采用FGM方法模拟钝体稳定火焰中NOx的生成

Modeling of Formation NOx in Bluff-Body Stabilized Flame Using FGM

关键词:  火焰面生成流形  钝体稳定火焰  辐射  光学薄模型

Key words:  Flamelet generated manifolds  Bluff-body stabilized flame  Radiation  Optically thin model

摘要:

为研究燃烧过程中NO预测精度的影响因素，采用火焰面生成流形(FGM)方法对Sydney大学CH4/H2钝体稳定火焰HM1中NO生成的预测模型进行研究。分别采用GRI 2.11和GRI 3.0甲烷反应机理构建了绝热和非绝热FGM查询表，辐射模型采用光学薄模型(OTM)。NO的预测是通过独立求解NO质量分数的输运方程实现，其反应速率由FGM查询表查表得到，包括两种处理方式:一种是直接采用FGM查询表中的反应速率；另一种是采用输运方程求解的NO质量分数替换逆反应速率中的NO质量分数。此外还研究了扩展反应进度变量对NO的预测精度的影响。对于流动和主要的热力化学参数，非绝热FGM方法和GRI 3.0可获得更高的精度；求解NO质量分数输运方程可有效提高NO的预测精度，而且第二种NO反应速率的处理方式得到的NO质量分数与试验吻合最好；考虑辐射提高了NO预测精度，并且采用GRI 3.0得到的NO预测结果与试验吻合最好；扩展反应进度变量对由FGM查询表和第一种NO反应速率处理方式得到的NO预测精度在加权因子达到1000时有较大的提升，而对第二种NO反应速率处理方式的结果没有明显的影响。非绝热FGM方法结合第二种NO反应速率处理方式求解NO质量分数的输运方程可以实现高精度的NO预测，此外化学反应机理对NO的预测精度有着非常重要的影响。

Abstract:

In order to study the factors affecting the accuracy of NO prediction in combustion， the NO prediction model was studied using Flamelet Generated Manifolds(FGM) in Sydney University CH4/H2 bluff-body stabilized flame HM1. Two methane chemical mechanisms of GRI 2.11 and GRI 3.0 were used to construct adiabatic and non-adiabatic FGM lookup table respectively， and Optically Thin Model (OTM) was utilized for radiation simulation. NO prediction was accomplished by solving the transport equation of NO mass fraction independently， the reaction rate of which was retrieved from FGM lookup table， including two methods: one was the reaction rate from FGM lookup table directly; the other was the NO mass fraction in the negative reaction rate replaced by that resulted from the transport equation of NO mass fraction. In addition the effects of the extended reaction progress variable on the accuracy of NO prediction were investigated. As for the flow and major thermo-chemical parameters， non-adiabatic FGM and GRI 3.0 chemical mechanism has higher accuracy. Solving the transport equation of NO can effectively improve the accuracy of NO prediction， and moreover the secondary method of dealing the reaction rate of NO has a best agreement with the experimental data. Inclusion of radiation improves the accuracy of NO prediction， and the best agreement with experimental data is obtained by GRI 3.0. Inclusion of NO mass fraction in the definition of reaction progress variable can significantly improve the accuracy of NO prediction resulted from FGM lookup and the first method of dealing the reaction rate of NO when weighted factor reaches 1000， but has no notable effect on the result from the second method of dealing the reaction rate of NO. High accuracy of NO prediction can be accomplished by the combination of non-adiabatic FGM and solving the transport equation of NO mass fraction with the secondary method of dealing the reaction rate of NO， and furthermore the chemical mechanism has an important effect on the accuracy of NO prediction.