基于机器学习的CH_(4)燃烧动力学机理优化  

Optimization of Kinetic Mechanism for Methane Combustion Based on Machine Learning

作　　者：曹双双 黄济勇 李伟[2] 张厚君 李象远[3] 韩优[1] CAO Shuangshuang;HUANG Jiyong;LI Wei;ZHANG Houjun;LI Xiangyuan;HAN You(School of Chemical Engineering and Technology,Tianjin University,Tianjin 300072,China;National Key Laboratory of Aerospace Chemical Power,Hubei Institute of Aerospace Chemotechnology,Xiangyang 441003,China;School of Chemical Engineering,Sichuan University,Chengdu 610065,China)

机构地区：[1]天津大学化工学院,天津300072 [2]航天化学能源全国重点实验室,湖北航天化学技术研究所,襄阳441003 [3]四川大学化学工程学院,成都610065

出　　处：《高等学校化学学报》2024年第10期69-77,共9页Chemical Journal of Chinese Universities

基　　金：航天化学能源全国重点实验室开放基金(批准号:STACPL220221B03);国家自然科学基金(批准号:T2441001,U20A20151)资助.

摘　　要：基于径向基函数插值算法构建的机器学习模型,以点火延迟时间(T=1084~2175 K,p=7.3×10^(4)~2.4×10^(6)Pa,φ=0.2~2.0)和层流火焰速度(T=293~600 K,p=5.1×10^(4)~1.1×10^(6)Pa,φ=0.4~2.0)实验数据为优化目标,对CH_(4)燃烧机理的指前因子和活化能进行优化,获得了可在宽工况范围内使用的CH_(4)燃烧机理.与初始的CH_(4)机理(Ori-CH_(4))相比,优化后的CH_(4)机理(Opt-CH_(4))在点火延迟时间上的预测平均误差下降了57.46%,在层流火焰速度上的预测平均误差下降了21.55%.使用Opt-CH_(4)机理对点火延迟时间、层流火焰速度和射流搅拌反应器中的组分浓度变化趋势进行了预测,Opt-CH_(4)机理均表现出优越的预测准确度.在T=1491.5 K,p=1.0×10^(5) Pa,4.988%CH_(4)\19.953%O_(2)\75.059%N_(2)(体积分数)工况下,CH_(3)+O_(2)■CH_(2)O+OH和CH_(2)O+O_(2)■HCO+HO_(2)在各个机理中的敏感性差异是优化前后CH_(4)机理预测准确度不同的主要原因.因此,机器学习方法在燃料燃烧反应动力学机理参数优化上具有广阔的应用前景.In this work,the experimental data of ignition delay time(T=1084—2175 K,p=7.3×10^(4)—2.4×10^(6)Pa,φ=0.2—2.0)and laminar flame speed(T=293—600 K,p=5.1×10^(4)—1.1×10^(6)Pa,φ=0.4—2.0)were taken as the optimization objectives based on the machine-learning model constructed by radial basis function interpolation method,and pre-exponential factors and activation energies of CH_(4)combustion mechanism were optimized,and a CH_(4)combustion mechanism that can be used in a wide range of working conditions was obtained.Compared with the Ori-CH_(4)mechanism,the mean error of the Opt-CH_(4)mechanism is reduced by 57.46%in the ignition delay times and 21.55%in the laminar flame speeds.The Opt-CH_(4)mechanism was used to predict the ignition delay times,laminar flame speeds and the variation tendency of species concentration in jet stirred reactor.The Opt-CH_(4)mechanism showed superior prediction accuracy.Under the conditions of T=1491.5 K,p=1.0×10^(5) Pa,4.988%CH_(4)\19.953%O_(2)\75.059%N_(2)(volume fraction),the difference of sensitivity of CH_(3)+O_(2)■CH_(2)O+OH and CH_(2)O+O_(2)■HCO+HO_(2)in each mechanism is the main reason for the difference of prediction accuracy of CH_(4)mechanism before and after optimization.Therefore,the machine learning method has a broad application prospect in the optimization of fuel combustion reaction kinetics mechanism parameters.

关 键 词：甲烷燃烧 机器学习 化学动力学 机理优化 

分 类 号：O643[理学—物理化学]