基于微粒群和满足质量约束的组炉方案优化方法

Optimization of charge design with quality constraints based on particle swarm optimization

中文关键词  组炉计划  质量设计  聚类分析  微粒群优化

英文关键词  charge design  quality design  clustering analysis  particle swarm optimization

中文摘要

      分析了面向订单生产的钢铁企业面临的市场需求与生产组织特点, 将炼钢组炉方案的优化设计归结为一个满足化学成份等质量因素约束的聚类分析问题. 在此基础上提出了基于微粒群优化的求解方法, 该方法利用主成分分析技术缩小了问题域的维度, 在传统的工艺约束、交货期约束和炉容量等约束的基础上, 引入质量相近产品成份取值范围约束来限定微粒的活动范围, 采用炼钢组炉计划与质量设计的集成模式, 在多约束下对成份相近的不同品种的候选组炉合同进行聚类分析, 实现了面向钢铁产品多品种小批量需求的满足质量约束的组炉方案的优化.

英文摘要

      To deal with the market demand and the production organization feature of steel making enterprises, we formulate the optimization of charge design as a cluster analysis problem with quality constraints on chemical compositions. A particle-swarm-optimization-based(PSO-based) solution is proposed for reducing the dimensions based on the principal component analysis(PCA) techniques. The range constraints of chemical compositions for products with similar quality are introduced in terms of traditional process constraints, due time constraints and furnace capacity constraints, etc. The solution adopts an integration schema for charge plan and quality design. It performs the cluster analysis for candidate products with similar chemical compositions and constraints to realize the optimal charge design under quality constraints on steel products of multiple varieties and in small batch demands.