改进的装炉组合问题建模与优化算法  

Improved Modelling and Optimal Algorithm for Combination Stacking

作　　者：王志刚[1] 刘全利[1] 王伟[1] 

机构地区：[1]大连理工大学信息与控制研究中心,辽宁大连116024

出　　处：《控制工程》2010年第2期197-201,204,共6页Control Engineering of China

基　　金：国家863高技术研究计划基金资助项目(2007AA04Z156)

摘　　要：针对罩式炉退火生产中的钢卷组合堆垛优化问题,建立了以最小化钢卷组炉总加热时间为目标的数学模型。模型综合考虑了钢卷自身属性以及生产工艺约束条件等因素对钢卷组炉加热处理时间的影响。在分析罩式炉退火加热工艺规范的基础上,提出了一种改进自适应遗传算法对模型求解。算法首先类比装炉组合问题与一维装箱问题的相似点分组编码染色体,借鉴装箱问题的优化思想改善初始解种群质量;然后在工艺规则的指导下对遗传基因进行启发式交叉和变异,变异率和交叉率随种群收敛程度自适应调整以保证种群多样性和全局收敛性;最后结合局部穷举搜索方法实现了对上述模型的优化计算。仿真对比实验以及现场实际应用效果均表明该算法相对其他算法的优越性。To the combinatorial optimization problem of stacking steel coils in bell-type batch annealing production process,a mathematical model is presented to minimize the total heating treatment time of batch stacking.The model thoroughly integrates the factors of affecting heating treatment time,such as characteristics of steel coils and production constraints.An improved adaptive genetic algorithm is proposed to calculate the model solution based on analysis of technical rules.Grouping encoded chromosome is adopted according to the similarity between combination stacking problem and one-dimension bin packing problem （BPP）.Optimization idea from BPP is utilized to improve initial population quality.Crossover and mutation operations are heuristic under the guidance of process rules.The crossover and mutation rate are adaptively tuned with convergence degree of population to guarantee diversity and convergence.A local search method is also added for enhancing performance.Simulation comparison and implementation result in a steel enterprise shows the superiority of the proposed algorithm.

关 键 词：罩式炉退火 组合优化 装箱问题 分组遗传算法 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]