炼钢-轧钢流程合同排程多目标变邻域搜索算法  

Multi-objective variable neighborhood search algorithm for order scheduling of steelmaking-rolling process

作　　者：徐端[1] 孙帅 刘士新[2] XU Duan;SUN Shuai;LIU Shixin(Informatization Business Department,Shanghai Baosight Software Co.,Ltd.,Shanghai 201900,China;College of Information Science and Engineering,Northeastern University,Shenyang 110819,China)

机构地区：[1]上海宝信软件股份有限公司信息化事业本部,上海201900 [2]东北大学信息科学与工程学院,辽宁沈阳110819

出　　处：《冶金自动化》2021年第1期1-7,共7页Metallurgical Industry Automation

基　　金：国家重点研发计划资助项目(2017YFB0304201)。

摘　　要：提高钢铁企业合同排程优化水平可以减少生产过程中品种及规格切换,稳定产品质量,降低能源消耗和生产成本。针对钢铁企业炼钢-轧钢流程合同排程优化问题,建立了多目标优化数学模型,设计了多目标变邻域搜索算法。针对问题特点,设计了3种邻域结构的局域搜索过程。通过实际案例对模型和算法进行了验证,并把算法嵌入到实际应用系统中。计算实验和实际应用效果表明模型和算法有效,可以求得分布较好的Pareto解集,便于决策者决策。Improving the optimization level of order scheduling in iron and steel enterprises can reduce variety and specification switches in the production processes,so as to stabilize product quality,reduce energy consumption and production cost. Aiming at the order scheduling optimization problem of steelmaking-rolling process in iron and steel enterprises,a multi-objective optimization mathematical model is established,and a multi-objective variable neighborhood search algorithm is designed.According to the characteristics of the problem,three kinds of local search procedures are designed based on different neighborhood structures. The model and algorithm are verified by a practical instance,and the algorithm is embedded into the practical application system. The computational experiments and practical application results show that the model and algorithm are effective and Pareto solution sets with good distribution can be obtained,which is convenient for decision makers.

关 键 词：炼钢-轧钢流程 合同排程 多目标 变邻域搜索算法 数学模型 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] F426.31[自动化与计算机技术—控制科学与工程]