A highly robust gas network simulation approach through an inherently solvable problem formulation for network states far from intended design points

Pipeline networks are an efficient and widespread transportation system for supplying natural gas as well as increasingly green gas to Europe. However, they are exposed to risks arising from environmental factors, accidents, crime and political issues. This work contributes to the assessment of gas transmission networks and enables a targeted improvement of their resilience. Therefore, the objective of this work is to develop a simulation tool that enables gas transmission system operators (TSOs) to identify weak spots, e.g. potential bottle necks, in their pipeline networks. To this end, a steady-state gas network simulation approach is developed that provides fast results for gas supply deficiencies for complete sets of multi-event disruption scenarios, enabling the identification of the most important network elements. This paper describes how to implement a computationally fast, numerically robust, and physically accurate gas network simulation and assessment tool. By solving mass and momentum balance equations, it accounts for key network elements such as flow control valves, pressure control valves, compressor stations, gas sources, and gas consumers. The central feature that distinguishes this approach from classical steady state solvers is the robustness of the gas flow calculation through the choice of a problem boundary condition formulation, which results in an inherently solvable system of equations. This feature is key in ensuring mathematical convergence of predictions for gas network states that strongly deviate from their original design point, where boundary conditions must not become mathematically invalid but need to reflect physically reasonable behavior. The capabilities of the system are demonstrated using a fictitious gas network and a representative national gas transmission network. Finally, the limits of applicability are outlined based on the size of the network under consideration and the types of analyses.