| Disaster chain,or cascading disaster,is a phenomenon in which an initial disaster triggers a sequence of other disasters,increasing the damage and causing the deterioration of the disaster situation.The initial event of the typhoon disaster chain is usually accompanied by strong winds,rainstorms,and storm surges,which in certain environment will cause second-order disasters,such as floods,landslides,debris flows,and seawater backflow,forming a complex chain of hazards.Critical infrastructure systems(CISs)provide basic goods and services for sustainable social production and people’s lives,and due to the fundamental importance,they are also known as the urban lifeline systems.The interdependency between CISs makes it easy for a failure within one system to spread to multiple systems.For instance,strong wind breaks the power system,which not only causes direct economic losses,sometimes may also ignite a series of cascading failures,causing large-scale power outages,water cuts,gas supply interruptions,traffic jams,and communication interruptions.The traditional singledisaster strategy for emergency management and decision-making is unable to cope with such multi-hazard,multi-object decision-making situation.In order to efficiently deploy resources and to improve decision-making ability,we should,first,understand the common patterns of the typhoon hazard chain and the mechanism of cascading failures in CISs;secondly,build models to make reasonable predictions of the development trends of disasters and the risks of the disaster-bearing bodies,so that a clearer pattern of the disaster chain can be pictured and we can be better prepared for the uncertainty.This thesis studies the typhoon disaster chain in CISs,and the work can be summarized as follows.(1)We summarize the typical patterns of typhoon hazard chains and how CISs are influenced by these hazards and their intrinsic interdependencies in reference to previous studies and literature.Based on the knowledge about disaster chain,we analyze the chain of disasters caused by the super typhoon ‘Mangkhut’.(2)We put forward a complex-network-based model to simulate the macroscopic typhoon disaster chain.The model takes CISs as nodes,the interdependencies as edges,and the state of a node as the destruction of CIM caused by the typhoon hazard chain.The model dynamically simulates the states of CISs under the attack of typhoon hazards,and these states are affected by their self-healing capacities,time delay coefficient,and hazard intensities.We also propose a key node identification method based on network topology for decision support.Compared with other complex-network–based models of disaster chain,our model innovatively takes the interdependencies between CISs as the triggering factors of disaster chains,and the spatial scale is unified,making it easier for decision-makers to focus on important aspects and identify key nodes.The model is validated using case ‘Mangkhut’.(3)Based on the macroscopic model of disaster chain,we further propose a microscopic model that simulates the spreading of cascading failures in CISs.The model dynamically predicts the state the coupled network of multiple CISs under the attacks of spatially distributed typhoon hazards.The spatial distribution attributes of both hazards and hazard-bearing bodies,the intensities of hazards,the interdependencies between CISs are all taken into consideration.We implement the model using CISs of power,communication,and subway in Shenzhen,as well as the data of strong winds and storm surge caused by "Mangkhut".Results show that the model accurately predicts the peak of the disaster and the overall trend,but it is still a virtually impossible task to quantitatively simulate the complex disaster system.The weaknesses of the model and future work are discussed at the end of the thesis. |
