以计量化学潜因回归分析对有机合成作响应面优化的新方法(英文)

A Novel Procedure of Response Surface Optimization for Organic Synthesis by Latent Regression Analysis of Chemometrics

研究开发了CEDOS（Chemometrics—basedExperimentalDesignandOptimizationSystemforOrganicSynthesis，计量化学试验设计与优化有机合成系统）。它借助潜因回归算法如偏最小二乘（PLS）、主成份回归（PCR）等技术及试验设计与优化方法对有机合成进行响应面优化，获得良好结果。

A novel procedure for response surface optimization, CEDOS, was developed with and without experimental design. The system based on latent regression analysis. The latent modelling methods by partial least squares (PLS) and principal component regression (PCR) for the expanded matrix of independent variables (x) are used for transformation to the PLS and PCR instead of the causal moddeling method by multiple linear regression (MLR) loadings into polynomial coefficients and detection of the coefficients and detection of the coordinates of the best response within the experimental domain. The results are presented both graphically and numerically. The system is validated on an optimization study of the yield of an famous organic reaction , Diles - Alder reaction.