本研究利用一自行設計開發之紊流尾流燃燒器，來觀測一由上往下傳播預混層焰與紊流尾流交相干涉之情形。由於預混焰受紊流尾流渦漩陣列的影響會產生拉伸效應，為了探討拉伸效應和紊流燃燒速度，本實驗使用高速雷射斷層攝影術及質點影像測速儀兩種技術，配合研究級甲烷或丙烷與乾燥空氣所混合之預混燃氣作一系列實驗。尾流渦漩基本上可分為二種，擴張應變渦對及壓縮應變渦對，實驗觀察到壓縮應變渦對與預混焰的交相干涉，比擴張應變渦對具有較高的燃燒速度，壓縮應變渦對可直接將火焰帶入順時針旋轉的渦心內燃燒，由渦心而層層迅速往徑向方向燃燒，其另一半逆時鐘旋轉渦漩則相對燃燒較慢，這是因為渦對軸心線與預混焰平面呈一非直角夾角再加上浮力效應，故會有此一不對稱燃燒情形發生。 沿著雷射斷層火焰面來計算火焰局部拉伸率(含曲率項和應變項)，發現越靠近擴張應變渦漩之渦心時，其拉伸率值越大，此結果與Driscoll et al. (1994)在單一軸對稱渦漩(軸心線與預混焰平面呈90°)之結果相同。當越靠近渦心時， 沿著火焰面切線方向的速度梯度(應變項)越大，故有較大的拉伸率。吾人發現曲率項對火焰拉伸之影響遠比應變項大，因此可用曲率項來近似火焰拉伸率。由拉伸率之機率密度函數分佈，可發現拉伸率會隨著時間的增加而減小，顯示非定常效應，若紊流能量不能由外界繼續提供，紊流尾流結構會迅速消散，加上燃燒生成物具高溫會大大提高生成物之運動黏滯係數(最高約是原反應物之25倍)，故皺摺火焰會有層流化的現象產生。 為了量測不同紊流強度對紊流燃燒速度的影響，吾人於平板上加裝不同厚度之擾流板來增加紊流強度，所得結果並與先前實驗和理論結果作比較。本實驗所得之紊流燃燒速度結果明顯大於先前我們在十字型燃燒爐(近似等向性紊流場)所獲之結果，主要原因為後者乃一三維紊流場，可觀測到有部分火焰會相互撞擊，造成整體火焰面積並非隨著紊流強度作線性增加，而使得紊流燃燒速度增加幅度不如本實驗。然而紊流尾流會隨時間而消散，從紊流燃燒速度的量測可明顯看出兩不同區域的存在，因此對本實驗所得之紊流燃燒速度，雖然其具有學術參考價值，但必須小心的看待，因其燃燒速度會隨時間作小改變。 This research uses a self-designed turbulent wake burner to investigate a downward propagating premixed flame interacting with a turbulent wake. Since the interaction of flame and wake can produce the so-called flame stretching effects. To study the stretching effects and ST, we apply high-speed laser tomography and particle image velocimetry (PIV), use research-grade methane or propane -air mixtures, and conduct a series of experiments. Basically, there are two types of wake vortices, a compressive strain (counter-rotating) vortex pair and an extensive strain vortex pair. It is found that flame-wake interactions in compressive strain vortices are more intense than in extensive strain vortices. This is because the compressive strain vortex pair can engulf directly the flame into the clockwise-rotating vortex core. Burning starts at the vortex core and through layers and layers radially. The other half, counter-rotating vortex has slower burning rates. This asymmetry of burning is due to the facts that the axis of the vortex pair is not perpendicular to the flame front and the buoyancy effect matters. We calculate the local flame stretching rate (including the curvature rate and the aerodynamic strain rate) along the wrinkled flame front at different times using laser tomography and PIV techniques. For the extensive strain vortex pair, values of the stretching rate increase along the wrinkled flame front from the vortex tip to the position which is close to the vortex core. This result is similar to that of Driscoll et al. (1994) using a single axisymmetric vortex pair interacting with a premixed flame. It is found that the curvature term is much more important than the strain rate term. Thus, the stretching rate may be approximated by the curvature term alone. From the probability density function of the stretching rate, we found that the stretching rate changes with time , indicating that the unsteady effect cannot be neglected. In order to measure the effect of turbulent intensity to ST, we put the thin disturbance slices with different heights on the edge of the sliding plate. The results are then compared with previous experimental and theoretical results. It is found that the present ST measurements are much higher than our previous data using a cruciform burner which can produce a near-isotropic turbulent flow field. This may be because in the 3-D near-isotropic turbulent flow field many flame-flame collisions and annihilations are frequently observed. Therefore, the increase of the total flame area is not linearly proportional to the turbulent intensity, values of ST obtained in 3-D near-isotropic turbulence field using the cruciform burner are found to be much less than that of the present measurements for 2-D flame-wake interactions. However, the latter is not stationary and must be viewed with caution.
