Ps. Skjaeraek et al., IDENTIFICATION OF DAMAGE IN REINFORCED-CONCRETE STRUCTURES FROM EARTHQUAKE RECORDS - OPTIMAL LOCATION OF SENSORS, Soil dynamics and earthquake engineering, 15(6), 1996, pp. 347-358
A method for the localization of structural damage in seismically exci
ted reinforced concrete (RC) structures using a measured acceleration
response time series is presented. From the measured response of some
or all storeys, the two lowest smoothed eigenfrequencies and mode shap
e coordinates are estimated. These estimated values are used as an inp
ut to a developed substructure iteration method where local storey dam
ages are estimated in such a way that these smoothed values are reprod
uced. The local damage indicator of a substructure is defined as the a
verage reduction of the stiffness matrix of the initial undamaged subs
tructure. The method is applied to simulated data of a six-storey, two
-bay test frame (scale 1:5) that is to be tested at the Structural Lab
oratory of Aalborg University, Denmark. The simulations are performed
using the non-linear finite element program SARCOF. Special emphasis i
s put on the investigation of the optimal location of measurement sens
ors, i.e. ac which locations along the structure is the most informati
on about the damage distribution gained. In all cases it is assumed th
at measurements are performed at top storey and ground surface, and th
e investigations are concentrated on putting one or two more measureme
nt points in between. The two cases where the structure is excited in
the first and second mode are investigated, and it is found that in ge
neral the sensors should be placed in the lower part of the structure.
Furthermore, it is found that the method provides good results even w
hen only the measurements at top storey and ground surface are used. C
opyright (C) 1996 Elsevier Science Limited.