Dello Russo, Angelo (2015) Seismic response of soil embankments in near-source conditions. [Tesi di dottorato]

[img]
Preview
Text
Dello_Russo_Angelo_27.pdf

Download (16MB) | Preview
[error in script] [error in script]
Item Type: Tesi di dottorato
Resource language: English
Title: Seismic response of soil embankments in near-source conditions
Creators:
CreatorsEmail
Dello Russo, Angeloangelo.dellorusso@unina.it
Date: 31 March 2015
Number of Pages: 297
Institution: Università degli Studi di Napoli Federico II
Department: Scienze della Terra, dell'Ambiente e delle Risorse
Scuola di dottorato: Scienze fisiche
Dottorato: Rischio sismico
Ciclo di dottorato: 27
Coordinatore del Corso di dottorato:
nomeemail
Zollo, Aldoaldo.zollo@unina.it
Tutor:
nomeemail
Sica, StefaniaUNSPECIFIED
Date: 31 March 2015
Number of Pages: 297
Keywords: Near-Source conditions, Seismic Response Embankment, Domain Reduction Method, Nonlinear Dynamic Analysis
Settori scientifico-disciplinari del MIUR: Area 08 - Ingegneria civile e Architettura > ICAR/07 - Geotecnica
Additional information: Dottorato svolto presso il Dipartimento di Ingegneria dell'Università degli Studi del Sannio (BN) quale sede consorziata
Date Deposited: 12 Apr 2015 09:45
Last Modified: 08 Oct 2015 08:07
URI: http://www.fedoa.unina.it/id/eprint/10536
DOI: 10.6092/UNINA/FEDOA/10536

Collection description

The research activity carried out in PhD period focused on numerical modelling of the seismic response of soil embankments in near-source conditions. The interest on such a topic comes from the fact that in Italy and worldwide there are many large earth dams placed very close to active faults. In such conditions the seismic response of the structure could be affected by near-source phenomena. Worth mentioning are the case-histories of Conza Dam during the Irpinia 1980 earthquake and Campotosto reservoir during the 2009 Abruzzo earthquake. In the first case, the embankment was partially jeopardized by the seismic event, being the dam site very close to the source ( ≈ 10 km). In the latter case, the epicenters of several aftershocks following the April 6, 2009 main event migrated in the NW direction, just below the 315-million-m3 Campotosto reservoir. In those days, the authorities in charge of the dam safety acted completely unprepared to face the emergency, due to the lack of pre-arranged predictive/interpretative tools. The investigated research theme is quite new in the geotechnical field. It requires a detailed knowledge of basic seismological aspects to properly simulate the seismic source and wave propagation pattern to get the input motion exciting the site and the embankment. As regards the implications in the field of civil engineering, an important feature of near-source phenomena is ground-motion asynchronism. This means that two points placed not very far each other, i.e. at a distance comparable to the dimensions of strategic infrastructures - as dams, road embankments, or bridges - may undergo quite different motion at their base (even not accounting for site effects). In such a context, the proper characterization of the motion at the bedrock level may provide a better prediction of the structure response to seismic loadings. This issue is crucial for large earth-dams or embankments. If the points placed at the base of these structures experience very different seismic motions (as expected in near-source conditions) significant differential settlements and fractures of the embankment may occur, with the consequent reduction of structure safety. In the first chapter the peculiarities of the near source seismic propagation will be illustrated referring to a detailed literature review on such an issue. At the end of the chapter some preliminary numerical results will be provided on typical embankments and simplified source models. In the second chapter, the mathematical formulation of the DRM approach, will be provided. According to DRM it is possible to divide the problem at the site scale from the problem of seismic motion generation which consists in the simulation of the source and the seismic propagation until the site of interest. Once known the nodal displacements, it is possible to determine the effective nodal forces (Bielak et al., 2003) which represent in the second model (model 2) the actions due to the seismic source. In this research project it was developed a novel algorithm to export the stiffness matrix to be used for calculating the effective nodal forces at the interface of the detail model at the site scale. This procedure has been implemented in a commercial software based on finite difference method. In the third chapter the methodology to simulate the reference source mechanism (1980 Irpinia earthquake) and the propagating media on a "regional scale" will be illustrated. A quasi-deteministic approach will be used, in which the determination of the seismic motion generated by the fault will be reproduced by the use of the empirical Green's functions in the frequency domain (Discrete Wave Number Method, by Cotton & Coutant, 1997). The source model was reconstructed from numerous publications on 1980 Irpinia source mechanism (Westaway & Jackson, 1984; Westaway & Jackson, 1987; Bernard & Zollo, 1988; Bernard & Zollo, 1989; Cocco & Pacor, 1993). In the fourth chapter, the mathematical formulation necessary to solve the boundary value problem in the geotechnical field will be presented. Some constitutive models suitable to represent soil behaviour under cyclic loads will be briefly described. In the fifth chapter, the model at the "site scale" of DRM approach will be described. It includes an interesting case history of an earth dam that in the 1980 Irpinia earthquake suffered huge damage, probably just due to the near-source seismic propagation. With the input motion provided by the large DRM model (step I), the seismic response of Conza dam will be evaluated. Comparisons will be provided between the predicted response of the dam by the DRM procedure and the measured one, as interpreted by Brigante (2010). Finally, the main differences between the DRM approach developed during the PhD activity, and a more traditional one (where a unique input motion is adopted) will be highlighted and the engineering implications discussed.

Downloads

Downloads per month over past year

Actions (login required)

View Item View Item