α-生育酚在超临界CO2中溶解度神经网络模型的建立

    Solubility Neural Networks Model of α-Tocopherol in Supercritical CO2

    • 摘要: 生育酚有很高的生理活性,油脂生产中得到的脱臭馏出物含有丰富的天然生育酚。作为萃取生育酚的基础,该文对甲酯化油脂脱馏出物中α-生育酚在超临界CO2中的溶解度进行了测试,并用Chrastil分子缔合模型和RBF神经网络模型对溶解数据进行了拟合。Chrastil分子缔合模型的相对误差为25.36%。对于RBF神经网络模型,经过网络学习和训练,训练集平均误差仅为0.32%,测试集误差为6.48%,效果比较理想。

       

      Abstract: Tocopherol has very high physiological activity. Soybean deodorizer distillate contains plentiful natural tocopherol. For extracting tocopherol from soybean deodorizer distillate, solubility of α-tocopherol in supercritical CO2 was measured. The solubility data were correlated respectively by Radial Basis Function(RBF) neural networks model and Chrastil model. Relative error of Chrastil model is 25.36%. By means of learning and training on RBF neural networks model, the average error of training set is 0.32% and that of testing set is 6.48%. RBF neural networks model is better comparatively.

       

    /

    返回文章
    返回