基于非线性最小二乘法的Van Laar常数计算

Compute Van Laar constant based on nonlinear least-squared method

以汽液平衡实验数据或以基于UNIFAC模型计算获得的汽液平衡数据为基础,采用非线性最小二乘法与数值计算的方法相结合计算VanLaar常数,对计算的过程和方法进行了详细的分析,并利用计算得到的VanLaar常数对20余个二元混合物进行模拟计算,其中包含了恒温和恒压条件下的数据计算,也包含了不同温度和不同压力条件下的汽液平衡数据计算,并应用计算结果进行验算,与相应的实验结果或通过UNIFAC模型计算获得的结果进行对比,计算所获得数据与实验数据相比,平均误差小于3%。同时,对不同的物系进行误差分析,非理想物系和强极性物系的误差较大,一般物系的平均误差小于1%,并使用计算获得的VanLaar常数进行汽液平衡曲线模拟,与实际汽液平衡曲线对比,效果理想。结果表明,采用非线性最小二乘法结合数值计算方法获得的VanLaar常数误差较小,可用于相关汽液平衡的模拟和计算。

Based on the experiment data of liquid equilibrium or the data from UNIFAC model of liquid equilibrium, the Van Laar constant is computed by the nonlinear least-squared method combined with numerical computation. More than 20 binary mixtures are simulated. It includes the condition of constant temperature, constant pressure, etc. Compared with the data of experiment, the average error of computation is less than 3%. The error of non-idea and strong polar mixture is larger than other. The error of general mixture is less than 1%. The method of nonlinear least squared to compute Van Laar constant is the ideal replacement of empirical method.