基于人工神经网络的咖啡因溶解度的建模

Model of solubility of Caffeine Based on the Artificial Neural Networks

利用激光法测定的0～50℃温度范围内咖啡因在水和乙醇中的溶解度数据,经过比较选择了较好的人工神经网络模型—2-2-1向后传播(BP)人工神经网络模型,利用训练后的2-2-1BP人工神经网络模型对非线性的咖啡因溶解度数据进行了再现回归、内插和外推,在水中溶解度数据的回归、内插的误差均在0.07％以内,外推的误差在0.2％以内;在乙醇中的溶解度数据的回归误差在0,07％以内,内插和外推误差均在4％以内,效果十分令人满意。

Based on the solubility of caffeine in water and ethanol from 0 to 501 measured by laser method, a 2 - 2 -1 backpropagation(BP) artificial neural networks(ANN) model was selected from many other ANN models. The regression, interposition and extrapolation of the data on caffeine solubility was taken with trained 2-2-1 BP ANN model. The outcome was very satisfactory.