以模糊基因演算法進行結構最佳化設計

Structure Design Optimization using Fuzzy-Genetic Algorithm

陳俊富(Chun-Fu Chen)

指導教授 ： 黃秀英

國立臺北科技大學/機電學院/車輛工程系所/碩士(2011年)

https://doi.org/10.6841/NTUT.2011.00429

摘要

在現今的車輛結構設計上，已逐漸藉由電腦輔助求得最佳結果，本研究係結合模糊演算法與基因演算法，提昇最佳化分析的效率，將本研究所提出的新演算法稱為模糊基因演算法（Fuzzy-Genetic Algorithm）。
本模糊基因演算法中，先確定結構最佳化問題的分析母體（Population），建立相對應的模糊規則庫，進行模糊化，將母體做分群（Group），再對分群使用基因演算法求得最佳化設計。研究中採用HyperStudy最佳化分析模組（Optimization toolbox）中的基因演算法（Genetic Algorithm）與研究中建立的模糊規則庫結合，形成模糊基因演算法，與分析軟體HyperWorks做結合而求得最佳化結果。本研究中並以平板模型為例，作演算法實例驗證。
研究結果中，平板模型的最佳化，比較僅使用基因演算法與模糊基因演算法結果，確定演算法的正確性。分析結果與傳統最佳化做比較，可有效地節省約21％的遞迴次數，對較大型的設計預期有更大改善效益。

關鍵字

模糊 ； 基因 ； 演算法 ； 最佳化分析應用

並列摘要

Nowadays, vehicle designs heavy rely on computer simulation. And optimization analysis becomes one of the key processes. The study investigates the feasibility of applying fuzzy and genetic algorithm to the structure optimization. A method combing with newly developed fuzzy rule set and genetic algorithm was proposed and called Fuzzy-Genetic algorithm.
In the study, a fuzzy logical set was defined based on the decent function and constraints. In the study, traditional gradient optimization, optimization combined with fuzzy, genetic optimization, genetic combined with fuzzy optimization were studied and compared. Fuzzification was defined and coded using Matlab. Optimization analysis was performed using the commercial code HyperStudy. For Fuzzy-Genetic optimization, genetic optimization was performed, and then fuzzy logic was applied. The proposed method intended to expedite optimization convergence.
Optimization results were compared. Results showed the proposed Fuzzy-genetic algorithm could help in reach optimization quick from the studied examples. It is improve the efficiency of iteration around 21%.

並列關鍵字

fuzzy ； genetic ； algorithm ； optimum analysis application

參考文獻

[1] C. Y. Chen and C. C. Shieh, "Fuzzy multiobjective topology optimization," Computers and Structures, vol. 78, no. 1, 2000, pp. 459-466.

[2] W. Pedrycz and A. Bargiela, "Fuzzy clustering with semantically distinct families of variables: Descriptive and predictive aspects," Pattern Recognition Letters, vol. 31, no.13, 2010, pp. 1952-1958.

[3] C. Carlsson and R. Fuller, "Fuzzy multiple criteria decision making: Recent developments," Fuzzy Sets and Systems, vol. 78, no. 2, 1996, pp. 139-153.

[4] K. Liu, W. Shi and H. Zhang, "A fuzzy topology-based maximum likelihood classification," ISPRS Journal of Photogrammetry and Remote Sensing, vol. 66, no. 1, 2011, pp. 103-114.

[5] A. Azadeh, M. Saberi, M. Anvari, A. Azaron and M. Mohammadi, "An adaptive network based fuzzy inference system-genetic algorithm clustering ensemble algorithm for performance assessment and improvement of conventional power plants," Expert Systems with Applications, vol. 38, no. 3, 2011, pp. 2224-2234.