论文部分内容阅读
为了更好地关联和预测共溶剂-超临界(sc)CO_2中固体的溶解度,本文采用经局部动量法和自适应理论优化的小波神经网络模型(WNN),分别以共溶剂的溶剂参数α、β、π~*和溶解度参数δ_d、δ_p、δ_h为影响因素,关联了4种共溶剂-SC CO_2体系中萘普生的溶解度,以预测其在乙醇-SC CO_2中的溶解度。其关联误差AARD分别为1.94%和2.12%;其预测误差AARD分别为8.14%和30.32%。以上结果表明共溶剂的溶剂参数α、β、π~*是共溶剂-SC CO_2中固体溶解度的主要影响因素;优化的WNN模型能较好地关联和预测共溶剂-SC CO_2中固体的溶解度。
In order to better correlate and predict the solubility of solid in cosolvent-supercritical carbon dioxide (CO2), a wavelet neural network model (WNN) optimized by local momentum method and adaptive theory was used. Solvent parameters α, β, π ~ * and solubility parameters δ_d, δ_p and δ_h were used as the influencing factors, and the solubility of naproxen in the four co-solvent-SC CO_2 systems was correlated to predict their solubility in ethanol-SC CO_2. The AARD of its correlation error was 1.94% and 2.12%, respectively; its prediction error AARD was 8.14% and 30.32% respectively. The above results show that the solvent parameters α, β and π ~ * of the co-solvent are the main factors affecting the solubility of the co-solvent-SC CO_2. The optimized WNN model can better correlate and predict the solubility of the co-solvent-SC CO_2.