Mine Ventillation Modelling at Fires: Dynamic Simulation of Toxic Gases and Heat Transfer, and Escape Routes Estimation

Abstract

This thesis explores the dynamics of mine ventilation systems in underground fire incidents using Ventsim Visual software dynamic simulation on the Maleevsky mine ventilation model. The fire sources is a load-haul-dump (LHD) loader, with 340 kg of diesel fuel, 306 kg of hydraulic oil, 2,200 kg of rubber tyres, and 2,000 kg of plastics, which represents a total fire load of 163 GJ. Four fire scenarios are tested during a three-hour simulation: an LHD fire where all ventilation fans are running (Scenario 2), and the same fire but the main underground fan fails at 60 minutes after ignition (Scenario 3). Six movable monitors measure CO concentrations, temperature, and velocity of airflow in the affected part of production. A further scenario (Scenario 4) involves the same fire, but with the main fan failing at 30 minutes (in the growth phase, not after the fully developed phase) to determine the impact of the timing of the fan failure on exposure to toxic gases. A fifth scenario (Scenario 5) is considered in which the main fan fails at the time of ignition (t = 0 min), and an additional seventh monitor is placed in Airway 4934 near the source of fire to capture the local response of the fire-source to this monitor. The findings indicate that in both cases the highest CO concentrations of about 2,300 ppm were recorded at the nearest monitor, which is almost two times more than the Immediately Dangerous to Life or Health (IDLH) of 1,200 ppm. An important counterintuitive result appears: the failure of fans at 60 minutes decreases the severity of the fire in general, as it restricts the oxygen supply, resulting in the fire switching to oxygen-restricted burning, instead of sustained combustion by flaming. Scenario 2 has a sustained plateau of CO above IDLH during 60-90 minutes; Scenario 3 has a peak and a rapid decrease, halving or decreasing by half the duration of IDLH. Peak CO at the farthest monitor is reduced to 350 ppm (Scenario 3) - a 74% decrease compared to 1,350 ppm (Scenario 2). In Scenario 4 (fan failure at 30 minutes), the effect is even more pronounced: maximum CO at the nearest monitor is limited to about 550 ppm (76 % less than Scenarios 2 and 3), the IDLH level is not exceeded in any location in the production section, and the cumulative CO exposure is reduced by 87 % compared with the baseline of all fans operating. In Scenario 5 (loss of the fans at t = 0 min) the local CO at the fire source is peaked at the highest level in any scenario (approximately 3,300 ppm in the fire airway, measured at the new Airway 4934 monitor) but the lateral spread through the rest of the production section is severely constrained: 12 of the 14 candidate escape destinations register a route-maximum CO of 0.0 ppm and full visibility, and all 14 routes remain viable, indicating that the pre-ignition fan loss produces the hazard footprint which is intense locally but spatially confined. Analysis of 14 evacuation destinations escape routes indicates that the spatial relationship between evacuation pathway and the fire gas transport corridor determines the viability of routes with routes passing through the fire plume recording lethal CO values of 1,9191937ppm and parallel routes recording only 178ppm. Failure of the fans provides a second evacuation window which is not available under sustained ventilation and is created after about 90120 minutes. These results are quantitative measures to the fan shutdown dilemma, which proves that controlled fan management can be used as an active fire suppression means in underground mines emergencies.