Coupling effect of multiple factors on the diffusion behavior of leaking natural gas in utility tunnels: A numerical study and PIV experimental validation

Abstract

The ignitable and explosive nature of natural gas poses a significant threat to its safe operation when incorporated into utility tunnels. Currently, the safety technology for gas pipelines in utility tunnels is not yet perfect, and understanding the diffusion law of leaked gas is fundamental to gas accident prevention and control technology. To accurately predict the gas diffusion pattern under the coupling effect of multiple factors, this study investigated the adaptability of Realizable k-ε model for simulating impinging jets through particle image velocimetry (PIV) experiments and computational fluid dynamics (CFD) simulations, simulated the gas diffusion behavior under various working conditions, and revealed the influence of different factors on the warning response time and the combustible gas cloud volume. The results demonstrated that the Realizable k-ε model accurately represented the flow field structure of the impinging jet, and the main characteristic parameters of each flow field zone closely aligned with the PIV results. Furthermore, the overall difference between the simulation results of this study and experimental results was within 4.8% in the validation of CH4 concentration, these findings indicated that the Realizable k-ε model was suitable for simulating gas diffusion processes in utility tunnels. The leakage direction significantly influenced the diffusion behavior within the first 10 s but had less impact later on. The leakage location, pipeline operating pressures, leakage size, leakage shape, and air change rate considerably impacted the warning response time, and the longitudinal plane over the center of the natural gas pipeline was the preferred plane for the installation of gas detectors. The coupling effects of air change rates and leakage size significantly affected the combustible gas cloud volume. Moreover, the coupling effect of multiple factors contributed to the most dangerous working condition in the range of this study, with the combustible gas cloud volume reaching a maximum of 8.25 m3. In conclusion, the numerical simulation model based on Realizable k-ε model could accurately predicted the diffusion behavior of leaked natural gas in utility tunnels under the coupling effect of multiple factors and built up the most dangerous gas leaking accident scenario. The findings of this study will contribute to the development of an accident warning and control plan for utility tunnels.