The survey of the damage occurred on land, buildings and infrastructures extensively affected by liquefaction, coupled with a comprehensive investigation of the subsoil properties enables to identify the factors that determine the spatial distribution of the phenomenon. With this goal, a database was created in a Geographic Information platform merging records of local seismicity, subsoil layering evaluated by cone penetration tests and groundwater level distribution for the relevant case study of San Carlo (Emilia Romagna- Italy) struck by a severe earthquake in 2012. Here liquefaction phenomena were observed on a portion of the village in the form of sand ejecta, lateral spreading and various damages on buildings and infrastructures. The location of damage allows to test possible relations with the factors characterizing susceptibility, triggering and severity of liquefaction. The relation among the different variables has been herein sought by training a specifically implemented Artificial Neural Network. A relation has thus been inferred between damage and thickness of the liquefiable layer and of the upper crust, seismic input and soil characteristics.

Prediction of liquefaction damage with artificial neural networks

L. Paolella
;
E. Salvatore;R. L. Spacagna;G. Modoni;
2019

Abstract

The survey of the damage occurred on land, buildings and infrastructures extensively affected by liquefaction, coupled with a comprehensive investigation of the subsoil properties enables to identify the factors that determine the spatial distribution of the phenomenon. With this goal, a database was created in a Geographic Information platform merging records of local seismicity, subsoil layering evaluated by cone penetration tests and groundwater level distribution for the relevant case study of San Carlo (Emilia Romagna- Italy) struck by a severe earthquake in 2012. Here liquefaction phenomena were observed on a portion of the village in the form of sand ejecta, lateral spreading and various damages on buildings and infrastructures. The location of damage allows to test possible relations with the factors characterizing susceptibility, triggering and severity of liquefaction. The relation among the different variables has been herein sought by training a specifically implemented Artificial Neural Network. A relation has thus been inferred between damage and thickness of the liquefiable layer and of the upper crust, seismic input and soil characteristics.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11580/73637
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
social impact