高压汽液平衡的统计回归与推算(Ⅰ)

Statistical Regression and Prediction of High-Pressure Vapor-Liquid Equilibrium (Ⅰ)

以最大似然原理为基础的统计回归方法在常压汽液相平衡数据回归中的应用已有一些报告，且回归结果非常令人满意．但此法在高压根平衡中的应用由于存在一定的困难．本文采用最大似然原理进行了国内外尚无完整研究报告的高压汽液平衡的统计回归，分别采用新的DDLC－SRK模型和SRK方程，用3种不同的目标函数对52套高压汽液平衡数据进行了统计回归，同时也进行了经典回归．得出结论，统计回归优于经典回归，且以双应答回归结果最好．

The method of statistical regresion on the basis of the maximum likelihood principle has beensuccessfully applied in the correlation of vapor-liquid equilibrium data at atmosphenc pressure. However,the application of the method in the correlation of high-pressure phase equilibria has not yet been reported in details so for. The high-pressure vapor-liquid equilibrium dara of 52 binary systems were correlated by the maximum likelihood principle with a new DDLC-SRK model and SRK equation and threekinds of objective functions in the study. The results were also compared with those obtained by the classical regression.It is confirmed that the statistical regression is much superior to the classical regression andthe dual response regression results are better than the single response regression ones.