The Resilience of Critical Infrastructures

THE RESILIENCE OF CRITICAL INFRASTRUCTURES Resilience term derives from the Latin word "resilio", which means to jump back. Recent natural disasters led to questioning how communities can recover, and the most consolidated approach is to study the interdependence between different networks. In recent years, also the behaviour of nonstructural components and systems during natural disasters is becoming relevant in the resilience field. City-scale simulations are a possibility to study the multiple cascading effects between critical infrastructures. Failures in portions of power grid can lead to consequence in networks such as the transportation and water or in the evacuation and disasters management. However, few real data about power components are available in the literature. In the first part of the thesis, two methodologies are proposed to design and analyze a distribution power network looking to substations and transformers. This work is part of a larger study on a virtual city called Ideal City that is developed to be a resilience-based designed community. This town is freely inspired to the city of Turin. Starting from the city map, population and apparent power charge density data, the system is designed neighbourhood by neighbourhood considering real parameters. Substations and equipment are installed inside buildings and their fragilities are related. Running a seismic scenario, a vulnerability analysis of the network is carried out and the resilience's response is given in terms of people without energy supply and Megawatts lost. The purpose is to identify quickly after a disaster the city areas more affected by power outages to maximize efforts in helping people and restoring the energy supply as soon as possible, temporarily with generators and then with new components. In the second part of the thesis, the focus is on the water distribution network. During a research period in the University of California, Berkeley a new methodology to assess a pipe network analysis in Python is proposed thanks to a project sponsored by EBMUD, the San Francisco water utility. Starting from the relevant literature, a case study is performed making a comparison between different methods to solve the hydraulics equations and to solve the mathematical non-linearity of the system. In the annexe, the python code with all the rows commented is provided to make this work available for further developments. In the last chapter, an Agent-Based Model (ABM) of the city of Turin is presented. The transportation network of the city is converted in python and the behaviour of agents moving on the roads is defined. This model works as a traffic model thanks to the storage of hundreds of data from Google about the real traffic in Turin, representing in the proper way cars movements during working days and weekends. This ABM represents the standard situation before the occurring of an earthquake and the improvement of this approach could be an evacuation model where agents move according to different rules. The main purpose of this work is to analyze the main critical infrastructures inside big cities to develop a comprehensive tool useful for city planners and managers to enhance the community resilience, referring to real data and proper indices.