Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Machine Design & Research》 2006-05
Add to Favorite Get Latest Update

Multiobjective Collaborative Optimization Based on BP Neural Network & Pareto Genetic Algorithm

ZHOU Sheng-qiang, XIANG Jin-wu (School of Aeronautic Science and Technology, Beijing University of Aeronautics and Astronautics, Beijing 100083, China)  
Multidisciplinary design optimization (MDO) problems are multiobjective in nature.Pareto Genetic Algorithm (PGA) can provide the Pareto optimal solution set which is helpful for design decision making.For computational efficiency problem resulted from directly using PGA in Collaborative Optimization (CO) framework,a method based on BP artificial neural network PGA is proposed to resolve multiobjective collaborative optimization problem.After employed design of experiment (DOE) method to select design point,neural network is used to establish the global response surface approximations of disciplinary subproblem optimizaton results.Then PGA is used to solve multiobject problem of the system-level optimization.The method is applied to two-objective optimization of the conceptual design of a trunkeliner.Comparision with the results of the multidisciplinary feasible(MDF) method using PGA,indicates that the proposed method can effectively approximate Pareto optimal front.
【Fund】: 新世纪优秀人才支持计划资助项目(NCET204201Q)
【CateGory Index】: TP18
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved