连铸机开浇炉次与时间决策的多目标优化

发布时间：2018-08-24 19:02

【摘要】：针对连铸机开浇决策中的炉次选择、排序与开浇时间确定的多目标优化问题,以炼钢厂生产批量计划执行情况的总惩罚、生产线积压金属量、优质铁水非有效利用量最小为目标函数,构建了连铸机开浇炉次与时间决策的多目标优化模型。针对该模型特点设计了改进的非支配排序遗传算法,以预选池内选择的炉次序号为基因的编码方式减小模型解的无效搜索空间,采取调整传统精英解集的计算顺序、限定计算拥挤距离个体数目的改进措施来减轻计算负荷,利用对Pareto解进行模糊选优的方法确定最终优化解。以某钢厂的生产实例数据测试表明,该模型有利于连铸生产炉次浇铸周期的稳定控制,算法效率优于传统的非支配排序遗传算法和强度Pareto进化算法。
[Abstract]:In view of the multi-objective optimization problems in the decision making of continuous casting machine, such as furnace selection, sequencing and casting time determination, the total penalty for the execution of batch production plan in steelmaking plant, and the backlog of metal in production line, are discussed. The minimum inefficient utilization of high quality hot metal is the objective function, and a multi-objective optimization model for the decision of the open casting furnace and the time of the continuous casting machine is established. According to the characteristics of the model, an improved genetic algorithm is designed, in which the sequence number of the furnace selected in the pre-selection pool is used as the coding method to reduce the invalid search space of the model solution, and the calculation order of the traditional elite solution set is adjusted. An improved method to limit the number of individuals to calculate the crowded distance is proposed to reduce the computational load, and the final optimal solution is determined by fuzzy selection of the Pareto solution. The test results of a steel plant show that the model is beneficial to the stable control of the casting cycle of the continuous casting plant, and the efficiency of the algorithm is superior to that of the traditional genetic algorithm for undominated ordering and the strength Pareto evolutionary algorithm.
【作者单位】：重庆大学材料科学与工程学院;攀枝花钢铁集团公司提钒炼钢厂;重庆大学经济与工商管理学院;
【基金】：国家自然科学基金资助项目(51474044) 重庆市科技攻关重点资助项目(CSTC2011AB3053)~~
【分类号】：TF777;TP18

【相似文献】