A coke oven gas (COG ) mechanism w as built and optimized based on sensitivity analysis and particle swarm optimization (PSO) in order to conduct further simulation research on the vehicle engine fueled with COG . Given the hierarchical structure and overall interrelationships between oxidation mechanisms for simple hydrocarbon fuels ,a COG mechanism (containing 27 species and 102 reactions) was constructed .Then the mechanism was analyzed by the models of ignition delay time sensitivity and laminar flame speed sensitivity ,which were established by using CHEMKIN code .A mechanism optimization model was built and the kinetic parameters of key reactions were optimized based on the PSO algorithm . The ignition delay time ,laminar flame speed ,in‐cylinder combustion pressure and NOx emission were simulated with COG mechanism .Results were compared with the relevant experimental data .Results show that COG mechanism has better agreement with the experimental data than GRI‐Mech 3 .0 under some conditions ,and the mechanism can accurately simulate the combustion process and NOx formation in COG engine .%为了开展焦炉气在车用发动机上的数值仿真研究，基于敏感性分析与粒子群寻优算法，搭建并优化焦炉气化学反应机理．根据简单碳氢燃料氧化机理的分级结构与总体关系，初步搭建焦炉气化学反应机理（27个组分，102步反应）．利用CHEMKIN软件搭建的滞燃期与层流火焰速度敏感性分析模型，对焦炉气化学反应机理进行滞燃期与层流火焰速度的敏感性分析．根据抑制局部最优的粒子群寻优算法，搭建化学反应动力学参数优化模型，对关键化学反应的动力学参数进行优化．根据相关的实验数据，对优化后机理的滞燃期、层流火焰速度、缸内压力与NOx 排放量进行对比验证．结果表明，所得的焦炉气化学反应机理较GRI‐Mech 3．0能够更加准确地预测滞燃期与层流火焰速度，且能够准确模拟焦炉气发动机缸内燃烧与NOx 生成过程．
展开▼
