Leakage Simulation and Prediction for High-Pressure Natural Gas Pipeline in a Confined Space

Abstract

With economic development and the burgeoning need for energy, numerous underground gas pipelines have been deployed for operation. Buried gas pipelines often suffer failures as a result of adverse factors, including corrosion, uneven subsidence, and damage caused by third parties. To investigate and predict the diffusion behavior following a natural gas pipeline leakage, Fluent software is utilized to develop a numerical model for the leakage and diffusion of high-pressure natural gas within a confined space. This model incorporates the Soave–Redlich–Kwong equation of state, which is widely recognized for its exceptional precision in characterizing the behavior of natural gas under high-pressure conditions. The study focuses on the analysis of leakage and diffusion behavior as well as the examination of how pipeline operating pressure and leakage diameter have an impact on the dispersion of leaked gas. Further, the prediction model for the diffusion distance of the hazardous area is developed, employing the least-square method and finite element calculations. The results show that, during the leakage process, a vortex and velocity region emerge, extending along the confined space. The farther away from the leakage hole above the pipeline, the higher the overall concentration of the gas. Moreover, the horizontal diffusion distance of gas at the bottom of the pipeline is considerably smaller than that above it. However, gas tends to readily accumulate in a high concentration area at the bottom of the pipeline. Elevating the leakage diameter and the operation pressure leads to a significant rise in gas concentration and the horizontal diffusion of the hazardous area. It is worth noting that the leakage diameter has a more pronounced effect on gas diffusion than does the pressure. The prediction model proposed in this study effectively anticipates the horizontal diffusion of the hazardous area within confined spaces.