Optimized control strategy for concentrate copper grade in copper flotation process based on data mining and NSGA-II

The optimized strategy for the copper flotation industrial process is studied in this paper, a key constituent in the mineral processing concentrator plant. To predict the concentrate copper grade accurately in the copper flotation industrial process, a nonlinear compensation strategy is adopted to eliminate the characteristics of nonlinear, strong coupling and time-delay that the copper flotation process can be approximated as a linear process. To discover valuable operational experience and rules, it could be meaningful to apply data mining to the data collected from field data of flotation industry. Moreover, knowledge based non-dominated sorting genetic algorithm (KNSGA-II) is proposed that it is more efficient when incorporating mining knowledge. The proposed method is considered desirable through the copper flotation industrial process.