Search

9 batches of unconfined methane-air deflagration tests were conducted to investigate the influential factors on the deflagration load. The methane-air mixture was filled in the cubic tents with different volumes of 1 m3, 8 m3and 27 m3. Methane concentrations varied from 7.5% to 11.5%, covering both lean and rich combustion regimes. The ground overpressure was recorded by using the piezoresistive sensors. An ideal film was validated before the tests. The effects of the methane concentration and gas volume on the deflagration load were analyzed. Based on the results, a new prediction model for the unconfined methane-air deflagration load was proposed. In addition, the model was compared with some existing models.

29 batches of vented explosion tests were conducted in a 12 m 3 concrete chamber filled with methane–air mixtures to investigate the effects of methane concentration and venting pressure on the development of overpressure inside the chamber. The deflagrations were vented from a square side window with a venting area of 0.64 m 2 upon rupture of the vent cover. The venting pressures were varied by using six different types of vent covers, and determined by performing a numerical simulation. Methane concentrations in the mixed gas varied between 6.5 and 13.5 vol.%, covering both lean and rich combustion regimes. The generation conditions of the four types of overpressure–time profiles with different overpressure transients were summarized. Among the overpressure transients, the Δ P 1 caused by failure of the vent cover and the Δ P 4 resulted from the coupling between acoustic mode and flame were basically dominant. The rate of Δ P 1 rise as well as peak value of Δ P 1 and Δ P 4 showed a same trend of first increasing and then decreasing with the methane concentration from lean to rich. They reached their maximum value at the methane concentration of about 9.5%, whereas the rate of Δ P 4 rise was found to be insensitive to the methane concentration. The peak value of Δ P 1 increased with the venting pressure, while that of Δ P 4 first increased and then decreased as the venting pressure increased. In addition, the rates of Δ P 1 and Δ P 4 rise were both insensitive to the venting pressure. The occurrence of Δ P 4 was depended on the methane concentration and the venting pressure. The differences in the behavior of these overpressure transients suggest their different generation mechanisms.