A Novel Approach to Model a Gas Network

Abstract

The continuous uninterrupted supply of Natural Gas (NG) is crucial to today’s economy, with issues in key infrastructure, e.g., Baumgarten hub in Austria in 2017, highlighting the importance of the NG infrastructure for the supply of primary energy. The balancing of gas supply from a wide range of sources with various end users can be challenging due to the unique and different behaviours of the end users, which in some cases span across a continent. Further complicating the management of the NG network is its role in supporting the electrical network. The fast response times of NG power plants and the potential to store energy in the network play a key role in adding flexibility across other energy systems. Traditionally, modelling the NG network relies on nonlinear pipe flow equations that incorporate the demand (load), flow rate, and physical network parameters including topography and NG properties. It is crucial that the simulations produce accurate results quickly. This paper seeks to provide a novel method to solve gas flow equations through a network under steady-state conditions. Firstly, the model is reformulated into non-linear matrix equations, then the equations separated into their linear and nonlinear components, and thirdly, the non-linear system is solved approximately by providing a linear system with similar solutions to the non-linear one. The non-linear equations of the NG transport system include the main variables and characteristics of a gas network, focusing on pressure drop in the gas network. Two simplified models, both of the Irish gas network (1. A gas network with 13 nodes, 2. A gas network with 109 nodes) are used as a case study for comparison of the solutions. Results are generated by using the novel method, and they are compared to the outputs of two numerical methods, the Newton–Raphson solution using MATLAB and SAINT, a commercial software that is used for the simulation of the gas network and electrical grids.

Investigation of the Multi-Point Injection of Green Hydrogen from Curtailed Renewable Power into a Gas Network

“…The approach taken to modelling the gas network is similar to the approach taken by Osiadacz (1987) [33] and it has been previously used by Ekhtiari et al (2019) [34]. Where using Matlab, a set of differential equations solve the flow and pressure requirements in the gas network.…”

“…Because of the relatively slow response time of pipeline variables and large time spans in hourly steps, a quasi-transient approach is taken in combining the Equations ( 6) and ( 9) (the continuity and pressure drop equations). The gas network model calculates network variables and parameters while using the Newton-Raphson numerical iteration method [34]. For a given pipe ij , then the volumetric flow rate Q ij through the pipe can be expressed by Equation (10).…”

Colebrook’s Flow Friction Explicit Approximations Based on Fixed-Point Iterative Cycles and Symbolic Regression

“…The Colebrook equation has been known since the end of the 1930s and is based on an experiment by Colebrook and White [16] that had been performed a few years earlier using pipes with different inner surface roughness, from smooth to very rough. As the Colebrook equation is empirical [1,16], its accuracy can be disputed [17,18] (e.g., the new Oregon and Princeton experiment related to pipe friction [18]); nevertheless, the equation is widely accepted as a standard and is in common use in the design of water and gas pipe networks [19]. It can be also adapted for special cases such as air flow through fuel cells [20], water flow in rivers [21,22] and blood flow in blood vessels [23].…”

“…The approach with Padé approximants offers an iterative procedure in which the repetition of the computationally expensive logarithmic function is avoided [24]. Using only one logarithmic function together with simple rational functions and avoiding computationally expensive functions, such as exponential functions or functions with non-integer powers, the presented method saves the processor time [7][8][9][12][13][14], which is essential for effective simulations of large pipe networks [19].…”

Ideas From Bounded Confidence Theory Applied to Dynamical Networks of Interacting Free-Bodies

“…Mathematics of networks have been also used in the modeling of gas networks, see Ekhtiari et al [12] and in the modeling of electrical power systems, see Dassios et al [13,14], Dassios [15], and Cuffe et al [16], were a key motivation is to begin to link power flow analysis with the mature literature, see Chung and Graham [17], on spectral graph theory.…”