自组织分治求解分布式约束优化问题-openrepositoryofnational.pdfVIP

计算机研究与发展 ISSN N 111777TP J ournal of Comp ut er Research and Develop ment 45 ( 11) : 183 1 1839 , 2008 自组织分治求解分布式约束优化问题 黄 　晶　刘大有 　杨 　博 　金 　弟 (吉林大学计算机科学与技术学院　长春 　1300 12) (吉林大学符号计算与知识工程教育部重点实验室 　长春 　1300 12) ( huangj ing @j lu . edu . cn ) A SelfOrganization Based Divide and Conquer Algorithm f or Distributed Constraint Optimization Problems Huang J ing , Liu Dayou , Yan g Bo , and J in Di ( Col leg e of Comp ute r S cience an d Technology , J i l i n Uni ve rs i ty , Chang ch un 1300 12) ( M i nis t ry of E d ucation Key L aboratory of S y m bol Comp ut ation an d K now le d g e Eng i nee ri ng , J i l i n Uni ve rs i ty , Chang ch un 1300 12) Abstract 　Di st ribut ed con st raint op timization p roblem ( DCO P ) i s a kind of op timization p roblem orient ed to lar gescale , op en and dynamic net wor k environment s , w hich ha s been widely app lied in many fiel ds such a s co mp ut atio nal grid , multimedia net wor k s , ebu sine ss , ent erp ri se re source p lanning and so o n . Be side s t he f eat ure s such a s nonlinear an d co n st raintsati sf action w hich t he t raditio nal op timization p roblem s have , DCO P ha s it s di stinct feat ure s inclu ding dynamic evolution , regional infor mation , localized cont rol and a synchro nou s up dating of net wor k st at e s. It ha s beco me a challenge for co mp ut er scienti st s to fin d out a lar gescale , p arallel an d int elligent solution for DCO P . So f ar , t here have been a lot of met ho ds for solvin g t hi s p roblem . However , mo st of t hem are not co mp let ely decent ralized and require p rior knowledge such a s t he globa