Premium
Damage evaluation models of reinforced concrete buildings based on the damage statistics and simulated strong motions during the 1995 Hyogo‐ken Nanbu earthquake
Author(s) -
Nagato Kenichiro,
Kawase Hiroshi
Publication year - 2004
Publication title -
earthquake engineering and structural dynamics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.218
H-Index - 127
eISSN - 1096-9845
pISSN - 0098-8847
DOI - 10.1002/eqe.376
Subject(s) - structural engineering , reinforced concrete , earthquake engineering , a priori and a posteriori , geology , statistics , geotechnical engineering , computer science , mathematics , engineering , philosophy , epistemology
We have tried to estimate the yield shear strengths of reinforced concrete (RC) buildings based on the damage statistics in Kobe surveyed after the Hyogo‐ken Nanbu, Japan, earthquake of 1995 and the non‐linear response analyses for synthetic waveforms calculated from a complex seismic source and a three‐dimensional basin structure. First, a set of building models that represented the RC building stock in Kobe was constructed and plausible non‐linear multi‐degree‐of‐freedom models with four different numbers of stories were created based on the current seismic code and construction practice. For response analysis the damage criterion and the strength distribution should be assumed a priori . When the damage ratios for these standard models were calculated it was found that the damage ratios were so high that we had to increase the average yield strengths in order to match the calculated damage ratios to those observed. After searching the best models it was found that the estimated average yield strengths should be much higher than those based on the code, especially for low‐rise buildings. Using this set of building models we succeeded in reproducing the belt‐shaped area with high damage ratios in Kobe. One can apply the proposed methodology to different countries if there is enough damage data, strong motion records, and building statistics. If there is sparse damage data at several locations only, then our models can be adjusted to reproduce observed damage data and used for damage prediction as a first‐order approximation. Copyright © 2004 John Wiley & Sons, Ltd.