论文部分内容阅读
对两个分支的平行反应,当两个分支的反应级数相同时两种产物的动力学谱线性相关;反之则与线性无关。基于这一原理,提出利用主成分分析法处理反应在线过程中测得的动力学-光谱二维数据矩阵,根据主成分数就能确定两分支的反应级数是否相同,并进一步推断反应机理是否相同。在此前提下,借助目标试验因子分析结合数值遗传算法可解析出动力学参数及有关组分的纯光谱。将上述方法应用于不同机理平行反应模拟数据的解析,获得的主成分数与已知反应机理相符,动力学参数和有关组分的纯光谱也与给定值吻合。
The parallel reaction between the two branches shows a linear correlation between the kinetic spectra of the two products when the reaction orders of the two branches are the same, and vice versa. Based on this principle, we proposed a two-dimensional matrix of kinetic-spectra measured by the principal component analysis (PCA) to deal with the reaction online. Based on the principal components, we can determine whether the reaction orders of the two branches are the same, and further infer whether the reaction mechanism is the same. Under this premise, kinetic parameters and the pure spectra of related components can be resolved by means of the target experimental factor analysis combined with numerical genetic algorithm. The above method is applied to the parallel reaction simulation data of different mechanisms. The principal components obtained are in accordance with the known reaction mechanism. The kinetic parameters and the pure spectra of the related components are also consistent with the given values.