Yield of C4-olefin based on partial least squares regression model

According to the empirical formula: C
4
-olefin yield equal to ethanol conversion multiply C
4
-olefin selectivity. It is known from reference [1] that in the process of preparing C
4
-olefin, temperature and catalyst combination will affect ethanol conversion and C
4
-olefin selectivity. Therefore, in order to make the C
4
-olefin yield as high as possible, choosing the appropriate temperature and catalyst combination seems especially significant. In this paper, the least square polynomial fitting model was established, with the temperature of each catalyst combination as the independent variable and ethanol conversion as the dependent variable, to study the influence of temperature on ethanol conversion when the catalyst combination is constant. The influence of temperature on C
4
-olefin selectivity can be studied in the same way. Then take temperature, ethanol concentration, loading ratio of Co/SiO
2
, loading ratio of Co/SiO
2
and HAP, and the total weight of loading as independent variables; ethanol conversion and C
4
-olefin selectivity were used as dependent variables. By taking logarithm of dependent variable, transform the multiple nonlinear regression into multiple linear regression. Establish the partial least squares regression(PLS) model to study the effects of different temperature and catalyst combination on ethanol conversion and C
4
-olefin selectivity. Then establish linear programming model to solve the maximum yield of C
4
-olefin temperature and catalyst combination.