An efficient parallel dynamic component tracking model of gas networks by decoupling hydraulic, thermal, and convection-diffusion processes

To ensure the fair trade of hydrogen blending natural gas in pipeline networks with multiple gas sources and different components, it's important to obtain the real-time gas component and quality at different nodes. The virtual metering of the gas components can be efficiently achieved by the accurate operation simulation of pipeline networks. This paper proposes a parallel and decoupling model to simulate the time-varying gas components and important parameters of the gas network. The simulation model rigorously considers the hydraulic, thermal, and convection-diffusion processes. The hydraulic process is initially simulated based on the iterative initial values for temperature and component at each time step. Subsequently, the thermal and convection-diffusion processes can be performed based on the variables at the initial moment, parameters calculated from the hydraulic process, and the iterative initial value of the component profile. Since the thermal process can be solved based on the iterative initial value of the component profile and the hydraulic variables, and the convection-diffusion process can be solved by the hydraulic variables, the simulations of the above two processes can be performed in parallel to accelerate the simulation speed. The Newton-Raphson algorithm is employed to solve the above three processes until the termination condition is met. The whole iteration process will cease when the disparities in gas temperature and composition profiles between successive iterations fulfill the convergence criteria. The effectiveness of the parallel and decoupling model is confirmed by two cases, and the dynamic simulation model is applied under various boundary conditions. The results show that the proposed model can efficiently reduce the computation time by 75.29 % compared to the original coupling model while maintaining accuracy in comparison with TGNET software. Additionally, the model exhibits robust adaptability across diverse boundary conditions.