CONSTRUCTION OF MULTIPLE REGRESSION EQUATIONS FOR STUDYING REGULARITIES OF CHEMICAL PROCESSES

Abstract

The article presents the results of a study of some regularities of the process of catalytic oxidation in the gas phase of methanol to formic acid. For this, a multiple regression equation for the process was created, the coefficients of the equation were calculated, and their significance was determined. The adequacy of the equation was checked and a more sensitive parameter was identified. Using the Fisher criterion, the statistical significance of the equation was proved. Pair correlation indicators of the coefficients of the equation are also calculated. It was also found that the model parameters are statistically significant. It has been established that the greatest influence on the yield of the target product - formic acid is exerted by changes in the temperature of the reaction medium. This technique for constructing a multiple regression model is universal and can be applied to other chemical-technological processes. Also checked the significance of the parameters of the multiple regression equation. The statistical significance of the regression coefficients is confirmed.