Abstract: At present, the steel plant adopts the manual method to carry out the ladle scheduling, the optimization results of the ladles selection are not good enough because of the uncertainty of ladle temperature, ladle life and ladle materials. Ladle scheduling is to select and distribute the ladles based on the heat plan, providing that the process time of heats and the process requirement of the dephosphorization for heats are given. The selection of dephosphorization ladle takes the performance index of maximum temperature, highest life, and the remaining online ladle life. Take the ladle temperature in the specified range, the ladle life within the specified range and the ladle maintenance end time as early as the furnace processing time as the constraints to determine the dephosphorization. This paper adopts consistency classification method to obtain the dephosphorization ladle rule, and use the rule-based reasoning to get dephosphorization ladle. The performance is tested by simulation experiments. The experimental results show that, compared with the manual scheduling, the algorithm enables the production work well and improve the production efficiency, and improve the economic efficiency of enterprises.

Key words: Steelmaking-refining-continuous, ladle scheduling, ladle matching, rule-based reasoning

摘要： 针对炼钢厂采用人工方式进行钢包调度，在钢包温度、寿命、材质方面随意性较大，导致钢包选配结果优化程度偏低的问题，在炉次的加工时间已知、炉次对钢包的脱磷工艺要求已知的条件下，以钢包温度最高、钢包寿命最长、钢包剩余在线时间最大为性能指标，以钢包温度在规定范围内、钢包寿命在规定范围内和钢包维护结束时间早于炉次加工时间为约束，采用分类一致性方法给出了脱磷钢包选配规则，通过规则推理为进行脱磷的炉次决策脱磷钢包。采用仿真实验测试算法的性能，实验结果表明，相对于现场人工调度，算法能够使钢包周转更加流畅，提升生产效率，提高了企业的经济效益。

关键词: 炼钢-精炼-连铸, 钢包调度, 钢包选配, 规则推理