CFD Simulation and ANN Prediction of Hydrogen Leakage and Diffusion Behavior in a Hydrogen Refuelling Station

Hydrogen refuelling stations are an important part of the infrastructure for promoting the hydrogen economy. Since hydrogen is a flammable and explosive gas, hydrogen released from high-pressure hydrogen storage equipment in hydrogen refuelling stations will likely cause combustion or explosion accidents. Studying high-pressure hydrogen leakage in hydrogen refuelling stations is a prerequisite for promoting hydrogen fuel cell vehicles and hydrogen refuelling stations. In this work, an actual-size hydrogen refuelling station model was established on the ANSYS FLUENT software platform. The computational fluid dynamics (CFD) models for hydrogen leakage simulation were validated by comparing the simulation results with experimental data in the literature. The effects of ambient wind speed, wind direction, leakage rate and leakage direction on the diffusion behaviors of the released hydrogen were investigated. The spreading distances of the flammable hydrogen cloud were predicted using an artificial neural network for horizontal leakage. The results show that the leak direction strongly affected the flammable cloud flow. The ambient wind speed has complicated effects on spreading the flammable cloud. The wind makes the flammable cloud move in certain directions, and the higher wind speed accelerates the diffusion of the flammable gas in the air. The results of the study can be used as a reference for the study of high-pressure hydrogen leakage in hydrogen refuelling stations.