutions were identified, and the top five competitive algorithms were determined. In the second phase, the stability and computational complexity of competitive algorithms were evaluated. This is the first extensive investigation of continuous beam problems (CBPs). The statistical analysis revealed that COA, SOS, SFS, GSK, and TS were the five competitive algorithms. According to the stability analysis results, COA was found to be the most stable algorithm for one- and two-span CBPs, whereas SFS was the most stable algorithm for three-span CBP. The SOS algorithm achieved a feasible solution for all problems within the shortest time based on the computation duration. According to the success-duration distance (SDD) metric introduced for the first time in this study, the COA algorithm is recommended for use in the optimal design of CBPs. Click here for the simulation environment containing the problem tools of the CBP benchmark suite: https://www.mathworks.com/matlabcentral/fileexchange/160641-rc-continuous-beam-problems-cbps-benchmark-suite