Towards Risk and Resilience Quantification of Gas Networks based on Numerical Simulation and Statistical Event Assessment

A Linearized Numerical Solution for Steady-State Simulations of Gas Networks

“…The developed method has been implemented in an algorithm that has an inner cycle, which first calculates the actual and linearized branch conductivities, then solves (1), obtains branch flow rates using (3), recalculates the hydraulic resistance coefficients with (12)- (15), pressure source values according to (26), (27) and self-consumption of compressors. The inner cycle ends with testing of convergence criteria.…”

“…A validation case study has been carried out to test the developed method with both flow equations and a method described in [15] and some of the results will be presented here. The settings for the convergence criteria were maxEPS 1 = 5 kPA (0.05 bar); maxEPS 2 = 0.01 m 3 /s; max-EPS 3 = 0.001 p.u;maxEPS 4 = 20 kPa (0.2 bar).…”

“…The node pressure results obtained by the proposed method with flow equation (4) (labelled as "Developed method 1" in graphs) and with modified equation (5) (labelled as "Developed method 2" in graphs), ones obtained by Leong as well as ones calculated using a method developed by Sebastian Ganter [15] are presented in Fig. 3.…”

“…This network was chosen as it represented a real network in the Mexico valley with the pressure values reported as measurement data [8], [11], [15], [41]. The following assumptions used by the other authors [8], [11], [15], [41] were adopted: S g = 0.65 p.u. ; T avg ≈ 191.67 K; E p = 0.95 p.u.…”

“…There is a method, which separates the linear equations obtained using Kirchof's laws and the nonlinear flow equations applying linearisation to the nonlinear part of the SSM [12]. Accurate results are presented with fast convergence but only for particular pipeline networks, but recently another paper has added an approach for modelling of non-pipeline elements [15]. A different method adapts the hydraulic conductivity of GTS elements so that the flow rate value remains the same, but it is a function directly of presure drop or drive power in case of compressors [8], [9].…”