應用柔性演算法於航太鋁合金銲接參數最佳化之研究

The Optimal Parameter Design of Aerospace Aluminum Alloy Weldment via Soft Computing

10.6220/joq.2014.21(3).05

張志平(Jhy-Ping Jhang)；劉孝先(Hsiao-Hsien Liu)

理想解類似度順序偏好法 ； 多重品質特性 ； 倒傳遞類神經網路 ； 基因演算法 ； 模擬退火法 ； 參數最佳化 ； 最佳化參數設計 ； TOPSIS ； multiple quality characteristics ； artificial neural network ； genetic algorithm ； simulated anneal ； parameter optimization ； optimal parameter design

品質學報

21卷3期（2014 / 06 / 30）

205 - 216

繁體中文

本研究以田口方法進行惰性氣體鎢棒電弧銲接實驗，探討非破壞性品質特性－銲道寬度、厚度、熔入深比以及破壞性品質特性－拉伸、衝擊值等五個銲接品質特性，再應用理想解類似度順序偏好法（Technique for Order Preference by Similarity to Ideal Solution）與倒傳遞類神經網路（Artificial Neural Network）搜尋最佳化參數設計，結合模擬退火法（Simulated Anneal）、基因演算法（Genetic Algorithm）等柔性演算法（Soft Computing）試圖找出航太鋁合金板材銲接參數最佳化。研究結果找出航太鋁合金銲接參數最佳化設計，可提供銲接相關業者針對航太鋁合金板材銲接參數作準確又實用的求解程序。

This research uses Taguchi method to proceed with the experiment of Tungsten In Gas (TIG), to discuss the nondestructive quality characteristics, welding width, welding thickness and the ratio of melting into the deep; and the destructive quality characteristics, tensile strength and shock value. It uses TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) and ANN (Artificial Neural Network) to train the optimal function framework of parameter design. It combines SC (Soft Computing) of SA (Simulated Anneal) and GA (Genetic Algorithm) to search the optimal parameters combination for the optimal parameter of weldment. To improve previous experimental methods for multiple characteristics, this research method employs SA to search the optimal parameter such that the potential parameter can be evaluated more completely and objectively. Additionally, the model can learn the relationship between the welding parameters and the quality responses of different materials to facilitate the future applications in the decision-making of parameter settings for automatic welding equipment. The research results can be presented to the industries as a reference, and improve the product quality and welding efficiency to relevant welding industries.

1. Chan, H. L.,Liang, S. K.,Lien, C. T.(2006).A new method for the propagation system evaluation in wireless network by Neural Networks and Genetic Algorithm.International Journal of Information Systems for Logistics and Management,2(1),27-34.
2. Hwang, G. L.,Yoon, K.(1981).Multiple Attributes Decision Making Methods and Applications.New York:Springer-Verlag.
3. Juang, S. C.,Tarng, Y. S.(2002).Process parameter selection for optimizing the weld pool geometry in the tungsten inert gas welding of stainless steel.Journal Materials Processing Technology,122(1),33-37.
4. Su, C. T.,Chiu, C. C.,Chang, H. H.(2000).Optimal parameter design via neural network and genetic algorithm.International Journal of Industrial Engineering,7(3),224-231.
5. Tong, L. I.,Wang, C. H.(2000).Optimizing multi-response problems in a dynamic system by grey relational analysis.Journal of the Chinese Institute of Industrial Engineers,17(2),147-156.

1. Wu, Pei-Yu,Jhang, Jhy-Ping(2017).APPLICATION OF ARTIFICIAL NEURAL NETWORK AND TOPSIS FOR THE OPTIMAL THRUST OF SMT DISPENSING PROCESS PARAMETERS.品質學報,24(5),324-334.
2. 羅惠瓊,張志平,朱佳荷(2019)。應用類神經網路與基因演算法於鎂合金與銅異種金屬銲接參數最佳化之研究。品質學報，26(6)，381-394。
3. 張志平、胡庭睿(2018)。鎂合金多重品質特性銲接參數最佳化之研究。品質學報，25(2)，106-119。
4. (2024)。應用機器學習於鈦與鎂異種金屬銲接參數最佳化。品質學報，31(1)，27-44。