With the wide application of hydrogen-doped natural gas (HBNG) in end users, laying pipelines in urban, comprehensive pipe corridors has become increasingly common. However, the leakage and diffusion of hydrogen-doped natural gas in confined or semi-confined spaces (e.g., utility corridors) can pose a severe safety hazard, as methane and hydrogen gas mixtures have a higher risk of explosion. Therefore, studying hydrogen-doped natural gas's leakage and diffusion behavior in the comprehensive pipe gallery is essential to ensure its safe operation. This paper establishes a numerical model to study the diffusion law of hydrogen-doped natural gas leakage, and the concentration distribution under different conditions is analyzed. The evolution of the leakage and diffusion of hydrogen-doped natural gas under different hydrogen doping ratios, wind speeds, inlet and outlet spacing, and leakage port sizes were simulated, and the effects of these factors on gas diffusion and explosion hazard volume were discussed. In addition, the backpropagation neural network (BPNN) combined with the global optimization capability of genetic algorithm and the nonlinear mapping capability of neural network is used in this paper to provide a reliable technical means for the accurate prediction of explosion risk volume, which has important application value in the safety design of pipeline corridor and accident prevention and control. The results show that the diffusion range of gas in the pipe corridor increases significantly with the increase of hydrogen mixing ratio. When the hydrogen mixing ratio decreases from 20% to 0, the explosion volume of CH4 decreases by 12.25%. The explosion volume of H2 is close to 0. At the same time, the greater the distance between inlet and outlet, the wider the spread of the dangerous area, when the distance between inlet and outlet is reduced from 200 m to 100 m, the explosion volume of CH4 is reduced by 99.92%, and the explosion volume of H2 is reduced by 100%. In the case of emergency, the reasonable design of Distance between inlet and outlet and air flow configuration can help to quickly discharge harmful gases, ensure the safe evacuation and equipment maintenance in the pipe corridor, and have guiding significance for the ventilation design of the pipe corridor.
Keywords: Backpropagation neural network; Hydrogen-blended natural gas; Leakage diffusion; Utility tunnel.
© 2025. The Author(s).