| 应用扩展贝叶斯进化算法求解混流装配调度问题 | |
| 羌磊 ; 肖田元 ; QIANG Lei ; XIAO Tian-yuan | |
| 2010-06-09 ; 2010-06-09 | |
| 关键词 | 优化调度 混流装配 统计优化方法 贝叶斯网络 optimal scheduling hybrid assembly statistic optimization Bayesian network O223 |
| 其他题名 | Model extended BOA to solve hybrid assembly scheduling problems |
| 中文摘要 | 为求解复杂混流装配线调度的问题,提出一种基于贝叶斯进化算法的优化方法,给出了基于同机工序和工艺相关工序关联分析的贝叶斯进化算法求解框架。在贝叶斯进化算法基本建模方法的基础上,引入有效解模式表征指标,以增强对有效解变量取值关系的表征和进化能力。同时,引入模式扩展关联机制,将变量层次的关联关系拓展到变量取值层次,进一步提高算法的进化搜索效率。最后,给出了算法实现过程中的小概率解模式保留策略,以避免优化信息的缺失。通过对混流装配调度算例及标准benchmark算例的仿真验证表明,在较大规模的问题求解中,本文算法与遗传算法和一般贝叶斯进化算法相比,优化效率得到了明显的提高。; A Model Extended Bayesian Optimization Algorithm(ME-BOA) was presented to solve hybrid assembly scheduling problems.The proposed algorithm firstly built an optimization schema based on relativity analysis of tasks with constraint on each other.Then an efficient solution pattern represent index was introduced to enhance the algorithm's ability of describing and reserving good solution patterns. And the extended Bayesian modeling strategy was then provided to enhance the algorithm's optimization efficiency.It could describe the dependency relationships among decision variable's particular values other than whole value sets.Finally,the algorithm adopted a pattern reserving mechanism to avoid the information losing of patterns with low probability.Scalable experiment results validated the efficiency of the new algorithm by comparing it with typical BOA and genetic algorithm.; 国家教育振兴计划基金资助项目(081100601)。~~ |
| 语种 | 中文 ; 中文 |
| 内容类型 | 期刊论文 |
| 源URL | [http://hdl.handle.net/123456789/57387]  |
| 专题 | 清华大学 |
| 推荐引用方式 GB/T 7714 | 羌磊,肖田元,QIANG Lei,等. 应用扩展贝叶斯进化算法求解混流装配调度问题[J],2010, 2010. |
| APA | 羌磊,肖田元,QIANG Lei,&XIAO Tian-yuan.(2010).应用扩展贝叶斯进化算法求解混流装配调度问题.. |
| MLA | 羌磊,et al."应用扩展贝叶斯进化算法求解混流装配调度问题".(2010). |
| 个性服务 |
| 查看访问统计 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。
修改评论