Optimization of triple-friction pendulum isolator using Kriging as a surrogate model under El Centro earthquake

Articles in Press

Document Type : Original Article

Authors

1 Ph.D. student, Department of Civil Engineering, Yazd University, Yazd, Iran.

2 Associate Professor, Department of Civil Engineering, Yazd University, Yazd, Iran

3 Assistant Professor, Department of Civil Engineering, Yazd University, Yazd, Iran

10.22065/jsce.2024.421572.3246

Abstract

The triple friction pendulum isolator is adaptable and can exhibit variable stiffness and damping in response to different earthquakes. Therefore, the need to optimize the effective parameters of this isolator to achieve the desired performance is significant. In this research, the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) algorithm is used to optimize the effective parameters of the triple friction pendulum isolator to minimize the maximum relative displacement between floors and the roof acceleration in two separate optimization processes. As optimizing complex models is computationally intensive and time-consuming, substituting the actual responses of the finite element model with cost-effective surrogate models results in significant time savings. In this article, structural analysis is performed using a nonlinear model implemented in OpenSees software, making the structural optimization process time-consuming and computationally expensive. To address this challenge, the Kriging model is employed as a surrogate to replace the finite element analysis computations necessary to calculate the responses, i.e., minimizing the maximum relative displacement between floors and the roof acceleration. The optimization results using the combination of the CMA-ES algorithm and the surrogate model are compared with optimization using the genetic algorithm. The findings reveal that utilizing the surrogate model for optimization reduces relative displacement between floors by 66.39% and roof acceleration by 12.28%. Additionally, computational costs are reduced by over 84%.

Keywords

Main Subjects

Journal of Structural and Construction Engineering

Articles in Press, Accepted Manuscript
Available Online from 08 April 2024

Files

History

  • Receive Date: 31 October 2023
  • Revise Date: 29 February 2024
  • Accept Date: 08 April 2024

Share

How to cite

Statistics