摘要
针对炼钢生产组炉计划编制问题,建立多目标数学模型,并提出了LS-MOEA多目标优化算法。该算法首先求解板坯需要的充足炉次,并以此为基础,采用0-1染色体编码方法,每个染色体代表一个炉次选择方案,使用启发式方法求解染色体对应的组炉方案,并采用局部搜索扩大求解范围,迭代后获得一组较优的Pareto板坯—炉次匹配方案。经实际生产数据测试,利用该算法可以在合理的时间内给出一系列较优组炉方案,为计划员提供足够的决策支持。
A multiple-objective mathematical model and a multiple-objective optimization algorithm,denoted as LS-MOEA,are proposed for the charge design problem of steel-making. The algorithm first finds sufficient number of charges,which serves as the baseline for designing LS-MOEA with 0-1 chromosome encoding. Each chromosome represents a selection scheme of charges. Heuristic method is used to calculate the fitness value for chromosome and better solutions are searched with local search.The algorithm gives a set of Pareto solutions for slab-charge matching. Simulation on real production data indicates that the proposed algorithm can obtain a set of the optimized matching solution within reasonable time,and can provide enough decision support for planners.
出处
《宝钢技术》
CAS
2015年第1期61-65,共5页
Baosteel Technology
关键词
炼钢
组炉
PARETO
steel-making
charge design
Pareto
