Genetic Algorithm for the optimal placement of Self-Centering Damage-Free joints in steel MRFs

• Published in: Journal of constructional steel research 2022-10, Vol.197, p.107489, Article 107489
• Main Authors: Pieroni, Ludovica, Di Benedetto, Sabatino, Freddi, Fabio, Latour, Massimo
• Format: Article
• Language: English
• Subjects: Genetic algorithm; Residual drifts; Seismic resilience; Self-centering damage-free joints; Steel moment resisting frames; Structural optimization
• ISSN: 0143-974X; 1873-5983
• DOI: 10.1016/j.jcsr.2022.107489

Nowadays' earthquake engineering is coping with the challenging task of providing low-cost seismic resilient structures. Among others, a viable solution for seismic resilient Steel Moment Resisting Frames (MRFs) is based on the use of Self-Centering Damage-Free (SCDF) joints at Column Bases (CBs) and Beam-to-Column Joints (BCJs), ensuring both the energy dissipation capacity and self-centering behavior of the structure. Past studies demonstrated the beneficial effects gained in damage and residual drifts reduction by including SCDF joints at all BCJs and CBs. However, this solution leads to the highest structural complexity, limiting the practical application. Significant improvements can be obtained including a limited number of SCDF BCJs, but there is a lack of generalized recommendations on the number required and their effective placement. In this work, a Genetic Algorithm (GA) is proposed to define the optimal placement of SCDF BCJs in steel MRFs. The GA is implemented in Matlab, and non-linear time-history analyses are performed in OpenSees to calculate the Fitness-Function. The results of the GA are validated against a Brute-Force Approach. An 8-story 3-bays steel MRF and a type of SCDF joint are selected for case study purposes, non-linear Finite Element Models are developed in OpenSees, and the GA is applied. The results show that the proposed GA is an efficient methodology to solve the considered optimization problem. •A Genetic Algorithm is developed to define the optimal placement of a limited number of Self-Centeing Damage-Free Beam-to-Column Joints.•A sensitivity analysis is performed to evaluate the effect of the input parameters on the final solution.•Validation of the Genetic Algorithm is developed comparing the proposed methodology with a Brute-Force approach.•The Genetic Algorithm is applied to a case study steel MRF with a different number of Self-Centering Damage-Free Joints.