Moving time domain blending operational strategy for alumina production by mixed integer optimization

The preparation of quality raw slurry is a complicated and crucial process in alumina production which determines the cost and efficiency of the entire production. In the mixing unit, some tanks with fully filled raw slurry are available to be selected in each round, and the components in these selected tanks are required to be mixed, meanwhile some ratios of chemical compositions in the final mixture must be strictly within the required level. This is the production procedure of quality raw slurry. To improve production efficiency and reduce operation time lags, this selection and proportioning problem is formulated as a mixed integer programming problem. An optimization algorithm developed based on the moving time domain is applied to find the optimal selection and the optimal mixing proportioning strategy such that the productivity of the quality raw slurry is maximized. Practical on-site comparison experiments are carried out using the proposed method and the method based on the experience of the operator. The results obtained show that the proposed approach is effective and has advantage in the preparation of quality raw slurry with high efficiency and maximal productivity.