本研究係使用田口法與遺傳演算法探討點字觸摸顯示方ABS塑膠零組件射出成型最佳化參數之研究,以避免射出成品產生短射、變形及表面缺陷等問題。研究中先採用參數化繪圖軟體進行實體繪製,接而利用Moldex3D執行模流分析,探討每一項射出參數對塑膠材料射出成型的影響。由分析結果顯示,模具溫度、融膠溫度、壓力切換和填充時間為顯著因子,並導入田口法利用直交表配合信號雜訊比與變異數分析,獲得最佳射出參數與因子的影響程度,最後再導入遺傳演算法並經輪盤選擇、基因交配和基因突變世代族群演化來找出最佳目標值。 本研究透過田口法與遺傳演算法皆可達成降低塑膠零組件變形、短射和表面缺陷等目標,從兩種方法所得結果比較顯示,遺傳演算法經過世代族群的演化最終的目標值會比田口法所獲得的目標值來得好,可證實遺傳演算法應用於射出成型加工因子最佳化搜尋的實用性。
The thesis uses Taguchi method and genetic algorithm to study the optimal parameters of injection molding to avoid the deformation, short shot and surface defects for the ABS plastic components of Braille cell. At first the parametric drafting software is applied for the entity design and rendering. Then the Moldex3D software is used for mold flow analysis on the influence of each injection parameters. The results show mold temperature, melt temperature, injection pressure, and filling time are the most influential factors on reducing component deformation. Besides, the orthogonal array of Taguchi method is applied to compare with S/N (Signal Noise Ratio) and ANOVA (Analysis of Variation). The influence degree of factors and the optimal parameters of injection molding are obtained. Finally, genetic algorithm with roulette wheel selection, crossover, and mutation is applied to create a new set of population. The optimal parameters of injection molding are also obtained by repeating this process. The deformation, short shot and surface defects of the plastic components in Braille cell are reduced by Taguchi method and genetic algorithm. The comparison of the results shows that the prediction accuracy of genetic algorithm is better than Taguchi method. It is proved that genetic algorithm is to be able to supply a useful optimal soft computing approach in the injection molding category.