摘要
针对连铸热轧一体化滚动计划问题,建立了动态约束满足模型,提出了滚动优化前后批量计划在浇次计划相似度和轧制计划相似度上的一致性度量方法。针对问题的模型及其特点,提出了基于动态约束满足的局部修复求解算法。算法通过基于域的板坯选择策略实现困难变量优先赋值,通过基于费用的板坯插入策略实现成功优先的值选择,在迭代过程中采取基于约束传播的初始变量域收缩策略和解空间更新策略提高搜索效率。基于企业实际生产数据的仿真实验结果,验证了模型和算法的有效性。
Aiming at the rolling plan problem of steel production continuous-casting hot-rolling,a dynamic constraint satifaction model was created.Besides,a consistency measurement method of batch plan before and after rolling optimization between cast plan similarity and rolling plan similarity was proposed for this problem.By studying the model and characteristic of the problem,a local repair algorithm based on dynamic constraint satisfaction was proposed.According to the domain-based slab selection strategy,the priority valuation of algorithm＇s difficult variables was realized;according to the cost-based slab insertion strategy,the priority selection of algorithm＇s successful value was exploited.To improve the algorithm＇s search efficiency,a Constraint Propagation Based（CPB） shrinkage strategy of the initial variables and a CPB update strategy of solution space were adopted during iterations.Validity of the model and algorithm were tested by simulation and calculation of the data from production practices.
出处
《计算机集成制造系统》
EI
CSCD
北大核心
2011年第10期2177-2184,共8页
Computer Integrated Manufacturing Systems
基金
宁夏医科大学特殊人才科研启动基金资助项目(XT2011003)
国家自然科学基金资助项目(70771008)
中央高校基本科研业务费专项资金资助项目(FRF-AS-09-007B)~~
关键词
钢铁生产
滚动计划
批量计划
动态约束满足
连铸热轧
steel production
rolling plan
batch plan
dynamic constraint satisfaction
continuous-casting hot-rolling
