Research On The Resilience Assessment And Enhancement Of Interdependent Infrastructure Networks

As a category of engineering facility network,power network,water network,gas network and other infrastructure networks produce,provide and distribute their products or services necessary for economic and social operation.Once disrupted or failed,they will seriously affect industrial production,people’s life,economic operation or public security.With the continuous development of science,technology and social economy,not only each infrastructure network becomes larger and more complex,but also the close and complex spatial and functional interdependencies among different types of infrastructure networks are formed and thus interdependent infrastructure networks(IINs)come into being.As a supporting system for the normal operation of modern society and economy,IINs protection has attracted extensive attention.Particularly in recent years,many governments,organizations and academics have reached consensus after learning from the experience and lessons of many disruptive events.In order to cope with increasingly unavoidable disruptive events,it is urgent to build highly resilient IINs,i.e.,to strengthen their comprehensive capacity of resisting,absorbing and recovering from the impacts of disruptive events.Given that assessing and enhancing IINs’resilience is an important basis to achieve this goal,only when resilience assessment job is completed,policy makers can grasp the "safe level" of the assessment object,find potential resilience enhancing strategies and measures,and explore its resilience enhancing path.Scientifically assessing and enhancing IINs’resilience has become the key to guiding and promoting practical activities in construction of highly resilient IINs.Therefore,based on IINs’ resilience connotation,this dissertation aims to assess and enhance IINs’ resilience under transient and gradual disruption scenarios that IINs may suffer from through analyzing,modeling,simulating and optimizing the disruption and restoration process of IINs,which takes advantages of multiple theory and methods,including complex network theory,random field theory,network maximal flow theory,sequential Monte Carlo simulation method and optimization method.The main research contents are as follows.(1)IINs’ resilience and its assessment basis.Firstly,IINs is defined,including the meaning of infrastructure network and spatial and functional interdependencies,and the modes and strength of the two kinds of interdependencies.And in time dimension,two kinds of disruption scenarios that IINs might suffer from,i.e.,transient and gradual disruption scenarios,are summarized.Secondly,on clarifying the meaning of IINs,resilience,a conceptual model of IINs’ resilience is developed.Finally,the disruption and restoration process of IINs and the corresponding performance curve are analyzed,and the assessment metric of IINs’ resilience is constructed.Furthermore,the influencing factors of IINs’ resilience are sorted out.(2)Research on the interdependencies modelling method and the performance estimation model of IINs.Establishing a performance estimation model in the process of disruption and restoration of IINs provides a basis for determining the performance curve of IINs and then assessing IINs’resilience.Firstly,physical failure state and functional failure state of IINs’components are defined and modeled,and common cause failure and chain failure caused by spatial interdependency,and cascading failure caused by functional interdependency are sorted out.Secondly,spatial interdependency and functional interdependency are modeled based on random field theory and master-slave logic,respectively.Thirdly,a mixed integer linear programming model for estimating IINs’ performance during the disruption and restoration process based on maximum flow theory is established,which depicts both spatial and functional interdependencies among IINs.Finally,the feasibility and effectiveness of the proposed model are verified by experiment analysis on artificial IINs.(3)Resilience assessment method of IINs based on simulation of disruption and restoration process.Resilience assessment methods for IINs under transient and gradual disruption scenarios are developed.Firstly,based on random field theory and sequential Monte Carlo simulation method,the disruption and restoration process of IINs are modeled and simulated under transient and gradual disruption scenarios respectively.Secondly,combined with the performance estimation model of IINs in Section(2).the assessment methods of IINs’resilience under transient and gradual disruption scenarios are given.Finally,the effectiveness of the proposed resilience assessment methods are verified by assessing the resilience of the interdependent electricity-gas-water network in Shelby County,Tennessee,and the interdependent electricity-water network in a certain city in China under certain disruption scenarios respectively.Further,the effects of network topology,spatial interdependency strength,functional interdependency strength,fragility of IINs’ components,and restoration strategies on IINs’ resilience are analyzed through simulation experiments on artificial IINs.(4)Research on resilience enhancement model of IINs.Resilience enhancement models of IINs are carried out from the perspectives of pre-event protection plan optimization and post-event restoration plan optimization respectively.Firstly,from the perspective of pre-event protection plan optimization,based on the idea of minimizing consequence under disruptive event then enhancing resilience,a mixed integer programming model for seeking the optimal pre-event protection plan of IINs’ components under protection budget constraint is developed,which considering both spatial and functional interdependencies among IINs,and then a genetic algorithm is designed.Secondly,from the perspective of post-event restoration plan optimization,based on the idea of minimizing cumulative consequences under disruptive event then enhancing resilience,another mixed integer programming model that integrates restoration selection and scheduling decisions under restoration resources constraint is constructed,which considering both spatial and functional interdependencies among IINs,and also a genetic algorithm is designed.Finally,the feasibility and effectiveness of the proposed models are verified through the interdependent electricity-gas-water network in Shelby County,Tennessee,USA and the interdependent electricity-water network in a certain city in China,and the resilience enhancement utilities of pre-event protection budget and post-event restoration resources are analyzed.In summary,this dissertation can enrich modelling methods of interdependencies among infrastructure networks,and expanded theories and methods for resilience assessment and enhancement of IINs.The research will also provide assistance for predicting consequence of IINs under disruptive event,assessing disruptive event response capabilities of IINs,and optimizing pre-event protection plan and post-event restoration plan of IINs.