Modelamiento numérico de desplazamientos horizontales inducidos por licuación.
Loading...
Date
2025
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
La licuación de suelos es un fenómeno crítico en zonas sísmicas que ocurre cuando suelos saturados y sueltos pierden rigidez y resistencia debido al aumento de presión de poros durante un sismo. Este proceso puede dar lugar a desplazamientos laterales significativos y a otros daños estructurales, afectando infraestructuras como puentes y muelles.
Los modelos constitutivos y el método de elementos finitos han emergido como herramientas esenciales en la ingeniería geotécnica para analizar la respuesta del suelo ante cargas dinámicas. En particular, el uso de modelos avanzados como el PDMY02 permite representar de manera realista la degradación de la rigidez y la acumulación de deformaciones plásticas, mientras que la implementación de elementos SSPQUAD-UP en plataformas como OpenSees facilita la simulación del comportamiento acoplado del suelo y la presión de poros durante la licuación.
En este trabajo se desarrolló una metodología integral que incluyó la validación de un modelo numérico mediante la simulación de ensayos centrífuga del proyecto LEAP, en los cuales se replicaron tendencias de desplazamientos, aceleraciones y excesos de presión de poros. Posteriormente, la metodología fue aplicada a un caso de estudio real, modelando el perfil geotécnico del sitio del puente Juan Pablo II afectado en el sismo del Maule 2010, utilizando datos obtenidos de sondajes SPT.
Los resultados obtenidos evidenciaron que, si bien el modelo numérico logró reproducir de forma general el comportamiento del suelo en términos de presiones de poros y desplazamientos relativos entre estratos licuables y no licuables, se observó una tendencia a sobreestimar los desplazamientos laterales. Esta discrepancia se atribuye a limitaciones en la representación de las condiciones de entrada, en particular al uso de registros sísmicos obtenidos a distancia del sitio, lo que resalta la necesidad de continuar perfeccionando los parámetros y condiciones del modelo para lograr una mayor precisión en la predicción de la licuación en escenarios reales.
Soil liquefaction is a critical phenomenon in seismic areas that occurs when saturated, loose soils lose stiffness and strength due to increased pore pressure during an earthquake. This process can lead to significant lateral displacements and other structural damage, affecting infrastructures such as bridges and docks. Constitutive models and the finite element method have emerged as essential tools in geotechnical engineering to analyze soil response under dynamic loads. In particular, the use of advanced models like PDMY02 allows for a realistic representation of stiffness degradation and the accumulation of plastic deformations, while the implementation of SSPQUAD-UP elements in platforms such as OpenSees facilitates the simulation of the coupled behavior of the soil and pore pressure during liquefaction. In this work, an integrated methodology was developed that included the validation of a numerical model through the simulation of centrifuge tests from the LEAP project, in which trends in displacements, accelerations, and excess pore pressures were replicated. Subsequently, the methodology was applied to a real case study, modeling the geotechnical profile of the Juan Pablo II Bridge site affected by the Maule 2010 earthquake, using data obtained from SPT borehole tests. The results obtained showed that, although the numerical model generally succeeded in reproducing the soil behavior in terms of pore pressures and relative displacements between liquefiable and non liquefiable strata, there was a tendency to overestimate lateral displacements. This discrepancy is attributed to limitations in representing the input conditions, particularly the use of seismic records obtained from stations located far from the site, which underscores the need to continue refining both the model parameters and conditions to achieve greater accuracy in predicting liquefaction in real scenarios.
Soil liquefaction is a critical phenomenon in seismic areas that occurs when saturated, loose soils lose stiffness and strength due to increased pore pressure during an earthquake. This process can lead to significant lateral displacements and other structural damage, affecting infrastructures such as bridges and docks. Constitutive models and the finite element method have emerged as essential tools in geotechnical engineering to analyze soil response under dynamic loads. In particular, the use of advanced models like PDMY02 allows for a realistic representation of stiffness degradation and the accumulation of plastic deformations, while the implementation of SSPQUAD-UP elements in platforms such as OpenSees facilitates the simulation of the coupled behavior of the soil and pore pressure during liquefaction. In this work, an integrated methodology was developed that included the validation of a numerical model through the simulation of centrifuge tests from the LEAP project, in which trends in displacements, accelerations, and excess pore pressures were replicated. Subsequently, the methodology was applied to a real case study, modeling the geotechnical profile of the Juan Pablo II Bridge site affected by the Maule 2010 earthquake, using data obtained from SPT borehole tests. The results obtained showed that, although the numerical model generally succeeded in reproducing the soil behavior in terms of pore pressures and relative displacements between liquefiable and non liquefiable strata, there was a tendency to overestimate lateral displacements. This discrepancy is attributed to limitations in representing the input conditions, particularly the use of seismic records obtained from stations located far from the site, which underscores the need to continue refining both the model parameters and conditions to achieve greater accuracy in predicting liquefaction in real scenarios.
Description
Tesis presentada para optar al título de Ingeniero/a Civil.
Keywords
Simulación por computadores, Terremotos, Licuación