固体材料构型和尺度是影响可燃特性的重要因素，由于构型的多样性以及构型和尺度耦合效应的复杂性，目前对其影响规律和机理尚未形成系统的认识。本文围绕固体材料可燃特性中的构型和尺度效应，循序渐进地开展了四个方面的实验研究工作：材料尺度和流动条件对一维（球形）材料燃烧与火焰熄灭影响机制；不同环境条件下二维（柱状）材料火焰传播特性，认识逆向（流动与火焰传播方向相反）、同向（流动与传播方向相同）、横向（流动垂直传播方向）火焰传播控制机理；用典型局部构型表征复杂材料结构特征，观测不同几何尺寸条件下两种典型构型材料火焰传播和可燃极限，研究复杂构型影响规律和燃烧机理；对向上火焰传播中的传热机制和尺度效应进行研究，分析远离熄灭极限和近极限火焰特征及尺度效应影响，可为深刻认识和改进向上传播测试方法提供参考。研究工作阐明了材料构型和尺度对可燃性影响机理，为全面认识和评价材料可燃性、发展可燃性评价方法提供科学基础。主要研究内容和结果如下：

开展落塔微重力实验，对静止环境中不同尺度的球形材料燃烧与火焰熄灭特性开展研究，并在地面常重力下对滞止流场中的球形材料可燃极限开展实验，揭示了材料尺度、流动速度、火焰拉伸、热损失等对火焰熄灭的影响机制。当球形材料直径增大时，火焰曲率减小引起火焰热反馈降低，而材料热惯性增加导致材料内部升温缓慢，固相传导热损失增大；火焰熄灭由固相热损失（包括表面辐射和传导热损失）控制，当材料直径较大时热损失达到临界值并引起火焰熄灭。在滞止流场驻点附近，球形材料燃烧的极限氧气浓度随火焰拉伸率a（与气流速度和材料直径相关）变化，二者构成U形可燃极限曲线；在火焰冷熄区，随着固相热损失减小可燃极限曲线向低氧浓度平移，可燃范围增大；在火焰吹熄区，可燃极限对固相热损失不敏感。球形材料火焰熄灭时的临界燃烧速率与环境流动、材料尺寸、固相热损失等有关，这些因素的影响可以用组合参数a^1/2ln(1+B)表征（B为传递数），临界燃烧速率随a^1/2ln(1+B)线性增加，这与文献中柱状材料在滞止流场中的燃烧速率变化规律相似，但是相同a^1/2ln(1+B)条件下，柱状材料有更大的临界燃烧速率。

对柱状材料表面逆向、同向、横向火焰传播开展实验研究，获得了流动速度、氧气浓度、火焰拉伸率等对火焰传播影响规律，分析环境因素和材料尺度的影响机制。关于逆向火焰传播，微重力实验揭示出低速流动中火焰传播的特点，结合地面实验结果将火焰传播速度与气流速度的关系分为三个不同的机理控制区：辐射区、传热区和化学反应区；尝试用火焰拉伸率关联不同气流速度时的火焰传播速度，发现由于不同尺寸材料的固相热损失不同，火焰拉伸率不能唯一地确定火焰传播速度。关于同向火焰传播，实验表明传播速度随气流速度增大而线性增加，随直径增大而减小，对于厚材料推导出了火焰传播速度对气流速度的依赖关系。火焰横向传播时，传播速度受气流速度和直径这两个参数共同影响，由于热厚材料的固相热损失不受尺度影响，火焰传播速度可与火焰拉伸率的1/2次方建立关联。

用凸台和凹槽结构代表典型的局部构型，研究复杂构型材料逆向、同向火焰传播与可燃极限特性。两种构型中的棱角位置附近，火焰向材料表面传递的热流量增加，材料局部的热惯性减小，因而此处的逆向、同向火焰具有最大的传播速度。逆向火焰传播中，相对于平板材料，棱角的存在使材料表面各处的火焰传播速度增大，并最终能够获得相同的传播速度；同向火焰传播中，材料表面火焰传播未达到稳定，且不同位置的火焰传播速度存在差异。将具有局部构型的材料等效成柱状材料，利用几何系数表征材料的受热程度，并基于柱状材料火焰传播理论，建立了复杂构型材料火焰传播速度的预测公式。实验结果和理论分析表明，复杂构型材料的可燃极限由其最小厚度决定，近极限条件下，材料最小热惯性主导火焰熄灭，复杂构型的极限氧气浓度与具有该构型最小厚度的平板材料的极限氧气浓度相同。

对材料可燃性测试方法—向上火焰传播试验的材料尺度（宽度和厚度）效应，以及近极限火焰特性进行研究。在不同的氧气浓度和环境压力下，通过详细测量火焰温度、固相内部温度、表面温度、质量损失速率等参数，获得热流量、传递数及变化规律，通过改变氧气浓度和环境压力，得到了这两个环境参数组合形成的不同可燃极限条件。远离熄灭极限时，随着平板材料宽度或厚度增加，火焰温度和火焰辐射热流量增加，传递数增大。近极限条件下，材料宽度不同时传递数差别不明显，但是厚度越小传递数越大，传递数随材料尺寸的变化规律与远离熄灭极限不同。与近极限的情况相比，远离熄灭极限条件时的传递数更大，而近极限时的传递数与氧气浓度有关，传递数呈现先减小再增大的非单调变化。根据传递数与火焰驻离距离之间的理论关系计算获得了远离极限和近极限条件下的传递数，与实验测量结果相比，远离极限条件下，由于未考虑燃料侧向扩散与固相热损失影响，理论计算结果略微偏大；近极限条件，理论计算与实验测量的差别更为显著，需要考虑近极限火焰特点对该理论进行完善。

The geometry configuration and scale of the solid materials are the main factors affecting the material flammability. Due to the diversity of geometry configurations and the complexity of geometrical configuration and scale coupling effects on the burning characteristics, a comprehensive understanding of their impact on the combustion mechanism is still lacking. Experimental research has been carried out step by step to investigate the effects of geometry configuration and scale on the flammability of solid fuels, involving four aspects. Firstly, the controlling mechanisms for the one-dimensional (spherical) fuel burning and flame extinction are determined. Secondly, the characteristics of flame propagation over two-dimensional (rod-shaped) fuel are examined under different flow conditions, including opposed flow (flow opposite to the direction of flame spread), concurrent flow (flow in the same direction as the flame spread), and horizontal flow configuration (flow perpendicular to the direction of flame spread). Thirdly, the effects of configuration and scale on solid fuel with complex geometries are investigated, using typical local configurations to characterize their burning behavior. Fourthly, heat transfer mechanisms and scale effects in upward flame spread configuration are examined, analyzing flame characteristics both far from and near the extinction limit, which offers valuable insights for a better understanding and enhancement of upward flame spread test methods. This work has clarified the controlling mechanism of material configuration and scale on flammability, providing a basis for comprehensive understanding and flammability evaluation. The main contents and conclusions of this paper are as follows:

The flammability characteristics of spherical solid fuel with different diameters are studied in the quiescent environment provided by the drop tower. Combined with the spherical fuel burning characteristics near the stagnation-point in stagnant flow in normal gravity, the effects of scale, flow velocity, stretch rate, and heat loss on flame extinction are investigated. The heat feedback from the flame to the fuel surface caused by the curvature effect is decreased as the increased of fuel diameter, and the larger thermal inertia results in a slower temperature rise of the fuel which further increases the total heat loss including surface radiation and solid-phase conduction. The surface radiation is not varied with the fuel diameter, and the larger solid-phase conduction will result in flame extinction when the diameter reaches the critical value. The limit oxygen concentration at the stagnation point is found to be U-shaped with the stretch rate. The solid phase conduction heat loss influences the flammable range in the quenching limit. The smaller the heat loss is, the larger the flammable range is. The heat loss has no significant effect on the flammability limit in the blowoff limit. The critical burning rate of spherical materials increases linearly with a^1/2ln(1+B). Even though the value of a^1/2ln(1+B) is the same, the critical burning rates for spherical and cylindrical fuel are still different and the cylindrical fuel has the larger burning rate.

Studies on opposed, concurrent, and horizontal flame spread over cylindrical fuel have been conducted to investigate the effects of flow velocity, oxygen concentration, and flame stretch rate on the flame spread. The opposed flame spread experiment in microgravity reveals flame spread behavior in low-speed forced flow. Combining the results from the microgravity and normal gravity experiments, three distinct regimes are identified to delineate the variation of flame spread rate with gas flow velocity, namely radiation regime, heat transfer regime, and chemical kinetics regime, and the flame spread characteristics at different regimes are analyzed in detail. The attempt to use the flame stretch rate to establish a correlation between flame spread rates and rod size under forced flow conditions reveals that the flame stretch rate alone cannot determine the flame spread rate, as solid-phase heat loss varies with rod size. The concurrent flame spread rate increases linearly with the forced flow velocity but decreases with diameter. The dependence of the flame spread rate on the forced flow velocity is derived for thick materials. The horizontal flame spread rate is affected by flow velocity and fuel diameter. Since the solid-phase heat loss of thermally thick materials is not affected by the scale, the horizontal flame spread rate is proportional to the square root of the flame stretch rate.

Convex and grooved solid fuels are chosen to examine the effect of complex geometry configurations on flammability, with both opposed and concurrent flame spread experiments conducted. Near the edge of the two configurations, the heat transfer from the flame to the fuel surface is increased and the local thermal inertia of the fuel is reduced. The flame spread rate along the edge is always the fastest in the opposed or concurrent flame spread. This result shows that the edge reflected the most dangerous position of the solid material. The opposed flame spread rate is accelerated across the fuel surface due to the presence of the edge and the flame spread rates eventually reach the same value. However, the flame will not reach a steady state in the concurrent flame spread and the flame spread rates are different across the fuel surface. Based on the flame spread theory of cylindrical fuel, a formula for estimating the flame spread rate over complex configuration fuels is established. The extinction experimental results and analysis show that the edge structure has no significant effect on the flammability limit. The thermal inertia of the solid phase dominates the flame extinction and the minimum thickness of the solid fuel determines its limiting oxygen concentration.

The scaling effects in the flammability test method, e.g. upward flame spread test, and the flame spread behavior near extinction limit are investigated under various oxygen concentration and atmospheric pressure conditions. Flame temperature, solid-phase temperature, surface temperature, and mass loss rate are measured. The distribution of heat flux and mass transfer number B are calculated. Different flammable limit conditions are obtained by changing the oxygen concentration and ambient pressure. The flame temperature and the flame radiation heat flux increase with the thickness or the width of fuel far from the extinction limit, which in turn increases the B number. The difference in the B number of the fuel with different widths is not obvious near the extinction limit but the B number increases with smaller thicknesses. The B number is greater when the flame is far from the extinction limit compared to when it is near the limit. The critical B number is firstly decreased and increased with oxygen concentration. The theory to estimate the B number far from the extinction limit is slightly larger than the experimental value because it does not take into account the influence of fuel lateral diffusion and heat loss. There are significant differences between theoretical calculations and experimental measurements to determine the B number, and the theory needs to be further improved to describe the flame behavior near the extinction limit.