Flame Interactions and Burning Rates of Discrete Liquid Fuel Arrays under Crossflow

Yuhang Chen,Yanli Miao,Shangqing Tao,Xiaolei Zhang,Fei Tang,Longhua Hu
DOI: https://doi.org/10.1016/j.combustflame.2024.113645
IF: 4.4
2024-01-01
Combustion and Flame
Abstract:Since multi-fires frequently occur outdoors in real scenarios, experiments were conducted to elucidate the flame interactions and burning rates of discrete liquid fuel arrays in crossflows. N-heptane pool was employed to generate diffusion flames, with the crossflow speeds u ranging from 0 similar to 3 m/s and spacings between liquid pools S ranging from 5 similar to 40 cm. All the cases were divided into 6 states, involving 2 merging conditions (Merging, Non-merging) multiplied by 3 flame interaction modes (Mode I: due to induced indraft in still air, Mode II: as a joint result of induced indraft and crossflow, Mode III: crossflow dominated) concerning the flame dynamics. The results showed that the burning rates of individual liquid pools based on burnout time were observed to decrease from the array center to the boundaries and corners in still air for merged flame but decreased from the leeward to the windward side in most cases under crossflow. The peak burning rate of the entire liquid-fuel pool array measured by balance and the global burning rate for liquid fuel array m(g)* based on the burn-out time against S/D grows firstly (Regime 1), then decreases (Regime 2), and finally levels-off (Regime 3) at both windless and windy conditions. Moreover, m(g)* continuously increases with the increased crossflow speed, and reaches a peak value within the range of S/D = 1.5-2. Both positive effects (PE) and negative effects (NE) were analyzed to explain the changing trends of burning rates of the entire array. PE is always dominant (m(g)*>= 1) for all cases due to the enhanced heat feedback and availability of oxygen under crossflow. The increment of global burning rates in crossflow compared to that in still air was characterized by their ratio, which appeared to be a function of Froude number employing different characteristic length scales for merged and non-merged flames of liquid fuel square array.
What problem does this paper attempt to address?