熱薄型燃料引燃及過渡至逆向火焰延燒之研究

Abstract

本論文以數值計算方法，探討重力場中自然對流情況下，熱薄型固體燃料引燃，及過渡至逆向火焰延燒之機制。另外，二維輻射熱傳對向下火焰延燒行為的影響，亦詳加探討。內容涵括三個部分: 第一部份，發展一非穩態、二維的燃燒數學模式，來模擬在正常重力場中，垂直放置的薄型固體燃料，受外界輻射熱源而引燃及過渡至逆向火焰延燒之過程。高斯分佈輻射熱源峰值強度為5 W/cm2，半寬1 cm，用來引燃在靜止空氣含氧濃度23.3 %的籤維質材料。引燃過程以引燃點分為加熱階段和火焰發展階段。固相吸收輻射熱量後溫度上升至裂解溫度時，開始釋放出可燃氣與外界氧混合。而在引燃瞬間一預混火焰形成，隨後火焰向下傳播成為擴散火焰，最後達到穩態火焰延燒階段。引燃延遲時間隨輻射熱源峰值強度和氧濃度的增加而減少，而低於峰值強度和氧濃度的下限則無法引燃。 第二部份，改進第一部份的暫態燃燒模式，加上燃料消耗至燃盡點時灰燼不殘留的考量，而成為"邊界移動"的燃燒模式。在改變重力場為變化參數下，探討其對向下火焰延燒的效應。從0.5到5.8 倍正常重力而氧濃度23.3 %的七種不同重力場被研究，火焰傳播速度隨重力場增加而減少。在5倍重力場時為為不穩定火焰傳播，而在5.8倍重力場時為吹滅限。引燃延遲時間隨重力場增加而增加。吹滅限預測值比先前研究更接近實驗值。 第三部份，探討輻射對向下火焰延燒之效應，氣相輻射模式以二維P-1近似法模擬。而研究參數主要為變化重力場，亦即變化Damkohler 數和輻射對傳導係數。與一維氣相輻射模式相較，本研究顯示流線方向的熱輻射有對原始燃料預熱的效果，因而增加火焰傳播速度。另一方面，在低重力場時，流線方向熱輻射卻提供較大量的能量損失。而輻射熱源的分佈，在近燃料表面附近最強，其次在火焰尾端亦有較強的輻熱源。

The research topics in this dissertation consist of three parts. In part one, a time-dependent, two-dimensional combustion model was developed to describe the radiant ignition and subsequent transition to downward flame spread over a vertical thermally-thin solid fuel in normal gravity. An external radiant flux with a strength of 5 W/cm2 and a half width of 1 cm is used as the heat source to ignite cellulosic material in a quiescent environment with 23.3% oxygen concentration. The process consists of a heating stage and a flame developing stage, separated by an ignition. The solid fuel temperature rises as it is subjected to an incident heat flux. Later, pyrolysis generates enough fuel vapor to form the flammable mixture for ignition. A hot gas layer is generated adjacent to the heated surface and a natural convection is established nearby. During the ignition, a premixed flame propagates into the unburned mixture. Then, the flame spreads downward to further support itself. The spreading flame is mainly a diffusion flame. Eventually, a steady flame spread is reached. The ignition delay time is decreased as the peak flux value or the oxygen index increases. There are two limiting peak flux values, and a lower bound for oxygen concentration beyond which no ignition can occur. In part two, the transient combustion model developed in part 1 was adopted but modified with a moving boundary of burnout point. It was used to study the downward flame spread, subjected to an opposed induced flow resulting from the buoyancy, over a thermally-thin solid fuel in various gravitational acceleration. The emphasis was on the transient period since the ignition behaviors were not affected by the removal of ash. Seven parametric cases at various gravity levels ranging from 0.5 to 5.8 times normal gravity under 23.3% oxygen concentration environment were conducted. At g=5.0, it was identified as an unstable flame spread. At g=5.8, blowoff extinction, occurred at 9.64s after irradiation, was predicted. Ignition delay time was found to increase with an increase in gravity level. Comparing to the measurements obtained by Altenkirch et al. (1980) and the prediction by steady model of Duh and Chen (1991), the present computed flame spread rates show excellent agreement with measurements in the experimental domain ranged from 1 and 4.25 . The predicted blowoff limiting value at 5.8 appears closer to the experimental value at 4.25 than that predicted by steady model, whose value is at 11 . In part three, investigates how radiation heat transfer influences downward flame spread by presenting a gas phase radiation model, described by a two dimensional P-1 approximation method, to incorporate with the combustion model of Duh and Chen (1991). The parametric study is based on the variation of gravity, which changes the Damkohler number (Da) and radiation to conduction parameter (1/N ) simultaneously. Comparing the results with the previous studies of Duh and Chen (1991) and Chen and Cheng (1994), which only considered the radiation effect in cross stream direction, the role of stream-wise radiation was identified. The stream-wise radiation contributes to reinforce the forward heat transfer rate subsequently increasing the flame spread rate. However, this model also provides more directional radiation loss than that of Chen and Cheng (1994) and, in doing so, draws more energy out from the flame to further reduce its strength. The results indicates that the effect of heat loss is greater than that of enhancing the upstream heat transfer since the flame spread rate in the present model is always lower than the one predicted by Duh and Chen (1991). Finally, a contour of the Planck mean absorption coefficient distribution is illustrated to demonstrate the effectiveness of gas radiation distribution. It reveals that the strongest radiation occurs near the pyrolyzing surface and the other significant one is in the plume region.