Please use this identifier to cite or link to this item: https://hdl.handle.net/1783.1/114358
Assessment of physical building vulnerability to landslides
| Author | Luo, Hongyu | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Issue Date | 2020 | ||||||||||||
| Abstract | Landslide hazards become more intense and frequent due to climate change and human activities, posing great risks to the built environment in landslide-prone areas. As large volumes of soil and/or rock are mobilized and travel long distances, these mass movement processes gain destructive power. It is impossible to stop such disastrous events of the nature; the only way is to reduce or avoid the risk. In such a case, risk assessment must necessarily be conducted to provide an objective measure of the risk. As one of the critical components in quantitative risk assessment (QRA), vulnerability assessment to landslides hazard is still in its infancy and considerable research gaps still exist with respect to the vulnerability assessment, hampering the development of QRA. Physical building vulnerability assessment establishes a relationship between debris flow intensity and building damage in a scientific manner. In the literature, little effort has been made to investigate the interaction mechanisms between mass flows and buildings, leading to a lack of understanding on the explicit failure mechanism in many vulnerability models. The mechanics-based building vulnerability model considering building-landslide interactions thus is still not available. In this study, the interaction processes between landslides and a typical reinforced concrete (RC-) building are discovered through an explicit finite element platform LS-DYNA. The Arbitrary Lagrangian–Eulerian (ALE) formulation, which allows automatic rezoning, is applied to simulate the landslide flow dynamics and the impact into the building. Three-dimensional RC-building is modelled using the Finite Element Method (FEM) considering detailed configurations. Seven impact cases with four different flow materials; namely low-intensity water flow, high-intensity water flow, debris flood, debris flow, low-intensity earth flow, moderate-intensity earth flow and high-intensity earth flow are studied. The evolution of physical building damage to landslides is investigated and different building failure modes to flows with increasing solid content, e.g., floods, debris flows and earth flows are compared. When a RC building is subject to a landslide impact, the frontal walls fail first due to their low out-of-plane strength. The side walls are damaged by the combination of the friction effect and lateral impact in the water flow, debris flood and debris flow cases whereas the side walls in the earth flow cases are mainly destroyed by the force transferred from the columns. Bending failure is the most common column failure mode. A forward pushover failure mode, which is a global failure of the building frame, is observed in the high-intensity water flow and debris flood cases. Upon impact by an earth flow, progressive local failures of the columns at the ground floor can occur and lead to backward collapse of the building. Upon impact by a debris flow, both the local column failure and the global pushover collapse of building can occur, as debris flow with high solid content is an intermediate flow sharing characteristics of both flood and earth flow. Five-class classification systems for RC-buildings impacted by an earth flow and a debris flow are proposed using multi-source information from field observations, numerical simulations and expert experience. The clear hierarchy of damage degree and the physical description of damage state provide more accurate evaluation of the damage state of buildings. Based on the identified physical damage process and failure mechanism, a novel reliability based building vulnerability model is established by explicitly considering the uncertainties on both landslide intensity and building material property. Two series of fragility models have been proposed based on practical debris-flow impact pressure models. Several debris flow intensity measures are investigated. A better indicator can be provided using the intensity measure that represents specific failure mechanism, for example, impact force (hv2) for force-dominated failures or overturning moment (h2v2) for moment dominated failures, where h and v are debris flow depth and velocity, respectively. The corresponding fragility surfaces best express the potential building damage. The intensity thresholds in the proposed fragility curves are consistent with those in empirical vulnerability curves. Further, the reliability based vulnerability analysis is extended to multi-hazard vulnerability assessment. The effect of hazard interactions on the physics of building damage is formulated from the perspective of triggering and temporal relations. A fragility analysis framework for generating physics-based vulnerability models for sequentially and concurrently occurring hazards is proposed. The framework is illustrated by evaluating the failure probability of a RC-building impacted by multiple surges of debris flows. Considering specific debris flow-building interaction mechanisms, nonlinear finite-element pushover analysis is adopted to obtain the building response under debris flow impact. By quantifying the physical damage caused by the primary debris flow, the cumulative damage effect of the sequentially occurring debris flows is formulated. The amplified damage effect of the concurrently occurring debris flows is however case specific. The interaction process between a building cluster and a landslide is studied based on two notable landslide cases, namely the Shen Zhen landslide and the Po Shan Road landslide. Blockage effect and domino damage effect of the building cluster are significant and should be considered in the landslide risk assessment and hazard mitigation work. | ||||||||||||
| Note |
Thesis (Ph.D.)--Hong Kong University of Science and Technology, 2020 xxvii, 272 pages : illustrations ; 30 cm HKUST Call Number: Thesis CIVL 2020 Luo |
||||||||||||
| Subject | |||||||||||||
| DOI | 10.14711/thesis-991012879661503412 | ||||||||||||
| Language | English |
||||||||||||
| Type | Thesis | ||||||||||||
| Access |
View details via DOIFiles in this item:
|
||||||||||||