论文部分内容阅读
杜利特尔算法作为1种高精密度低计算量的数值分析法,在数学等诸多领域得到了广泛的应用,但在分析化学方面的应用则不多见。这种算法通过三角分解矩阵、降阶,简化矩阵运算而实现,与传统多元校正法相比具有更高的灵敏度。本文首次将杜利特尔法引入分析化学中的条件优化,以影响色谱分离度的3个因素为例,均匀实验设计癸烷和十二烷烃体系分离,得到分离度范围为50.124 2～98.694 2的不同实验体系,而后分别运用杜利特尔法和经典的最小二乘法建立的回归方程拟合实验结果,结果表明运用杜利特尔法所得出回归方程的拟合误差仅为-2.50%～2.61%,明显优于最小二乘法的结果(-24.67%～-57.47%)。说明杜利特尔多元校正法运用于条件优化方面比较理想,所拟合出的回归方程可以直接预测和推算实验条件的优劣性,在分析化学中具有广泛的应用前景。
As a numerical analysis method with high precision and low computational complexity, Doolittle’s algorithm has been widely used in many fields such as mathematics, but its application in analytical chemistry is rare. This algorithm is realized by triangulating the matrix, reducing the order and simplifying the matrix operation, and has higher sensitivity compared with the traditional multivariate calibration method. This paper, for the first time, introduced the Doolittle method into the analytical conditions of the analytical chemistry. Taking the three factors affecting the chromatographic resolution as an example, the experimental design of decane and dodecane system was separated experimentally and the separation range was 50.124 2 ~ 98.694 2 The experimental results were fitted by the regression equations established by Doolittle method and classical least squares method. The results show that the fitting error of the regression equation obtained by Doolittle method is only -2.50% ~ 2.61%, which is obviously better than the least squares method (-24.67% -57.47%). It shows that Dolerian multivariate calibration method is ideal for conditional optimization. The fitted regression equation can directly predict and calculate the advantages and disadvantages of experimental conditions and has a wide range of application prospects in analytical chemistry.