发展一个具有宽范围预测能力的燃烧反应动力学机理是预测燃料燃烧特性、实现能源清洁高效利用的前提，而小分子核心机理是实际燃料燃烧机理中不可或缺的重要组成部分。针对核心机理的发展现状以及对于宽温度和压力范围动力学参数的需求，本工作采用理论计算方法对核心机理C4组分的燃烧反应动力学机理进行了系统性的研究。 首先对丁烯自由基异构体(C4H7)的势能面进行了高精度量子化学计算并运用动力学方法计算了速率常数。之后将计算结果用于模型更新，并基于更新后的机理进行了点火延迟时间、层流火焰燃烧速度和重要中间组分浓度的模拟。通过路径分析和灵敏性分析总结归纳了不同的骨架结构对丁烯异构体燃烧活性的控制机制，基于模型对比揭示了温度和压力依赖效应在不同工况下对模型预测能力的影响。 其次研究了1,3-丁二烯及其异构体的异构和解离机制，结合量化计算和动力学计算探究了四种C4H6异构体的主要分解路径，并结合模型分析对前人热解实验中的观测现象进行了解释。研究发现，模型中遗漏的“well-skipping”解离反应是实验对1,3-丁二烯初始消耗机制的描述出现矛盾的原因。更新后的模型能较好地解释和预测宽范围下的实验现象。 之后对核心机理C4组分中对苯环生成具有重要贡献的共振稳定 的C4H5异构体的动力学进行了研究，通过计算新的光电离截面数据结合前人光电离质谱实验量化了C4H5异构体在碳氢燃料火焰中的组成。同时运用RRKM/主方程方法计算了C4H5异构体的异构和解离速率常数并进行了动力学模拟，给出了碳氢燃料火焰中C4H5异构体的比例。进一步计算了2- C4H5和12-C4H5与乙炔反应的温度压力依赖速率常数，探索了上述共振稳定C4H5异构体在火焰中对苯环生成的贡献。 最后，基于一个典型的C4体系势能面，结合人工神经网络-高维模型表征方法(ANN-HDMR)对RRKM/主方程模拟过程中输入参数的不确定度传递机制进行了探索，揭示了典型小分子反应温度压力依赖速率常数及其分支比的不确定性控制因素，进一步总结不同类型反应的不确定度随温度和压力变化的规律。基于上述研究评估了C4组分燃烧动力学参数的不确定度，提供了核心机理重要C4组分的温度和压力依赖的速率常数及其不确定性信息。

The development of a combustion reaction kinetic mechanism that able to accurately predict the combustion property of fuels in a wide range is the premise of achieving clean and efficient energy utilization. And the core mechanism which is also referred to as the foundational fuel chemistry is an indispensable and important component of the combustion mechanism of practical fuels. Considering the development status of the core mechanism and the demand for the kinetic parameters at a wide temperature and pressure range, this work systematically studied the reaction kinetics of the C4 component of the core mechanism by theoretical methods. The potential energy surface of the isomeric butenyl radicals (C4H7) was calculated with high precision and the RRKM/master equation method was used to calculate the rate constant. Then the new calculation results were used for model development. The ignition delay time, laminar flame speed and the mole fractions of important intermediates are simulated by the updated model. Simulation results were further explained by rate of production (ROP) analysis and sensitivity analysis. The controlling kinetic mechanism of different skeleton structures on the combustion reactivity of butene isomers is summarized. In addition, the effect of pressure-dependent kinetics on the prediction ability of the model under different operating conditions is revealed based on the model comparison. The isomerization and dissociation mechanism of 1,3-butadiene and its isomers were studied. The main decomposition pathways of four C4H6 isomers were investigated by quantum chemistry calculation and kinetic theory. The experimental observation in previous studies on the initiation mechanism of 1,3-butadiene in pyrolysis and flame conditions were explored. The present study found that the "well-skipping" reaction that was missing in the core models is the reason for the contradictory observation of different experiments in describing the initial consumption pathway of 1,3-butadiene. The updated model can better explain and predict experimental phenomena over a wide temperature and pressure range. Then the kinetics of the resonance-stabilized C4H5 isomers - important C4 components of the core mechanism which have a large contribution to the formation of benzene was studied. The new photoionization cross section data were combined with previous photoionization mass spectrometry experiments to quantify the composition of the C4H5 isomers in hydrocarbon flames. At the same time, the isomerization and dissociation rate constants of C4H5 isomers were calculated by RRKM/master equation method and the kinetic simulation was conducted. The ratio of C4H5 isomers in hydrocarbon fuel flame was deduced from both experimental data and modeling results. The temperature-dependent rate constants of the reaction of 2-C4H5 and 12-C4H5 with acetylene were further calculated, and the contribution of the above-mentioned resonance-stabilized C4H5 isomer to the formation of benzene in the flame was revealed. Based on a typical potential energy surface of a C4 reaction system, the artificial neural network-high dimensional model representation method (ANN-HDMR) is used to explore the uncertainty propagation of input parameters in RRKM/main equation simulation process. The controlling factors that determine the uncertainty of rate constants and their branching ratio were revealed. Based on the above studies, the uncertainty of the combustion kinetic parameters of the C4 component was evaluated, and a C4 sub-mechanism containing temperature and pressure-dependent parameters with uncertain information was provided.