溶鋼鍋引当計画最適化アルゴリズムの開発

抄録

The reduction of molten steel temperature from the tapping to the casting is crucial for energy conservation in the steel making process. The one of the reasons that the steel temperature reduction is the heat transfer from molten steel to ladle. After the end of casting, the ladle is empty and its temperature continues to drop until next assignment to the appropriate tapping which is scheduled in advance. Therefore, it is important to schedule the ladle assignment plan with taking account into shorten the ladle empty time. The goal of this study is development of the algorithm which can calculate the optimal plan satisfied the several complex operational constraints. We developed the algorithm based on the breadth-first search algorithm. It has features to search the wide range of solution space while limiting search space in order to search optimal solution efficiently. This algorithm is simplified to treat the change and addition of operational constraint flexibly. As the result of test using actual data, this algorithm could calculate the plans satisfied all the constraints while the representative optimization methods, which are using genetic algorithm and mixed integer liner programming, couldn’t. Furthermore, the result shows 20% reduction of the ladle empty time compared with the conventional result scheduled by operator.