SharePoint
Aide
Faire avancer la sûreté nucléaire

La Recherchev2

Publications

Seismic evaluation of existing nuclear facility using ambiant vibration test to caracterize dynamic behavior of the structure and microtremor measurements to caracterize the soil : a case study.


Fermer

Authentification

Email :

Mot de passe :

Ph. VOLANT and N. ORBOVIC
Tenth international conference on soil dynamics and earthquake engineering : SDEE 2001, Philadelphia, Pennsylvania, USA, October 7 - 10, 2001

Type de document > *Congrès/colloque

Mots clés > séismes, bruit de fond, effet de site, séismes

Unité de recherche > IRSN/DEI/SARG/BERSSIN

Auteurs > VOLANT Philippe

Date de publication > 06/10/2001

Résumé

Ambient vibration test and microtremor measurements are used in a seismic evaluation of an existing nuclear facility built in 1962. The goal of the measurements for the evaluation is to calibrate numerical models of the structure and to check the site conditions.

The nuclear facility lies in the French Atomic Energy Commision center of Cadarache (South East of France). This center is located just few kilometers from a main active fault called the South Eastern Durance Fault. The study of the historical seismic activity on this fault shows that on average one earthquake (with magnitude around 5-5.5) occurs every century. The last one occured in 1913 with a magnitude of 5. For this reason, the seismic hazard analysis is of the highest importance in this area.

The studied building is composed of three different and independent blocks. The structure is a reinforced concrete frame with masonry infills. Total height of the building is about 17 m and its masse is about 6700 tons. The building contains an independent massive concrete cell. There is no gap between the frame and the cell.

The first series of measurements was done with 7 velocimeters placed on the high roof of the building (3 on the first block, 2 on the second and 2 en the third block) and one velocimeter on the ground close to the building. The goal was to find the modal shape at the high roof level and to study the behavior of the joints between independent blocks. The second series of measurements was done with 5 velocimetres placed at five levels of the building, one velocimeter on the cell and one on the ground. The goal was to find the modal shape of the main column in elevation and to study the interaction between the frame and the cell.

We then computed the spectral ratios and evaluated the natural frequency. The results show two distincts modes at 4 Hz and 6,5 Hz for the three blocks. The modal shape of the first mode and of the second mode were then estimated both in plan and in elevation. The results show that the three independent blocks behave as they were connected and that a strong interaction exists between the building and the cell. The modal shapes show also influence of masonry infills in dynamic response of the building.

During the experiment on the building, a series of measurements has been made on the soil. A dense microzonation (61 points measurements) using the H/V method (Nakamura method) has been done all around the building in order to characterize the soil quality. The results of the H/V campaign point to intermediate frequency amplification effects (1.5-2 Hz). The building is built on rock (limestones). However, the results of the H/V measurements showed a possible site effect. Moreover, the frequency of the structure without masonry infills (1.4-2.1 Hz) and the frequency of the soil are close (1.5-2 Hz).

A last campaign has been done to give a definitive answer about the existence or not of a site effect. Four stations has been installed (3 around the building and one on a rock site, 400 meters south of the studied building) with permanent recording. The goals of this last experiment was (1) to study the stability of H/V ratios during night and day and (2) to record earthquakes in order to compute spectral ratios between the three stations around the building and the reference station on rock.

The definitive results show a strong variability of H/V ratios during time But the record of earthquakes allowed to compute spectra ratios. The results show that there is no amplification of seismic motion in the vicinity of the building. The paradoxal results of H/V ratios are difficult to explain, probably due to local unidentified noise source effects.

All the measurements on the building will be done again, after the seismic retrofitting planned by the owner. Il will be possible in this way to check the effect of the retroffiting.