Experimental study of smoke back-layering length under coordinated ventilation in underground interconnected tunnel

Abstract

Extensive research has been conducted on smoke back-layering length (L) in single-tube tunnels, but studies on underground interconnected tunnels remain limited. This study conducted 408 experiments, considering four fire locations, six heat release rates (HRRs) and various ventilation schemes. The variation in smoke backflow under different fire locations during coordinated ventilation in underground interconnected tunnel was investigated. It is found that higher HRR results in larger backflow. When the branch tunnel is ventilated separately, smoke spreads more easily to the tunnel entrance as the fire source moves further into the branch tunnel. A predictive model for smoke back-layering was developed for individually ventilated branch tunnel, based on the Richardson number and accounting for the variation in fire location. During coordinated ventilation in underground interconnected tunnel, back-layering length is inversely proportional to the velocity of the branch tunnel (v₁). In contrast, the influence of the main tunnel velocity (v₂) on back-layering depends on the fire location. When the fire source remains outside the branch tunnel, v₂ facilitates smoke movement downstream, reducing L as v₂ increases. Conversely, when the fire source enters the branch tunnel, v₂ impedes smoke movement downstream, causing L to increase as v₂ rises. Taking into account variations in fire location, a predictive model for smoke back-layering length in underground interconnected tunnel under coordinated ventilation has been established.