Se Woon Choi

GA-Based Multi-Objective Optimizationźfor Retrofit Design on a Multi-Core PC Cluster

This article presents a distributed nondominated sorting genetic algorithm II NSGA-II for optimal seismic retrofit design using buckling restrained braces BRBs on a cluster of multi-core PCs. In the formulation, two conflicting objective functions of the initial BRB installation cost required for seismic retrofitting and damage cost that can be incurred by earthquakes expected during the life cycle of the structure were minimized. Because time-consuming nonlinear structural analyses are required for fitness evaluations of individuals in every generation, parallelism at candidate design level or individual level is exploited by assigning fitness evaluations for individuals to slave core processors evenly. The distributed algorithm is applied to seismic retrofit design of 2D steel frame structure and 3D irregular reinforced concrete structure. The performance of the distributed NSGA-II was assessed based on three criteria: convergence of the distributed algorithm, efficiency of distributed computing, and quality of optimal solutions. Implementation of the distributed algorithm on the multi-core cluster consisting of up to 64 core processors resulted in relatively high speedups or efficiencies of the distributed optimization without deteriorating the quality of the optimal solutions.

Deformation Monitoring of a Building Structure Using a Motion Capture System

Conventional 1-D or 2-D displacement sensors are occasionally used to measure the deformation of a structure. However, a motion capture system (MCS) can measure the 3-D movements of markers attached to a target structure with high accuracy and a high sampling rate. Because markers can be easily attached to a structure, an MCS is useful for monitoring the dynamic motions of complex structures, such as buildings, using multiple markers. This study proposes a deformation measurement method for building structures using an MCS. The suggested measurement method consists of four stages: 1) setup of the MCS; 2) data acquisition; 3) coordinate transformation into a structural coordinate system; and (4) generation of the deformed shape. The feasibility of the suggested MCS-based measurement method was validated using a free-vibration test of a three-story experimental frame model. The displacement and deformed shape that were measured using an MCS were compared to the displacement and deformed shape measured using a laser displacement sensor, which is a conventional displacement sensor. The comparison results indicated that the MCS can overcome the limits of 1-D displacement sensors and easily and accurately obtain the deformed shape of a structure.

Sensor-Free Stress Estimation Model for Steel Beam Structures Using a Motion Capture System

As a new paradigm for safety assessment of beam structures, this paper proposes a sensor-free monitoring model to estimate the stress distribution of steel beam structures with uncertain loads. The model consists of three steps: the use of a sensor- and wire-free measurement step using a motion capture system; a deformed shape estimation step employing cubic smoothing spline interpolation; and a stress distribution and maximum stress estimation step using the radius of curvature of the estimated deformed shape or a finite-element method. Therefore, unlike the conventional sensor-based structural health monitoring with lengthy cables, the model can determine the maximum stress in a non-contact manner without the deterioration of the measurement accuracy. The sensor-free stress estimation method was applied to estimate the stress distribution of a steel beam structure subjected to static loads applied at the mid-span of the beam by a hydraulic jack. The validity of the proposed method was confirmed by comparing the estimated stresses with stresses measured in static loading tests.

Wind-induced response control model for high-rise buildings based on resizing method

A variety of methods have been applied to reduce the effect of the wind-induced vibration of a high-rise building as the excessive wind-induced vibration at the top of a high-rise building can cause physical and psychological discomfort to the user or the residents. For structural engineers, the most effective approach to control the wind-induced responses of high-rise buildings would be to control the stiffness or natural frequency of the building. This paper presents a practical design model to control the wind-induced responses of a high-rise building. In the model, the stiffness of a high-rise building is maximized to increase the natural frequency of the building by the resizing algorithm. The proposed design model is applied to control the wind-induced vibration of an actual 37-storey building during the initial stage of its structural design.

Analytical models for estimation of the maximum strain of beam structures based on optical fiber Bragg grating sensors

AbstractThe structural safety of a beam structure is assessed by a comparison between the maximum stress measured during monitoring and the allowable stress of the beam. However, the strain directly measured from a fiber Bragg grat- ing (FBG) strain sensor may not be identical with the actual maximum strain induced in the structural member. Unless a FBG strain sensor is installed exactly on where maximum strain occurs, the reliability of the evaluated safety based on the measured strain depends on the number and location of sensors. Therefore, in this paper, analytical models are presented for estimation of the maximum values of strains in a linear elastic beam using the local strains measured from FBG sensors. The model is tested in an experiment by comparing estimated maximum strain from FBG sensors and directly measured strain from electrical gages. For the assessment of safety of typical beam structures in buildings and infrastructures, analytical models for various loading and boundary conditions are...

A Proposal of the Gage-Free Safety Assessment Technique for the Steel Beam Structure Under Uncertain Loads and Support Conditions Using Motion Capture System

Estimating the maximum stress through stress distribution of a structure is an important indicator for structural safety evaluation. Structural health monitoring can be used to do this with a variety of measuring equipment such as strain gage, LVDT, LDS. All the measuring equipment, however, has some weakness in the configuration of complex wire network and some inconvenience of replacing faulty sensors. Therefore, this paper suggests a technique that can estimate stress distribution of steel beam structure under uncertain load and support conditions by using motion capture system (MCS). MCS is a Vision-based Monitoring System, which measures 3D coordinates of multiple markers attached to the surface of steel beam without installing the complex wire network. In this study, the stress distribution is estimated from an analytic model by using displacement values measured by MCS. For the evaluation of the estimated stress distribution, comparing with the measured stress from ESG is performed.

Genetic Algorithm Based Optimal Seismic Design Method for Inducing the Beam-Hinge Mechanism of Steel Moment Frames

In this paper, the optimal seismic design method for inducing the beam-hinge collapse mechanism of steel moment frames is presented. This uses the non-dominated sorting genetic algorithm II(NSGA-II) as an optimal algorithm. The constraint condition for preventing the occurrence of plastic hinges at columns is used to induce the beam-hinge collapse mechanism. This method uses two objective functions to minimize the structural weight and maximize the dissipated energy. The proposed method is verified by the application to nine story steel moment frame example. The minimum column-to-beam strength ratio to induce the beam-hinge collapse mechanism are investigated based on the simulation results. To identify the influence of panel zone on the minimum column-to-beam strength ratio, three analytic modeling methods(nonlinear centerline model without rigid end offsets, nonlinear centerline model with rigid end offsets, nonlinear model with panel zones) are used.

A Study on the Minimum Column-To-Beam Moment Ratio of Steel Moment Resisting Frame with Various Connection Models

This paper presents the minimum column-to-beam moment ratio of steel moment resisting frame (SMRF) with various connection models using the non-dominated sorting genetic algorithm-II (NSGA-II). We evaluate the minimum column-to-beam moment ratio of SMRF required to ensure the beam-sway mechanism by the pushover analysis. We investigate the influence of the connections on the minimum column-to-beam moment ratio. We consider WUF-W and RBS as the connection models. The optimal algorithm used in this paper is to minimize the structural weight and the column-to-beam moment ratios through the objective functions and to induce the beam-hinge mechanism through the constraint which is to prevent the formation of plastic hinge at the column part consisting of joints. Using the 3-story example, we compare the minimum column-to-beam moment ratio of SMRF required to ensure the beam-sway mechanism according to the connection models.