Structural Damage Assessment Through Both Centralized and Sensor-level Vibration-based Data Analysis

Abstract

The concept of vibration-based damage detection (VBDD) provided a platform for structural system identification, damage detection and localization, and quantification. In this study the VBDD approach, through both centralized and sensorlevel data analyses, will be discussed for structural damage assessment. Damage assessment is focus on damage detection and localization using both ambient and the earthquake response measurements. Finally, for damage quantification a model updating technique called efficient model correction method (EMCM) is applied in this study to correct/update the analytical mass and stiffness matrix respectively using the identified frequencies and mode shapes from recursive subspace identification (RSI) algorithm. Combined the result from RSI and damage detection algorithms, the percentage of stiffness reduction and the time of damage occurrence can be identified. To demonstrate the capability of the proposed recursive subspace identification algorithm and the damage assessment technique, dynamic response data generated from the shaking table test of a twin tower steel structure is used to verify the algorithms on damage detection and quantification.