An improved 3-pentanone high temperature kinetic model using Bayesian optimization algorithm based on ignition delay times, flame speeds and species profiles
An improved 3-pentanone high temperature kinetic model using Bayesian optimization algorithm based on ignition delay times, flame speeds and species profiles
复制标题
使用基于点火延迟时间、火焰速度和物质分布的贝叶斯优化算法改进的 3-戊酮高温动力学模型
DOI：
10.1016/j.fuel.2020.118540
复制
发表时间：
2020-11
期刊：
影响因子：
7.4
通讯作者：
Haiyang Shi
中科院分区：
文献类型：
--
作者：
Qianjin Lin;Junmei Zheng;Chun Zou （通讯作者）;Jia Cheng;Jiarui Li;Wenxiang Xia;Haiyang Shi
摘要在Sun等人提出的3-戊酮（二乙基酮，DEK）高温反应动力学模型（HUST_Model）的基础上进行了改进。(Sun_模型）在本研究中。主要更新的反应如下：由于分子结构的相似性，通过与丁酮类比，得到了DEK与H O ~ 2、O ~ H和O ~ 2的夺氢反应的反应速率常数; DEK的两个分解反应的速率常数由Dames等人的工作得到。（Dames_Model）。此外，DEK的点火延迟时间（IDT）在2.65巴和10巴的压力下测量;当量比为0.5和2.0;温度范围为1100-1500 K。利用文献中大量的实验数据，包括IDT、火焰速度和物种分布，以及本研究所测得的IDT，采用贝叶斯优化算法对H-O-2、O-至今H、O2和DEK分解反应的DEK氢原子提取反应的速率系数进行了优化. IDT和火焰速度的实验数据用于HUST_Model的优化，物种分布的实验数据用于优化的HUST_Model的验证。在优化过程中，这四对反应的分支比保持不变。结果表明，HUST_Model不仅对IDT和火焰速度的预测性能优于Dames_Model和Sun_Model，而且对组分分布的预测性能也优于Dames_Model和Sun_Model。通过灵敏度分析和反应通量分析，比较了HUST_模型与Dames_模型和Sun_模型对IDT、火焰速度和CH 3浓度分布的预测结果。结果表明，BOA法对HUST_模型中DEK的H O ~ 2、O ~ H、O ~ 2抽氢反应和DEK分解反应的反应速率常数进行全局优化是合理可靠的。
Abstract A 3-pentanone (diethyl ketone, DEK) high temperature kinetic model (HUST_Model) was improved based on the model proposed by Sun et al.(Sun_Model) in the present study. The main updated reactions are as following: the reaction rate constants of the DEK H-atom abstraction reactions by H O ̇ 2, O ̇ H and O 2 were obtained by analogy with butanone due to the similarities in the molecular structure, and the rate constants of two DEK decomposition reactions were obtained from the works of Dames et al.(Dames_Model). Moreover, the ignition delay times (IDTs) of DEK were measured at pressures of 2.65 bar and 10 bar; equivalence ratios (ϕ) of 0.5 and 2.0; and temperatures in the range of 1100–1500 K. The rate coefficients of DEK H-atom abstraction reactions by H O ̇ 2, O ̇ H, O 2 and DEK decomposition reactions can be optimized by Bayesian Optimization algorithm using a very large number of experimental data from the literatures, including IDTs, flame speeds and species profiles, and IDTs measured in the present study. The experimental data of the IDTs and the flame speeds were used for the optimization of HUST_Model, and those of the species profiles for the validation of the optimized HUST_Model. The branching ratios of these four pairs of reactions were maintained unchanged during the optimization process. The results of the evaluation and validation show that the performance of HUST_Model is much better than Dames_Model and Sun_Model not only for the prediction of the IDTs and the flame speeds, but also for the prediction of the species profiles, especially the CH 3 profiles. The comparisons of HUST_Model with Dames_Model and Sun_Model were conducted by the sensitivity analysis and reaction flux analysis for the predictions of the IDTs, the flame speeds and the CH 3 profiles. The results of the comparison and analysis show that, the global optimization of the reaction rate constants of DEK H-atom abstraction reactions by H O ̇ 2, O ̇ H, O 2 and DEK decomposition reactions in HUST_Model by the BOA is reasonable and reliable.
登录
查看更多内容
DOI：
10.1039/b605650b
发表时间：
2007-06
期刊：
--
影响因子：
--
作者：
P. Seakins
通讯作者：
P. Seakins
影响因子：
3.4
作者：
Weiqi Ji;Zhuyin Ren;C. Law
通讯作者：
Weiqi Ji;Zhuyin Ren;C. Law
DOI：
10.1021/jp107167f
发表时间：
2010-10
期刊：
The journal of physical chemistry. A
影响因子：
--
作者：
Z. Serinyel;N. Chaumeix;G. Black;J. Simmie;H. J. Curran
通讯作者：
Z. Serinyel;N. Chaumeix;G. Black;J. Simmie;H. J. Curran
影响因子：
10.4
作者：
An, H.;Yang, W. M.;Chua, K. J.
通讯作者：
Chua, K. J.
影响因子：
4.3
作者：
G. Buxton;C. Greenstock;W. P. Helman;A. Ross;W. Tsang
通讯作者：
G. Buxton;C. Greenstock;W. P. Helman;A. Ross;W. Tsang