Keywords: collapse mechanism, nonlinear dynamic analysis, plastic hinge, earthquake damage, Zemmouri earthquake May 21, 2003.

On May 21, 2003, a major shallow earthquake of magnitude 6.8 (Mw) shook the eastern part of the capital city of Algiers (Algeria). The epicentre was located at 36.89N-3.78E just offshore from the locality of Zemmouri in the province of Boumerdes about 50 km from Algiers. The intensity of shaking was very strong in the epicentral region and attained remote areas. The maximum acceleration amplitude of ground motion was measured as 0.34 g for EW direction, 0.26g for NS direction and 0.25g for vertical direction at the Keddara1 station about 20 km from the epicentre (National Earthquake Engineering Research Centre, Algiers). Post-earthquake surveys documented and categorised well the damages [1]. However, the principal problem with engineering involvement in the damage assessment process is a lack of consistent structural damage assessments. The earthquake damage information contained in this paper is based on a reconnaissance survey performed two weeks after the earthquake in the Boudouaou region (15 km from the epicentre). The survey put a special emphasis on the ruin mechanism of particular reinforced concrete frame buildings that suffer damage ranging from minor cracks to total collapse. The damaged buildings were examined and the mode of local failure of the columns and beams were identified. The data of field investigations helped to determine the parameters that potentially influence the performance of the structures under seismic excitations and were introduced in the numerical models to carry out nonlinear dynamic analyses [2]. The effects of uncertainties in material properties and simulated structural faults on the overall response of the buildings were examined.

The results showed that structure models having well detailed ductile sections (good concrete quality) would withstand the earthquake ground motion. However, the response of a model with actual poor concrete properties was instable, characterised by severe damage at lower storeys and considerable loss of overall stiffness. The damage predicted analytically portrayed in general the observed damage.

The choice of the hysteresis parameters: stiffness degradation, strength deterioration and slip-bond (pinching) is important. These parameters may vary within a range for a given material properties disparity. It was found that small variation in the slip-bond parameter to account for the observed slip may affect remarkably the overall behaviour of the structure. This is demonstrated using a model with 10% change in the pinching parameter which provoked the model to collapse after 7.6 seconds into the earthquake. These results explain the disparity of the observed damage sustained by several buildings.

Therefore, it is evident that inappropriate construction methods resulting in low strength capacity of some strategic element may cause dramatic failure to the entire system. The approach is potentially useful in understanding and evaluating the unexpected structural damage, that which was not foreseen in the design process.

1

M. Belazougui, M.N. Farsi, A. Remas, "Zemmouri earthquake of May 21, 2003 : Building damage assessment and causes", Proceedings of the International Conference on risk, vulnerability and reliability in construction Vol. 2, pp. 653-660, 2003.

2

R.E. Valles, A.M. Reinhorn, S.K. Kunnath, C. Li, A. Madan, "IDARC 2D, program for the inelastic damage analysis of buildings", Technical Report NCEER-96-0010, 1996.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £135 +P&P)