不锈钢炼钢计划排程问题的数学模型及遗传算法  

Mathematical model and genetic algorithm for solving steelmaking scheduling problems of stainless steel

作　　者：董翠连 廖德诚 DONG Cuilian;LIAO Decheng(MES Software Business Department,Shanghai Baosight Software Co.,Ltd.,Shanghai 201900,China;Production Management Section,Fujian Fuxin Special Steel Co.,Ltd.,Zhangzhou 363100,China)

机构地区：[1]上海宝信软件股份有限公司MES软件事业部,上海201900 [2]福建福欣特殊钢有限公司产销管理组,福建漳州363100

出　　处：《冶金自动化》2023年第4期35-42,共8页Metallurgical Industry Automation

摘　　要：炼钢计划编制是钢铁企业生产运作管理的核心任务之一,合理的炼钢计划有利于炼钢生产平稳有序进行,同时优化调控各订单的生产节奏。以国内某不锈钢生产线为背景,考虑修磨机组的生产能力以及保温罩、保温炉等会影响产品组合的配套设施,通过建立数学模型,设计遗传算法(genetic algorithm,GA)实现炼钢计划的排程优化,解决了不锈钢产线炼钢计划排程中由修磨瓶颈导致的高温/低温装炉钢种交替生产、产品规格由宽到窄变化、长浇浇次和短浇浇次交替生产等复杂约束下的排程问题。试验结果表明,本文方法实现了以修磨机组为瓶颈设备控制生产节奏的智能炼钢作业计划编制,确保铸机、保温罩、修磨、保温炉与下游热轧工序协同顺畅,有效提升了炼钢计划的编制效率,每次计划编制时间由2 h缩短到1 min,计划编制自动化率达到95%,系统自动编制的炼钢计划满足了现场使用要求。Steelmaking production planning is one of the key tasks of the operation management of iron and steel enterprises.A reasonable production planning is fundamental to the steady and orderly operation of steelmaking,and helps to optimize and control the production rhythm of each order.Based on a specific domestic stainless steel production line,c onsidering the production capacity of the grinding unit and the supporting facilities such as insulation covers and furnaces that affect the product mix,a n optimization method for steelmaking scheduling by use of genetic algorithm(GA)was proposed.The method solves the scheduling problems caused by grinding bottleneck,such as the alternate production of high temperature/low temperature charging steel grades,the change of product width in descending order,and the alternate production of long casts and short casts.Experiment results show that the method realizes the intelligent steelmaking operation planning which takes the grinding units as the bottleneck equipment to control the production rhythm,and ensures the smooth cooperation between the casting machine,insulation cover,grinding,insulation furnace and the downstream hot rolling processes.The efficiency of steelmaking planning is effectively improved,and the planning time of each plan is shortened from 2 h to 1 min.The rate of automation planning reaches 95%,and the steelmaking plans meet the production requirements.

关 键 词：不锈钢生产 炼钢排程 瓶颈工序 数学模型 遗传算法 自动排程系统 

分 类 号：TF764.1[冶金工程—钢铁冶金] TP18[自动化与计算机技术—控制理论与控制工程]