Application of multiple linear regression in cement burden

In the process of cement burden, as the changing composition of the four raw materials, the oxide content in the raw meal product fluctuates. When the three-rate value is unqualified, the manual cannot judge the adjustment type and quantity of raw materials. This process has the disadvantage of large amount of calculation and difficult to accurately adjust the ratio of raw materials. Finally, the quality of raw materials fluctuates greatly and the qualified rate is low. So as to improve the pass rate of raw cement, this paper explored the optimal allocation of raw materials, which obtained the oxide data by near-infrared online analyzer, using segmented multiple linear regression to obtain regression parameters. Through the target value of four oxides of raw materials, the regression coefficient is used to solve the adjustment scheme of raw materials. Finally we use a variant of NSGA-II to solve the optimal blending scheme of raw materials.