A new model order reduction strategy adapted to nonlinear problems in earthquake engineering.

Area of Science:

• Structural Engineering
• Computational Mechanics
• Seismic Analysis

Background:

• Dynamic response analysis of large, nonlinear structures during earthquakes is computationally intensive.
• Improving seismic performance of critical facilities like medical complexes requires efficient analysis methods.
• Base isolation using frictional pendulum bearings is a key strategy for seismic protection.

Purpose of the Study:

• To develop and present a novel model order reduction (MOR) strategy based on proper orthogonal decomposition (POD).
• To demonstrate the practical applicability of this MOR strategy to a realistic building structure.
• To enhance the seismic performance analysis of complex structures, reducing computational expense.

Main Methods:

• The proposed MOR strategy utilizes proper orthogonal decomposition (POD) to create a transformation matrix from 'snapshots' of structural response.
• This transformation matrix generates reduced-order models for analyzing structures under various earthquake excitations.
• Nonlinear low-order representations are solved, drastically cutting computation time compared to full-order models.

Main Results:

• The new MOR strategy significantly reduces the time required for dynamic response analysis of complex structures.
• Reduced-order models accurately approximate the physical responses of the structure, validated against full-order computations.
• The accuracy of the MOR strategy is contingent on capturing the probable structural behavior within the POD snapshots.

Conclusions:

• The developed MOR strategy offers a computationally efficient and accurate approach for seismic performance analysis of large, complex structures.
• This method is particularly effective for base-isolated buildings, improving the speed of seismic response prediction.
• The POD-based MOR technique provides a valuable tool for engineers seeking to optimize structural designs under seismic loads.