Please use this identifier to cite or link to this item:
標題: Data-mining assisted structural optimization using the evolutionary algorithm and neural network
作者: Chen, T.Y.
Cheng, Y.L.
關鍵字: structural optimization;data mining;evolution strategy;artificial;neural network;global optimization;multimodal functions;genetic algorithms;minimum;search;design
Project: Engineering Optimization
期刊/報告no:: Engineering Optimization, Volume 42, Issue 3, Page(s) 205-222.
The use of evolutionary algorithms for global optimization has increased rapidly during the past several years. But evolutionary computations have a common drawback: they need a huge number of function evaluations. This makes them inadequate for structural optimization. To overcome this difficulty, the authors propose a method that integrates the evolutionary algorithm with data mining and approximate analysis to find the optimal solution in structural optimization. The approximate analysis is used to replace exact finite element analyses and the data mining is employed to identify feasible solutions. These combined efforts can reduce the computational time and search the feasible region intensively. As a result, the efficiency and quality of structural optimization using evolutionary algorithms will be increased. Some test problems show that the proposed method not only finds the global solution but is also less computationally demanding.
ISSN: 0305-215X
DOI: 10.1080/03052150903110942
Appears in Collections:機械工程學系所

Show full item record

Google ScholarTM




Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.