The number of tunnels in seismic regions has grownsignificantly in recent decades.It has usually been assumed that tunnels perform better than surface structures during seismicevents. However, recent cases have shown that tunnels can be significantly damaged byseismic events. Thus, an evaluation of their response to earthquakes has become increasinglynecessary. This paper presents a FEM blind prediction of centrifuge tests on a reduced scaletunnel. Themain objective of the paper is to evaluate the numerical model that reproduces theresponse recorded in the centrifuge. The centrifuge tests involved a tunnel in dry sand. Thenumerical simulationwas performed on the physical-scalemodel of the transverse direction ofthe tunnel, which is of prime importance, as it can show large stress–strain levels in the tunnellining. The tunnel behaviour was assumed to be visco-linear-elastic, while the soil behaviourwas assumed to be visco-elastic-perfectly plastic. The soil model parameters were calibratedon the basis of laboratory tests performed on the sand used for the test. The comparisonbetween the experimental and numerical results is presented in terms of acceleration inthe time and frequency domains. The experimental and numerical settlements of the sandsurface and displacements of the sand-tunnel system, as well as the bending moments andhoop forces acting in the tunnel are also compared. Increments of the bending moments andhoop forces are also evaluated using the closed-form solution proposed by Wang (Seismicdesign of tunnels: a simple state-of-the-art design approach. Parson Brinckerhoff, New York,1993) and Penzien (Earthq Eng Struct Dyn 29:683–691, 2000). A very good agreementbetween the experimental and numerical results is achieved in terms of horizontal accelerationtime-histories and their Fourier spectra, as well as in terms of vertical displacements of thesand surface. Moderate differences exist between the experimental and numerical bending moments and hoop forces; experimental, numerical and analytical increments of the bendingmoments and hoop forces are in a quite good agreement with each other.
Numerical modelling of centrifuge tests on tunnel-soil systems
ABATE G.;MASSIMINO, MARIA ROSSELLA;
2015-01-01
Abstract
The number of tunnels in seismic regions has grownsignificantly in recent decades.It has usually been assumed that tunnels perform better than surface structures during seismicevents. However, recent cases have shown that tunnels can be significantly damaged byseismic events. Thus, an evaluation of their response to earthquakes has become increasinglynecessary. This paper presents a FEM blind prediction of centrifuge tests on a reduced scaletunnel. Themain objective of the paper is to evaluate the numerical model that reproduces theresponse recorded in the centrifuge. The centrifuge tests involved a tunnel in dry sand. Thenumerical simulationwas performed on the physical-scalemodel of the transverse direction ofthe tunnel, which is of prime importance, as it can show large stress–strain levels in the tunnellining. The tunnel behaviour was assumed to be visco-linear-elastic, while the soil behaviourwas assumed to be visco-elastic-perfectly plastic. The soil model parameters were calibratedon the basis of laboratory tests performed on the sand used for the test. The comparisonbetween the experimental and numerical results is presented in terms of acceleration inthe time and frequency domains. The experimental and numerical settlements of the sandsurface and displacements of the sand-tunnel system, as well as the bending moments andhoop forces acting in the tunnel are also compared. Increments of the bending moments andhoop forces are also evaluated using the closed-form solution proposed by Wang (Seismicdesign of tunnels: a simple state-of-the-art design approach. Parson Brinckerhoff, New York,1993) and Penzien (Earthq Eng Struct Dyn 29:683–691, 2000). A very good agreementbetween the experimental and numerical results is achieved in terms of horizontal accelerationtime-histories and their Fourier spectra, as well as in terms of vertical displacements of thesand surface. Moderate differences exist between the experimental and numerical bending moments and hoop forces; experimental, numerical and analytical increments of the bendingmoments and hoop forces are in a quite good agreement with each other.File | Dimensione | Formato | |
---|---|---|---|
79_Massimino.pdf
solo gestori archivio
Tipologia:
Versione Editoriale (PDF)
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
3.14 MB
Formato
Adobe PDF
|
3.14 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.