Modelling and simulation of natural gas distribution networks in the presence of hydrogen injection

The 2050 long-term strategy, defined by the European Commission, leads towards zero greenhouse gas emissions by 2050. Reduction of carbon dioxide emissions can be achieved substituting high carbon fossil fuels (coal and oil) with natural gas, renewable sources and green fuels. In the next years, the gas system will play a central and crucial role in the global energy market. Due to modifications of international gas trade flows and rise of demand, the existing gas infrastructures will necessarily have to be expanded, upgraded and renovated in the immediate future. Furthermore, power-to-gas technology is a potential solution to support and accelerate the penetration of renewable sources and the decarbonization of the energy sector. The excess of power generated by renewable energy sources is used by power-to-gas facilities to produce alternative green fuels. The resulting gas, such as hydrogen or synthetic natural gas, can be injected and stored into the existing gas grid. Subsequently, the green low/zero-carbon fuel blended with the traditional natural gas would enable to reduce carbon dioxide emission of industrial, commercial and residential gas customers. In this new scenario, it is essential to study, model and simulate the integration and operation of gas networks in the energy system. It is also very important to evaluate the impact of alternative fuel injections on the properties and composition of the gas delivered to the users connected to the gas grid. In this thesis, a steady-state and dynamic one-dimensional gas network tool, named "Gas Network Solver", is developed. The research focuses on mathematical modelling of city gate station (source), pipe, reducing station, valve, demand node and interchange node elements, which compose a gas distribution network. Particular attention is dedicated to the implementation of the mathematical model of the gas and the algorithm for quality tracking in order to analyse and simulate multiple types of gas sources. The tool proposed is validated by comparing results of three test cases to solutions obtained with a commercial software application, named "Scenario Analysis Interface for Energy Systems" (SAInt), and data from other models available in the literature. Finally, a case study considering a real medium-pressure and low-pressure gas distribution network, composed by about 2289 elements and located in a hilly area of central Italy, is analysed. After the simulation and analysis of the network in the actual scenario, a possible solution to decarbonize the network is carried out. The installation of a power-to-gas facility, associated effects on behaviour of the network and quality of the gas delivered are studied. The investigation also aims to evaluate the maximum amount of hydrogen injectable respecting gas standards defined by the Italian Regulatory Authority for Energy, Networks and Environment.