引用本文:李中华,谭洪舟,张雨浓,毛宗源.基于免疫算法的午饭时期层际高峰交通电梯群控制的动态优化[J].控制理论与应用,2007,24(2):177~182.[点击复制]
LI Zhong-hua,TAN Hong-zhou,ZHANG Yu-nong,MAO Zong-yuan.Dynamic optimization of elevator group control based on artificial immune algorithm for inter-floor peak traffic during lunch-time[J].Control Theory and Technology,2007,24(2):177~182.[点击复制]
基于免疫算法的午饭时期层际高峰交通电梯群控制的动态优化
Dynamic optimization of elevator group control based on artificial immune algorithm for inter-floor peak traffic during lunch-time
摘要点击 1430  全文点击 896  投稿时间：2004-10-15  修订日期：2006-10-08
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/j.issn.1000-8152.2007.2.003
  2007,24(2):177-182
中文关键词  电梯群控制  人工免疫算法  层际高峰交通  动态优化
英文关键词  elevator group control  artificial immune system  inter-floor peak traffic  dynamic optimization
基金项目  国家自然科学基金资助项目(60575006,60636004);广州市科技局科技攻关项目(108B204017);中山大学青年教师基金资助项目.
作者单位
李中华,谭洪舟,张雨浓,毛宗源 中山大学电子与通信工程系, 广东广州 510275
华南理工大学自动化科学与工程学院, 广东广州 510640 
中文摘要
      层际高峰交通是上行高峰、下行高峰与层间交通客流的融合,午饭交通就是发生在现代办公楼内的层际高峰交通之一.基于午饭时期层际高峰交通的特点,本文提出了一种利用人工免疫算法实现电梯群控制动态优化的策略.该策略采用两级优化控制, 利用常规的THV算法对层站召唤分配实施局部优化,采用人工免疫算法定时对层站召唤分配实施全局优化. 同单纯的THV算法相比,本文提出的组合算法将平均的系统等待时间降低了21.26%.仿真结果表明, 本文提出的优化控制策略能够改善午饭时期电梯系统的服务性能, 具有较好的现实意义和研究价值.
英文摘要
      Inter-floor peak traffic is a combination of up-peak traffic, down-peak traffic and inter-floor traffic. Lunch-peak traffic is a typical pattern of vertical transportation within a modern office building. Based on the characteristics of lunch-peak traffic,a dynamic optimization strategy of elevator group control isproposed in this paper by utilizing artificial immune algorithm. This strategy has a two-level control structure. One structure is the locally optimal assignment of a hall call performed by THV algorithm; the other is the globally optimal assignment of all hall calls, which is executed periodically by artificial immune algorithm.Compared to THV algorithm, the newly proposed strategy decreased the average system waiting time by 21.26%.Simulation results demonstrated that the performance of elevator system during lunch-time is greatly improved under control of the proposed strategy in this paper.