【作者】 安江涛；

【导师】 蒋勇；

【作者基本信息】 中国科学技术大学 ， 安全科学与工程， 2013， 博士

【摘要】 实际火灾是湍流燃烧过程,而在湍流燃烧中湍流流动与复杂化学是时空多尺度耦合作用的。研究湍流与复杂化学的相互作用是一项既有挑战也十分有意义的课题。湍流的空间尺度可以分为含能尺度、惯性子区尺度和耗散尺度三个层次。湍流涡旋从最小的耗散尺度开始就与火焰有着强烈的相互作用。数值模拟研究湍流与复杂化学的相互作用需要解决耦合详细化学反应机理的耗散尺度湍流反应流模拟,和海量模拟结果诊断分析的两个难题。本文采用一维湍流模型(one-dimensional turbulence, ODT)和化学反应爆炸模式分析方法(chemical explosive mode analysis, CEMA)解决了这两个难题,建立了ODT+CEMA分析湍流与复杂化学相互作用的数值分析平台,并对典型灭火剂与氢气射流火焰的相互作用进行了数值模拟和深入分析。复杂化学研究中,详细化学反应机理的分析及简化对实现湍流耦合复杂化学的数值模拟十分重要。本文中采用关系图法对典型碳氢燃料反应机理进行了简化研究。不但研究了在保持一定计算精度情况下简化机理能达到的最小规模,还深入分析了方法参数对简化结果的影响,并且引入强关系组分群的概念,分析了复杂化学反应机理中强关系组分群的聚集、分布情况。为了加快反应机理的计算,本文还做了利用图形显卡并行计算求解大型化学反应机理的研究,在大型机理情况下计算加速效果明显。一维湍流模型ODT,可以实现耦合复杂化学的耗散尺度湍流反应流模拟。本文介绍了ODT模型的计算框架、湍流模拟机制和数值求解方法。并用此方法模拟分析了氢气射流火焰,以及添加典型灭火剂后的射流火焰基本特性。高精度的湍流反应流模拟会产生海量的计算数据,传统采用典型系统参数如温度、关键组分的浓度,甚至一阶导数的分析方法已经不适合处理如此庞大、精细的计算结果。化学反应爆炸模式分析方法是一种分析湍流反应流局部系统特征值的方法。本文系统介绍了化学反应爆炸模式分析方法的数学基础和基于ODT模拟数据的湍流反应流信息分析方法。通过对添加灭火剂的氢气空气预混热自燃模型的化学反应爆炸模式分析发现,系统反应进行最剧烈的时候也是最大正特征值发生明显变化的时刻。添加了灭火剂后,热自燃时间不但明显推迟,其反应进程和最大正特征值的突变性更加明显。化学反应爆炸模式分析可以直观的观察到湍流反应流局部系统状态的变化,对诊断湍流火焰中的热自燃、熄火、重燃等行为十分有效。局部Damkohler数是控制火焰结构,研究湍流与复杂化学相互作用的重要参数。在本文分析中重新定义了Damkohler数。通过分析湍流火焰在新Da数空间的散点分布发现：新定义的Da数是判定局部熄火的良好标准,而灭火剂的作用途径主要是通过增大局部反应系统的热自燃时间,进而减小Da数,增加熄火概率。通过分析,本文还解释了Re数增加时灭火剂灭火效果增加的原因。研究结果明确指出了主要组分和主要基元反应在射流火焰中的作用区域,以及在火焰抑制效果不同时,火焰中心区域主要作用组分和基元反应的差异。本文的主要创新点和贡献在于：为研究湍流与复杂化学相互作用,建立了ODT+CEMA的数值分析平台,提供了一种分析湍流与复杂化学相互作用的新方法。更多还原

【Abstract】 The real fire is turbulent combustion process. The turbulence interacts with complex chemistry on multi-scale space. The research about interactions between turbulence and complex chemistry is a task of great importance and challenge. The spatial scales of turbulence can be divided into energetic scale, inertial sub-range scale and dissipation scale. The strong interaction between the turbulence and complex chemistry starts from the smallest scale of the turbulence, the dissipation scale.There are two important issues to deal with to discuss the interactions between turbulence and complex chemistry:the high accuracy simulation of turbulent flame coupled with detailed chemical reaction mechanism and the challenge of systematically and comprehensively extracting salient information from the massive output using rigorous computational utilities. In this paper, one-dimensional turbulence model and chemical explosive mode analysis, CEMA, were used to solve these two problems. The new computation platform built in this paper is useful to discuss the interactions between turbulence and complex chemistry. Simulations of the typical hydrogen jet flame and the effect of extinguishing agent is made to introduce the validation of the new computation platform.The analysis and reduction of the detailed reaction mechanism is very important for the rigorous turbulence combustion simulation. The robustness and usefulness of the two mechanism reduction methods, direct relation graph method and directed relation graph with error propagation method, was analyzed in this paper. A new concept of "strong ties species group" was introduced in this paper. The analysis of the numbers of the strong ties species group clearly showed the aggregation behavior of the species in the detailed mechanism. In this paper, the research about the parallel computing for detailed mechanism with the Graphics Processing Unit (GPU) was also introduced.One dimensional turbulence model, or ODT for short, can achieve dissipation scale turbulence simulation coupled with detailed chemical mechanism. In this paper, the computing framework, turbulence mechanism and numerical solution method of the ODT model were introduced.Scientific method is needed to process the huge datasets that was produced by the turbulence simulation with detailed chemical kinetic mechanisms. The chemical explosive mode analysis (CEMA) is a systematical method to detect ignition, extinction, premixed flame fronts and the control species and reactions in the flame. The math theory and application methods coupled with ODT model was illustrated in this paper.The CEMA of auto-ignition and ODT simulations with fire extinguishment agent showed that the CEMA is a useful method to identify the change point of auto-ignition and extinction. The results showed that the agents can significantly delay the ignition time and reduce the chemical explosive mode. The fire extinguishment agent can lower the Da number of the local system. When the Da number is lower than1.0, the flame will extinguish.The CEMA of the hydrogen jet flame clearly showed the major control variables and reactions in different acting areas.The main contribution and innovation can be summarized as follows:a new numerical analysis platform named ODT+CEMA was built to analyze the interactions between turbulence and complex chemistry.更多还原