The great importance of underground facilities in terms of transportation and industrial utility is highly in contrast with the lack of a detailed and ruled seismic design procedure. Compared to above-ground structures, the underground ones have shown better performances in seismic events. Nevertheless, the high risk associated with their damage or collapse increases their vulnerability: the serviceability of wide network may be strongly affected and reduced in those eventualities. Seismic behaviour of underground structures is quite complex to understand and to describe. First of all, the nature of those structures itself is unique: they present a full interaction with surrounding geo-materials, along their whole external surface. Thus, design of underground structures, in the seismic stage, is highly sensitive to any simplified assumption, made to take into consideration soil-structure interaction (SSI). Stress-strain field observed in the lining is influenced by three crucial features: soil and structural constitutive models (non-linear), soil-structure relative flexibility and interface strength properties. Earthquake characteristics and geological site-specific features also affect structural performance. In most cases, underground facilities have a significant 3D geometry (e.g. tunnels of great longitudinal extension), which increases the degree of complexity of the investigation. Seismic design of underground structures has been often neglected, for the lack of specific codes or guidelines and also because, as mentioned above, underground structures are considered less vulnerable to earthquake loading. Moreover, it must be noticed that those kind of facilities are very difficult to be monitored and possible damaging is hard to be detected in detail. A poor case-history is available nowadays, while a great number of uncertainties and open problems still remains. Experimental campaigns has recently started, exploiting the great progresses made in scale-modelling, through geotechnical centrifuge facilities, even though their outcomes still require extreme care to be correctly interpreted. This study is intended to clarify some aspects related to seismic behaviour of underground structures, comparing experimental results (i.e. geotechnical centrifuge test performed within the project TUNNELSEIS) with numerical predictions. In particular, the behaviour of a segment of rectangular lining has been investigated. For the numerical simulation, the finite element method is implemented, using ABAQUS. The analyses have been performed on prototype-scale models under plane strain conditions. While the tunnel behavior is assumed to be elastic, the soil non-linear behavior during shaking is simply modeled using a Mohr-Coulomb failure criterion, coupled with a non-associative plastic flow-rule. The model parameters are adequately calibrated using common G-gamma-D curves, available in literature. The soil–tunnel interface is also accounted and simulated adequately. A parametric analysis is presented in the ensuing chapters, testing the role of different features on final outcomes, such as sand state conditions (i.e. peak or critical), interface strength properties, plasticity. Finally, the internal forces of the tunnel lining are also evaluated with available closed form solutions, usually used in the preliminary stages of design and compared with the experimental data and the numerical predictions. The numerical analyses can generally reproduce reasonably well the recorded response. Differences between the experimental data and the numerical results are mainly attributed to the simplification of the used model and to differences between the assumed and the actual mechanical properties of the soil and the tunnel during the test.
La grande importanza rivestita dalle opere in sotterraneo, nel trasporto e nell'industria, è altamente in contrasto con la mancanza di una procedura di progettazione antisismica regolata e dettagliata. Le strutture in sotterraneo, rispetto a quelle superficiali, hanno mostrato prestazioni migliori in occasione di eventi sismici. Tuttavia, l’alto rischio associato al loro danneggiamento o collasso incrementa la loro vulnerabilità: il funzionamento di un’ampia rete potrebbe essere fortemente compromessa e ridotta, in queste eventualità. Il comportamento sismico delle strutture in sotterraneo è abbastanza complesso da capire e da descrivere. Innanzitutto, la natura stessa di queste strutture è unica: esse presentano un’interazione continua con il terreno circostante, lungo tutta la loro superficie esterna. Perciò, la progettazione antisismica di strutture in sotterraneo è molto sensibile a qualunque ipotesi semplificativa, fatta per tenere in considerazione l’interazione terreno-struttura (SSI). Lo stato di sforzo-deformazione osservabile nel rivestimento è influenzato da tre fattori cruciali: i comportamenti meccanici (non lineari) di terreno e struttura, la flessibilità relativa terreno-struttura e le caratteristiche meccaniche dell’interfaccia. Anche le caratteristiche intrinseche dell’evento sismico e le particolarità geologiche sito-specifiche influenzano la performance strutturale. Nella maggior parte dei casi, le opere sotterranee hanno una geometria 3D non trascurabile (e.g. tunnel di grandi estensioni longitudinali), che aumenta il grado di complessità dell’analisi. Il progetto antisismico di opere in sotterraneo viene spesso trascurato, per la mancanza di regolamenti o linee guida e anche perché, come detto, tali strutture sono considerate meno vulnerabili nei confronti carichi sismici. Inoltre, è necessario notare che questi tipi di infrastrutture richiedono estesi sistemi di monitoraggio e il loro possibile danneggiamento è difficilmente riconoscibile. Ad oggi, è presente una case-history limitata, mentre permane un gran numero di incertezze e problemi irrisolti. E’ recentemente iniziata una serie di campagne sperimentali che sfruttano i grandi progressi fatti nel campo della modellazione in scala, attraverso la centrifuga geotecnica, anche se i risultati richiedono ancora molta attenzione nell'essere interpretati. Questo studio si propone come obiettivo quello di chiarire alcuni aspetti relativi al comportamento sismico di strutture sotterranee, mettendo a confronto risultati sperimentali (i.e. quelli provenienti da un test effettuato in centrifuga geotecnica, all'interno del progetto TUNNELSEIS) con quelli provenienti da analisi numeriche. In particolare, si è investigato il comportamento di un segmento di tunnel a sezione rettangolare. Per quanto riguarda le simulazioni numeriche, si è fatto uso del metodo degli elementi finiti, attraverso il software ABAQUS. Le analisi sono state effettuate alla scala del prototipo, in condizioni plane strain. A differenza del rivestimento, il cui comportamento meccanico è stato considerato elastico, il comportamento non lineare del terreno, durante la fase di scuotimento, è stato modellato in maniera semplificata, facendo uso di un criterio Mohr-Coulomb, accoppiato ad una legge di flusso plastico non associativa. I parametri del modello sono stati calibrati facendo uso di curve G-gamma-D fornite da letteratura. L’interazione terreno-struttura è stata tenuta in debita considerazione e modellata adeguatamente. Nei capitoli successivi viene presentata un'analisi parametrica, per verificare gli effetti causati da una serie di fattori sui risultati finali, come le condizioni rottura per la sabbia (i.e. di picco o critiche), le caratteristiche meccaniche dell'interfaccia, le deformazioni plastiche. Infine, sono state valutate le forze interne al rivestimento, in accordo con forme chiuse disponibili, comunemente utilizzare nelle fasi preliminari di progettazione, e confrontate con i dati sperimentali e con le predizioni numeriche. Le analisi numeriche riproducono in maniera ragionevole buona la risposta sperimentale. Le discrepanze tra dati sperimentali e risultati numerici vengono principalmente attribuite alle semplificazioni di cui si è fatto uso nei modelli processati e soprattutto alla differenza tra proprietà meccaniche reali e presunte del terreno, nonché alla loro evoluzione durante il procedere dell'esperimento.
Experimental and numerical investigation on the seismic response of rectangular underground structures
GATTI, FILIPPO
2013/2014
Abstract
The great importance of underground facilities in terms of transportation and industrial utility is highly in contrast with the lack of a detailed and ruled seismic design procedure. Compared to above-ground structures, the underground ones have shown better performances in seismic events. Nevertheless, the high risk associated with their damage or collapse increases their vulnerability: the serviceability of wide network may be strongly affected and reduced in those eventualities. Seismic behaviour of underground structures is quite complex to understand and to describe. First of all, the nature of those structures itself is unique: they present a full interaction with surrounding geo-materials, along their whole external surface. Thus, design of underground structures, in the seismic stage, is highly sensitive to any simplified assumption, made to take into consideration soil-structure interaction (SSI). Stress-strain field observed in the lining is influenced by three crucial features: soil and structural constitutive models (non-linear), soil-structure relative flexibility and interface strength properties. Earthquake characteristics and geological site-specific features also affect structural performance. In most cases, underground facilities have a significant 3D geometry (e.g. tunnels of great longitudinal extension), which increases the degree of complexity of the investigation. Seismic design of underground structures has been often neglected, for the lack of specific codes or guidelines and also because, as mentioned above, underground structures are considered less vulnerable to earthquake loading. Moreover, it must be noticed that those kind of facilities are very difficult to be monitored and possible damaging is hard to be detected in detail. A poor case-history is available nowadays, while a great number of uncertainties and open problems still remains. Experimental campaigns has recently started, exploiting the great progresses made in scale-modelling, through geotechnical centrifuge facilities, even though their outcomes still require extreme care to be correctly interpreted. This study is intended to clarify some aspects related to seismic behaviour of underground structures, comparing experimental results (i.e. geotechnical centrifuge test performed within the project TUNNELSEIS) with numerical predictions. In particular, the behaviour of a segment of rectangular lining has been investigated. For the numerical simulation, the finite element method is implemented, using ABAQUS. The analyses have been performed on prototype-scale models under plane strain conditions. While the tunnel behavior is assumed to be elastic, the soil non-linear behavior during shaking is simply modeled using a Mohr-Coulomb failure criterion, coupled with a non-associative plastic flow-rule. The model parameters are adequately calibrated using common G-gamma-D curves, available in literature. The soil–tunnel interface is also accounted and simulated adequately. A parametric analysis is presented in the ensuing chapters, testing the role of different features on final outcomes, such as sand state conditions (i.e. peak or critical), interface strength properties, plasticity. Finally, the internal forces of the tunnel lining are also evaluated with available closed form solutions, usually used in the preliminary stages of design and compared with the experimental data and the numerical predictions. The numerical analyses can generally reproduce reasonably well the recorded response. Differences between the experimental data and the numerical results are mainly attributed to the simplification of the used model and to differences between the assumed and the actual mechanical properties of the soil and the tunnel during the test.| File | Dimensione | Formato | |
|---|---|---|---|
|
2014_04_Gatti.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
89.63 MB
Formato
Adobe PDF
|
89.63 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/89641