Landsileds are very frequent events in many areas of the world due to the environment evolution. Case by case the at rest conditions of the slopes can be largely variable, passing from largely to poorly stable. For this reason, even small earthquake can trigger collapse phenomena of deposits which are in weakly or mederately stable state before the seismic excitation. "Earthquake-induced landslides have been documented from as early as 1789 B.C. and represent, still nowadays, a major concern, since they can result in remarkable damage in terms of both economic and human losses" [39]. It is sufficient to think that for many earthquakes, the landslides have caused much more damages compared to the ones caused by other effects of the seismic phenomena. So it is straightforward the reason why the study of such phenomena is of remarkable interest in the field of geotechnical engineering. The term "earthquake induced landslides" includes divers collapse events, whose triggering depends on a wide-ranging mechanical and environmental factors. In particular there can be distuinguished three main categories in line with the material type, landslides movements and geomorphology: (i) "falls" of boulders and/or blocks due to the detachment from a rock mass; (ii) "coherent landslides" mainly connected to a rotational/translational shear mechanism; (iii) "flowslides" (often associated with the liquefaction instability phenomena) which pecualiarity is to have a "fluid-like" movement of the sliding mass. In the present dissertation the attention is focalized on the third category of the previous mentioned: "earthquake indiced landslides" due to the triggering dynamic liquefaction instability phenomenon. For such a purpose, the main target is to define a constitutive model which can reproduce the main features of the liquefation instability phenomenon. The assumption of a simplified 1D geometry will be useful in order to analyze the mechanical effect of the liquefaction for granular soils excited by a dynamic solicitations. The present work is in fact a numerical/theoretical study on the relation between the assumptions made at the constitutive model level and the results coming from the dynamic monodimensionl numerical analysis (obtained by means of a Finite Element Method - FEM - code). The work comprises five chapters, whose contents is here summarized. In the Chapter 1 the problem of the liquefaction instability phenomenon is presented. In the Chapter 2 the concept of dilation is introduced: a deep explaination of the role of such parameter in analyzing the mechanical response of granular soils is then explained. A short bibliographic excursus will help to follow the logical path which will bring to the definition of the constitutive equations. In the Chapter 3 the constitutive model used for the dynamic monodimensional numerical analysis is formulated and described in the detailed. Several loading cases, for different sand in different conditions, will be analyzed in order to check the model reliability. The definition of such a model will proceed in step: at each step major complexity level will be added until the final version. The Chapter 4 comprises a brief introduction about the numerical code and the algorithm used for the field equations solution, defined in the simplified 1D geometry. Although its limitations, the 1D geometry can be conveniently used in order to analyze the liquefaction hazard for an infinite extendend slope. At the end of the chapter the results will be discussed focusing the attention on the limitation of the imposed constrained and the assumption of 1D geometry. Finally, in the last Chapter pros and cons of the constitutive model will be discussed. Limitations and improvements of the numerical code will be treated as well.
Le frane sono eventi frequenti in molte aree del mondo a causa della continua evoluzione ambien- tale. Caso per caso le condizioni a riposo dei pendii possono essere largamente variabili, passando da ampiamente a poco stabili. Per questo, anche deboli eventi sismici possono essere sufficienti ad innescare fenomeni di collasso di depositi che si trovano in uno stato debolmente o moderatamente stabile prime della scossa sismica. "Le frane indotte da eventi sismici furono documentate già nel 1789 D.C. e sono, ancora oggi, di grande interesse nel campo dell’ingegneria civile, dell’ambiente e del territorio a causa dei notevoli danni economici e sociali" [39]. E’ sufficiente pensare che in molti terremoti, i fenomeni franosi hanno provocato molti più danni rispetto a quelli derivanti da altri rischi connessi ai fenomeni sismici. Perciò è evidente il motivo per cui la valutazione della stabilità dei pendii in condizioni sismiche è di notevole importanza nel campo dell’ingegneria geotecnica. Il termine "frane sismoindotte" include diversi fenomeni di collasso, il cui avvenimento dipende da svariati fattori meccanici e ambientali. In particolare si possono distinguere tre categorie prin- cipali in funzione del tipo di materiale, di movimento del corpo di frana e della geomorfologia: (i) "caduta" di massi e/o blocchi a seguito del distacco da un ammasso roccioso; (ii) "frane coe- renti" principalmente connesse a meccanismi di taglio traslazionali/rotazionali; (iii) flussi detritici (spesso associati al fenomeno della liquefazione) per i quali si ha un movimento del corpo di frana paragonabile a quello di un fluido viscoso. Nel presente elaborato l’attenzione è focalizzata sulla terza categoria, cioè su quelle frane che si verificano a seguito di un collasso per liquefazione. A tale scopo, l’obiettivo principale è quello di formulare un modello costitutivo che sia in grado di riprodurre i gli aspetti principali connessi al fenomeno di instabilità per liquefazione. L’assunzione di una geometria semplificata 1D permetterà di analizzare gli effetti meccanici di instabilità per liquefazione di terreni incoerenti dovuta a sollecitazioni dinamiche. Il presente ela- borato è difatti uno studio numerico/teorico sulla relazione tra assunzioni a livello di modello costitutivo e risultati forniti da analisi dinamiche monodimensionali. L’elaborato comprende cinque capitoli, il cui contenuto sarà qui brevemente sintetizzato. Nel Ca- pitolo 1 si presenta il problema del fenomeno di instabilità per liquefazione. Nel Capitolo 2 si introduce il concetto di dilatanza e si spiega approfonditamente il ruolo chiave giocato da tale parametro nella descrizione della risposta meccanica di terreni incoerenti. Un breve excursus bibliografico aiuterà a seguire il percorso logico che porterà a definire il modello costitutivo imple- mentato nel codice numerico per analisi dinamiche monodimensionali. Prima dello studio del problema tramite analisi numeriche, nel Capitolo 3 è descritto nel dettaglio il modello costitutivo, per il quale si introducono via via successivi gradi di complessità. La bontà del modello nel riprodurre la risposta meccanica di diversi tipi di materiale in diverse condizioni di carico sarà supportata da svariate analisi parametriche e da considerazioni teoriche. Nel Capitolo 4 è introdotto brevemente il codice numerico e l’algoritmo di calcolo utilizzato per la risoluzione delle equazioni di campo per la geometria semplificata descritta in precedenza. No- nostante i suoi limiti intrinseci, il modello 1D può essere opportunamente usato per determinare il rischio di liquefazione di un pendio infinitamente esteso. Infine, nell’ultimo capitolo saranno evidenziati i pro e i contro di tale modello costitutivo, e le limitazioni e i miglioramenti che riguardano l’utilizzo del codice numerico utilizzato per l’analisi del rischio di liquefazione di pendii infinitamente estesi.
Liquefazione sismoindotta : modellazione costitutiva e analisi numeriche
CASAPENTA, GIUSEPPE
2016/2017
Abstract
Landsileds are very frequent events in many areas of the world due to the environment evolution. Case by case the at rest conditions of the slopes can be largely variable, passing from largely to poorly stable. For this reason, even small earthquake can trigger collapse phenomena of deposits which are in weakly or mederately stable state before the seismic excitation. "Earthquake-induced landslides have been documented from as early as 1789 B.C. and represent, still nowadays, a major concern, since they can result in remarkable damage in terms of both economic and human losses" [39]. It is sufficient to think that for many earthquakes, the landslides have caused much more damages compared to the ones caused by other effects of the seismic phenomena. So it is straightforward the reason why the study of such phenomena is of remarkable interest in the field of geotechnical engineering. The term "earthquake induced landslides" includes divers collapse events, whose triggering depends on a wide-ranging mechanical and environmental factors. In particular there can be distuinguished three main categories in line with the material type, landslides movements and geomorphology: (i) "falls" of boulders and/or blocks due to the detachment from a rock mass; (ii) "coherent landslides" mainly connected to a rotational/translational shear mechanism; (iii) "flowslides" (often associated with the liquefaction instability phenomena) which pecualiarity is to have a "fluid-like" movement of the sliding mass. In the present dissertation the attention is focalized on the third category of the previous mentioned: "earthquake indiced landslides" due to the triggering dynamic liquefaction instability phenomenon. For such a purpose, the main target is to define a constitutive model which can reproduce the main features of the liquefation instability phenomenon. The assumption of a simplified 1D geometry will be useful in order to analyze the mechanical effect of the liquefaction for granular soils excited by a dynamic solicitations. The present work is in fact a numerical/theoretical study on the relation between the assumptions made at the constitutive model level and the results coming from the dynamic monodimensionl numerical analysis (obtained by means of a Finite Element Method - FEM - code). The work comprises five chapters, whose contents is here summarized. In the Chapter 1 the problem of the liquefaction instability phenomenon is presented. In the Chapter 2 the concept of dilation is introduced: a deep explaination of the role of such parameter in analyzing the mechanical response of granular soils is then explained. A short bibliographic excursus will help to follow the logical path which will bring to the definition of the constitutive equations. In the Chapter 3 the constitutive model used for the dynamic monodimensional numerical analysis is formulated and described in the detailed. Several loading cases, for different sand in different conditions, will be analyzed in order to check the model reliability. The definition of such a model will proceed in step: at each step major complexity level will be added until the final version. The Chapter 4 comprises a brief introduction about the numerical code and the algorithm used for the field equations solution, defined in the simplified 1D geometry. Although its limitations, the 1D geometry can be conveniently used in order to analyze the liquefaction hazard for an infinite extendend slope. At the end of the chapter the results will be discussed focusing the attention on the limitation of the imposed constrained and the assumption of 1D geometry. Finally, in the last Chapter pros and cons of the constitutive model will be discussed. Limitations and improvements of the numerical code will be treated as well.| File | Dimensione | Formato | |
|---|---|---|---|
|
document.pdf
non accessibile
Descrizione: Testo della tesi in formato .pdf
Dimensione
15.38 MB
Formato
Adobe PDF
|
15.38 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/10589/140291