本研究透過反應曲面法結合蟻群演算法，建立製程參數與品質特性之 回歸模型並進行最佳化搜索，以獲得最佳製程參數，提升材料之縫合線強度。以15%短碳纖維含量之尼龍6 作為射出材料，並使用Moldex3D 與 SolidWorks 進行射出成型模具設計，再來使用Box-Behnken 設計法規劃實驗並建構數學模型，並將其作為蟻群演算法之適應函數，最後各別以反應曲面法與蟻群演算法求解製程最佳化問題，探討尼龍6 纖維複合材料於射出成型製程中，不同的製程參數對於其縫合線強度之影響，並比較反應曲面法與蟻群演算法兩種最佳化方法之優劣。 實驗結果顯示，縫合線強度受到熔膠溫度的影響最為顯著，熔膠溫度越高，縫合線之結合性越好，使其抗拉強度有明顯的提升。反應曲面法最佳化之預測與實驗誤差為1.53%，縫合線強度改善率為3.89%;另一方面，蟻群演算法最佳化之預測與實驗誤差為0.68%，縫合線強度改善率為4.54%，根據上述結果表示，蟻群演算法之預測誤差較小，且縫合線強度改善率較高，代表其最佳化製程參數之能力較佳，但綜觀結果來看，反應曲面法與蟻群演算法都能夠有效地最佳化製程參數，提高縫合線強度。 本研究成功地利用CAD 與CAE 工具完成模具設計，有效降低試模成 本，並成功地結合反應曲面法與蟻群演算法，找出最佳的製程參數組合，提 升尼龍6 纖維複合材料之縫合線強度，降低實驗成本並節省時間，提供技 術人員另一種有效的製程最佳化方法。;This study combines response surface methodology (RSM) and ant colony optimization (ACO) to establish the regression model between process parameters and specimen quality. This combination performs optimization to obtain the optimal process parameters and improve the weld line strength of specimen. First,nylon 6 containing 15% short carbon fiber was used as the injection material, and Moldex3D and SolidWorks were used for mold design. Next, the Box-Behnken design (BBD) was employed to conduct the experiment and establish amathematical model which was the fitness function of the ACO. Finally, RSM and ACO were used to solve the optimization problem. The influence of different process parameters on the weld line strength of the specimen was discussed. The optimization results of RSM and ACO were compared with each other. The experimental results show that the weld line strength is significantly affected by the melt temperature. The higher the melt temperature, the better the bonding of the weld line which significantly improves its tensile strength. The prediction error of weld line strength using RSM is 1.53%, and its improvement rate is 3.89%. Moreover, the prediction error of weld line strength using ACO is 0.68%, and its improvement rate is 4.54%. According to the present results, it shows that ACO has a smaller prediction error and a higher improvement rate of weld line strength. This study indicates that ACO has better ability to optimize process parameters. In conclusion, both RSM and ACO can effectively optimize process parameters and improve weld line strength, which provide technicians with another effective method for process optimization.
