Abstract Due to the size and complexity of industrial-scale systems and computing limitations, semi-empirical soot models are often employed in CFD simulations rather than computationally-expensive detailed models. Many of these models were developed for specific applications with unique characteristic timescales and/or validated only under fuel-air combustion conditions. Hence, their use in different contexts, such as oxy-combustion, could lead to inaccurate predictions. In this study, twelve semi-empirical models (1-step or 2-step) are evaluated on their ability to respond to changes in stoichiometric mixture fraction (Zst) and strain in a series of ethylene counterflow flames which span the experimental sooting-to-non-sooting (yellow to blue) transition. A unique approach of plotting soot formation rate in normalized, local equivalence ratio space is introduced to aid in the analysis. Results show that no existing model is able to predict a blue flame when Zst is increased beyond the experimentally-measured sooting limit. Many models give the counter-intuitive result of increasing peak and/or integrated soot volume fraction (svf) as Zst increased, contrary to experimental observations. Two models are able to successfully predict a blue flame upon increase of strain to the sooting limit value. For two-step models, the strong dependence of growth rate on surface area results in high sensitivity of svf to changing flame boundary conditions, but with mixed accuracy. There remains a significant need for a robust semi-empirical model which can accurately predict soot fraction in systems with oxygen-enrichment or variable strain.

中文翻译：

---

评估具有增加的化学计量混合分数和应变的非预混火焰的半经验烟尘模型

摘要 由于工业规模系统的规模和复杂性以及计算限制，在 CFD 模拟中通常采用半经验烟尘模型，而不是计算成本高的详细模型。许多这些模型是为具有独特特征时间尺度的特定应用而开发的，和/或仅在燃料空气燃烧条件下进行验证。因此，它们在不同环境（例如氧燃烧）中的使用可能会导致预测不准确。在这项研究中，评估了 12 个半经验模型（1 步或 2 步）对一系列乙烯逆流火焰中化学计量混合分数 (Zst) 和应变变化的响应能力，这些火焰跨越了整个实验过程。 - 无油烟（黄色到蓝色）过渡。绘制标准化碳烟形成率的独特方法，引入局部当量比空间以帮助分析。结果表明，当 Zst 增加超过实验测量的烟尘限制时，没有现有模型能够预测蓝色火焰。许多模型给出了随着 Zst 增加而增加峰值和/或综合烟尘体积分数 (svf) 的违反直觉的结果，这与实验观察相反。两种模型能够成功预测应变增加到烟尘极限值时的蓝色火焰。对于两步模型，增长率对表面积的强烈依赖性导致 svf 对不断变化的火焰边界条件的高度敏感，但精度参差不齐。仍然非常需要一个强大的半经验模型，该模型可以准确预测富氧或可变应变系统中的烟尘分数。