| Summary: | 本研究利用月桂酸和醋酸視黃酯以正己烷當作有機溶劑，進行 Novozym®435 酵素催化合成月桂酸視黃酯，並藉由反應曲面法 (response surface methodology, RSM) 及類神經網路 (artificial neural network, ANN) 探討最適化研究。以三階四變中心混成實驗設計法 (central composite design, CCD) 設計出 27 組實驗組數，其影響因子分別為：酵素用量 (10~50 mg)、溫度 (40~60℃)、時間 (2~6 小時) 和基質莫耳數比 (酸/醇) (1~5)，其得到之最高相對轉化率為 82.64 ± 1.94%，而最低之相對轉化率為 14.52 ± 1.85%。RSM 分析結果之 R2 為 0.9887，變異數分析 (ANOVA) 以酵素用量和基質莫耳數比 (酸/醇) 影響較為顯著，將實驗之 27 組數值進行 ANN 系統分析，得到之最適分析模型為：訓練模式 LM、轉換函數 Tanh、循環次數 104、隱藏層神經元個數 6，均方根誤差 (root mean squared error, RMSE) 和 R2 分別為 0.22347 和 0.99994，而 RSM 和ANN 最適化條件組之相對轉化率分別為 84.7 ± 1.16% 和 88.31 ± 0.3%，酵素重覆利用次數則在第 5 次開始其結果有明顯下降趨勢。驗證組部分 ANN 和 RSM 之 RMSE 分別為 0.04 和 0.12，結果表示以 ANN 預測系統較 RSM 系統精準。 In this study, immobilized lipase from Candida antarctica (Novozym® 435) was used in enzymatic catalysis to generate retinyl laurate by esterification of retinyl acetate with lauric acid in n-hexane. This reaction was model and optimized by response surface methodology (RSM) and artificial neural network (ANN). A 3-level-4-factor central composite design (CCD) was employed in this study to design 27 experiments. The experimental variables included enzyme amounts (10~50 mg), temperature (40~60℃), time (2~6 hours) and substrate molar ratio (acid/alcohol=1~5). Within the 27 experiments, the highest and lowest relative conversion are 82.64 ± 1.94 and 14.52 ± 1.85%, respectively. The coefficient of determination (R2) calculated from the design data of RSM is 0.9887, and analysis of variance (ANOVA) showed that enzyme amounts and substrate molar ratio exhibited the stronger effect on the molar conversion than other variables. As ANN was employed to model the biosynthesis, the optimal architecture of ANN was founded as follows: learning algorithm (LM), transfer function (Tanh), iterations (104) and the nodes of hidden layer (6). The root mean squared error (RMSE) and R2 were 0.22347 and 0.99994, respectively, suggesting that ANN was better than RSM in data fitting. The relative conversion of RSM and ANN optimal verification test were 84.7 ± 1.16% and 88.31 ± 0.3%, and the relative conversion decreased obviously when reusing enzymes by the fifth times. The results of verification test suggested that the RMSE of RSM and ANN were 0.04 and 0.12, respectively. This result shows ANN is more appropriate as the system of data prediction in this study. 摘要 I Abstract III 目錄 V 圖目錄 VIII 表目錄 XIII 第1章 前言 1 第2章 文獻回顧 3 2.1維生素 A 3 2.1.1介紹 3 2.1.2視黃酯合成之相關研究及應用 5 2.2酵素 8 2.2.1酵素簡介 8 2.2.2脂解酵素 10 2.2.3 Novozym®435 12 2.3最適化方法 13 2.3.1最適化方法簡介 13 2.3.2反應曲面法 16 2.3.3類神經網路 18 第3章 材料與方法 24 3.1實驗材料 24 3.1.1藥品 24 3.1.2儀器設備 25 3.2合成方法 26 3.2.1實驗架構 26 3.2.2實驗方法 27 3.2.3合成方法 28 3.3分析方法 29 3.3.1分析條件 29 3.3.2產物之相對轉化率 (Relative Conversion) 計算方式 30 3.3.3統計分析 30 3.3.4類神經網路 (Artificial Neural Network) 31 第4章 結果與討論 32 4.1圖譜分析 32 4.2變數分析 34 4.2.1酵素用量 34 4.2.2超音波功率 36 4.2.3溫度 38 4.2.4時間 40 4.3反應曲面法 42 4.3.1實驗設計及月桂酸視黃酯相對化率實驗結果 42 4.3.2酵素催化合成月桂酸視黃酯之變異數分析 45 4.3.3酵素催化合成月桂酸視黃酯之曲面分析 47 4.4類神經網路 54 4.4.1訓練模式 (Learning Algorithm) 54 4.4.2 轉換函數 (Transfer Function) 和隱藏層神經元 (Node) 個數 56 4.4.3循環次數 (Iterations) 58 4.4.4倒傳遞類神經網路 60 4.5最佳組實驗驗證和酵素重複再利用 66 4.6反應曲面法和類神經網路比較 68 第5章 結論與未來工作 70 5.1結論 70 5.2未來工作 71 參考文獻 72
|
|---|