Abstract
This paper presents an optimization approach for the preliminary seismic design of self-centering moment resisting frames (SC-MRFs). Two problems are tackled herein: The design of SC-MRFs with energy dissipation (ED) bars at the beam-column connections and; The design of SC-MRFs with fluid viscous dampers (FVDs). As the design of new SC-MRFs is considered, all the parameters are set as design variables, including the element's cross-section properties, pre-stress forces and cross-section area of the cables, as well as the ED bars or FVDs parameters. The goal of the optimization is to minimize the total cost of the structural systems while the desired seismic performance level of the frame is obtained by constraining the peak inter-story drifts under a suit of ground motions. In addition, other constraints are adopted to avoid significant plastic deformations to the main frame elements under the Design Basis Earthquake intensity level and to ensure that a self-centering behavior is achieved. Due to the large number of design variables to be optimized, an efficient gradient-based optimization approach is utilized together with the discretized-then-differentiate adjoint sensitivity analysis for the gradient derivation. Moreover, to achieve a practical design in terms of the elements' cross-section properties, discrete material optimization functions are used. The structural responses are evaluated using a nonlinear time history analysis approach that holds both computational efficiency and accuracy. Finally, the efficiency of the methodology is shown using three numerical examples, including an irregular seatback frame.
Original language | English |
---|---|
Article number | 192 |
Journal | Structural and Multidisciplinary Optimization |
Volume | 66 |
Issue number | 8 |
DOIs | |
State | Published - Aug 2023 |
Keywords
- Fluid viscous dampers
- Seismic design
- Self-centering
- Structural optimization
All Science Journal Classification (ASJC) codes
- Software
- Control and Systems Engineering
- Computer Science Applications
- Computer Graphics and Computer-Aided Design
- Control and Optimization