在本論文中使用了一組四個參數的描述因子，包括：∑MV(ter)R(ter)、LF、∆XSB和∑PEI來估計84個高分子的玻璃轉化溫度，並且分別以多元線性迴歸分析與逆傳遞類神經網路來建造玻璃轉化溫度的估計模型。在本文中建立了以4-8-1層架構的最佳類神經網路，經過訓練組的訓練與驗證組的驗證測試，分別得到了均方根誤差為3.3 K(R2=0.9975)與13.9 K(R2=0.9513)的結果。從測試的結果顯示，以類神經網路模式估計高分子的玻璃轉化温度可以得到非常準確的估計值。
In this thesis, a set of four-parameter descriptors, ∑MV(ter)R(ter), LF, ∆XSB and ∑PEI were used to correlate with glass transition temperatures for 84 polymers. Multiple linear regression analysis and back-propagation artificial neural network (ANN) were used to generate the model. The final optimum neural network with 4–8–1 structure produced a training set root mean square error(RMSE) of 3.3 K (R2=0.9975) and a validation set RMSE of 13.9 K (R2=0.9513). The results show that the ANN model obtained in this study is accurate in the estimation of values for polymers.
