Multi-objective optimization of rolling schedules for tandem hot rolling based on opposition learning multi-objective genetic algorithm

An opposition learning multi-objective genetic algorithm based multi-objective optimization approach of rolling schedules for tandem hot rolling is proposed. According to actual rolling process, the power distribution and rolling energy consumption are selected as cost functions. Then the multi-objective model of rolling schedules is established. The opposition learning multi-objective genetic algorithm is applied to the model. Result shows that the proposed method decreases the values of two objective functions simultaneously compared to actual rolling schedules. Furthermore the proposed approach archives better Pareto solution set of rolling schedules with less running time compared with NSGA II. The validity of the proposed approach is confirmed.