基于表现型共享的多目标粒子群算法研究-福州大学学报.pdf

35 3 ( ) V o.l 35 No. 3 2007 6 Journa l of F uzhou U n iversity( N a tura l Sc ience) Jun. 2007 1000 - 224 3( 2007) 03- 0365 - 05 , , (, 350002) , , , Pa reto , . , P are to , (M OPSO ) . , , EPSHYP R2 . , , P areto. ; ; TP 39 A M ultiobjective particle swarm optimization research based on phenotype sharing HUANG M in, CH EN Guo- long, GUO W en - zhong ( College o f M athem atics and Com puter Science, Fuzhou U niv ers ity, F uzhou, F uj ian 350002, Ch ina) Abstract: In mu ltiob ject ive particle sw arm opt mi ization, a p artic le flies according to its histo ry best position and the leaders, therefo re how to define the fitness function in order to gu ide the search to w ards the g loba l Paretooptmi al reg ion and m ain tain population d iversity in the nondom inated front is the key to su ccess. To solve the above prob lem, a fitness funct ion based on the phenotyp e sh aring is designed con sidering both the Pareto dom inance and the ne ighborhood den sity o f the ob jective sp ace. Then a m ultiobjective p artic le sw arm optmi ization a lgorithm b ased on the phenotype fitness function is proposed. In order to va lidate the proposed algorithm, the D om inance ranks indicator is u sed