適應性網路模糊推論系統於方杯深引伸成形極限最佳化料片之研究

A Study of Optimum Blank of the Forming Limit in the Square Cup Deep Drawing Process by Using Adaptive Network Fuzzy Inference System

林千森(Chien-Sen Lin)

指導教授 ： 李經綸

淡江大學/工學院/機械與機電工程學系碩士班/碩士(2008年)

https://doi.org/10.6846/TKU.2008.00291

摘要

本文係結合動顯函有限元素分析與適應性網路模糊推論系統，以前向模式預測深引伸貫穿成形後之杯高，並發展一套適應性網路模糊推論系統之逆向模式，透過此逆向模式來預測最佳化料片之初始輪廓外形，以達到成形極限目標杯高之目的。
本文首先進行圓形初始料片之方杯深引伸成形極限分析，結合適應性網路模糊推論系統(ANFIS)，將不同直徑之圓形初始料片輪廓外形之節點編碼(N)、節點座標(X與Y)，及變形後之杯高分佈值(H)等資料建立前向模式知識規則庫，利用ANFIS進行方杯深引伸成形極限之最佳化初始料片輪廓外形分析。方杯深引伸成形極限分析之圓形初始料片面積為4417.86 ，而成形極限之最佳化料片面積則為4273.92 ，將可節省3.4%之材料面積；而成形後之最低杯高由29mm升高至33mm，可使極限杯高提升14%，其對應之容積分別為25902.77 與29489.04 ，故經最佳化後之工件容積可增加13.8%。由初始料片輪廓外形之最大距離除以方形沖頭平行邊寬所定義之極限引伸比(LDR)得知，圓形初始料片之LDR為2.5，而成形極限最佳化料片之LDR則為2.659，故LDR提升了6.4%。經數值模擬與實驗結果比較沖頭負荷與衝程關係，杯高分佈，變形歷程，成形極限圖，應力應變分佈圖等，顯示本動顯函有限元素程式與適應性網路模糊推論系統可精確的預測方杯深引伸成形極限之最佳化初始料片輪廓外形。

關鍵字

成形極限 ； 方杯深引伸 ； 動顯函有限元素 ； 最佳化 ； 適應性網路模糊推論系統

並列摘要

The aim of this paper was to combine the dynamic-explicit FEM and adaptive network fuzzy inference system (ANFIS) to predict the height of the cup after the deep drawing by the forward model. Meanwhile, an ANFIS inverse model was developed to predict the profile of the optimum blank to get the goal of the target height of the cup in the forming limit.
This study firstly explored the analysis of the forming limit in the circular initial blank during square cup deep drawing, and combined with the adaptive network fuzzy inference system (ANFIS) to establish a rule data base according to the information of the circular initial blank with different diameters of the node number, the node coordinates of the initial blank and the deformed height of the cup, and then used the ANFIS to perform the optimum blank profile of the forming limit. In the forming limit analysis of the square cup deep drawing, the area of the circular initial blank was 4417.86 , and the area of the optimum blank was 4273.92 , so the area of the material would reduce by 3.4%. After drawing, the lowest height of the cup could increase from 29mm to 33mm, so the limit height of the cup could rise by 14%. Also, the volume corresponding to the height of the cup 29mm and 33mm was 25902.77 and 29489.04 respectively, so the volume of the workpiece after optimizing could increase by 13.8%. According to the definition of limit drawing ratio (LDR), the LDR of the initial circular blank was 2.5 and the LDR of the optimum blank was 2.659, so the LDR could increase by 6.4%. The simulation results included the punch load and punch stroke, the distribution of the cup height, the deformation history, forming limit diagram, and the distribution of stress and strain. The comparing results of experiment and of simulation showed that the dynamic-explicit FEM and ANFIS can exactly predict the optimum blank of the forming limit in the square cup deep drawing process.

並列關鍵字

Forming limit ； Square cup drawing ； Dynamic-explicit finite element ； Limit drawing ratio ； Optimum ； ANFIS

參考文獻

9.鄧維明，「金屬板材深引伸成形極限之動顯函有限元素分析」，碩士論文，淡江大學機械與機電工程學系，2006年。

1.Takuda, H., Mori, K., Fujimoto, H. and Hatta, N., “Prediction of forming limit in bore-expanding of sheet metals using ductile fracture criterion,” Journal of Materials Processing Technology, Vol. 92-93, pp.433–438 (1999).

2.Zimniak, Z., “Implementation of the forming limit stress diagram in FEM simulations,” Journal of Materials Processing Technology, Vol. 106, pp.261–266 (2000).

3.Moshksar, M. M. and Mansorzadeh, S., “Determination of the forming limit diagram for Al 3105 sheet,” Journal of Materials Processing Technology, Vol.141, pp.138–142 (2003).

4.Ozturk, F. and Lee, D., “Analysis of forming limits using ductile fracture criteria,” Journal of Materials Processing Technology, Vol. 147, pp.397–404 (2004).