Sparse particle methods for turbulent premixed combustion

新颖的燃烧器概念主要是为精益预混燃烧而设计的。然而，为了保证点火和运行安全，需要一定程度的局部燃料分层，从而导致燃料-空气混合的不均匀性。更高的吞吐量和燃烧室尺寸减小的额外要求可能会增加湍流水平，从而使主要的火焰状态从小火焰状态转移到薄火焰状态。目前的燃烧模型通常基于小火焰假设，目前的模型不能可靠地预测燃料分层和偏离小火焰状态。提出的研究将开发一种基于粒子的蒙特卡罗方法来模拟湍流预混火焰。每个蒙特卡罗（PDF）粒子表示流体元素的瞬时局部解。然后，这些瞬时解允许近似气相化学成分的联合概率密度函数（PDF），从而近似预混火焰的内部结构。借助大涡模拟，在欧拉框架下求解了速度场的守恒方程。分子和湍流扩散的建模提出了主要的挑战：将标准混合模型应用于预混燃烧会导致火焰前缘不受控制和非物理混合，从而阻碍火焰结构甚至燃烧速度的准确预测。提出的工作的新颖之处在于使用稀疏粒子方法计算气相组成PDF。传统的粒子方法需要每个LES细胞20到50个粒子，它们的数值解可能非常昂贵。当使用稀疏方法时，粒子数可以降低到每10个LES单元中小于1个粒子，并且可以显着减少模拟次数。这是通过在参考场上调节混合算子来实现的。正是这个参考场可以将扩散模拟为一个多步骤过程，它可以防止火焰前缘的混合，确保预热区颗粒温度的持续升高，并限制颗粒着火和热量释放到反应区。首先，简单的、统计上一维火焰的直接数值模拟（DNS）将有助于开发新的闭包。然后，DNS配置将扩展到包括燃料分层。在最后一步，新的封闭将应用于真实的实验室火焰，并将在现有实验数据的帮助下进行验证。

Novel burner concepts are primarily designed for lean premixed combustion. To guarantee ignition and operational safety, some local fuel stratification is needed, however, leading to inhomogeneities of the fuel-air mixture. The additional requirements of higher throughput and a size reduction of the combustion chambers may increase the turbulence levels such that the predominant flame regime shifts from the flamelet to the thin-flame regime. Current combustion models are usually based on the flamelet assumption, and no current model can reliably predict fuel stratification and deviations from the flamelet regime.The proposed study will develop a particle based Monte Carlo method for the modelling of turbulent premixed flames. Every Monte Carlo (PDF) particle represents an instantaneous, local solution of a fluid element. These instantaneous solutions then allow for the approximation of the joint probability density function (PDF) of the chemical composition of the gas phase and thus of the inner structure of the premixed flame. The conservation equations for the velocity field are solved in a Eulerian framework with the aid of large-eddy simulations (LES).The modelling of molecular and turbulent diffusion pose the major challenge: the application of standard mixing models to premixed combustion leads to uncontrolled and unphysical mixing across the flame front that hinder the accurate prediction of the flame structure and even of the burning velocity. The novelty of the proposed work is the use of a sparse particle method for the computation of the gas phase composition PDF. Conventional particle methods require between 20 and 50 particles per LES cell and their numerical solution can be extremely costly. The particle number can be lowered to less than 1 particle per 10 LES cells when using sparse methods, and simulation times can be significantly reduced. This is realized by conditioning of the mixing operator on a reference field. And it is this reference field that can be used to model diffusion as a multistep process that prevents mixing across the flame front and ensures a continuous increase of particle temperature in the preheat zone and limits particle ignition and heat release to the reaction zone.First, direct numerical simulations (DNS) of simple, statistically one-dimensional flames will aid the development of new closures. Then, the DNS configuration will be extended to include fuel stratification. In a final step, the new closures will be applied to real laboratory flames and will be validated with the aid of available experimental data.