吸附数据的方程拟合和给定条件的自适应最小二乘样条

Adsorption Data Fitting and Adaptive Least Squares Spline Which Meets Given Conditions

本文提出了一类给定条件的样条拟合方法。作者导出了能满足给定条件的最小二乘样条拟合正规方程组,讨论了样条节点数及其位置分布对拟合精度的影响,以函数值变化激烈处设置较密节点为原则,成功地开发了一种自适应样条拟合算法。为达到预先规定的拟合精度,这种算法能自动确定节点数及其位置。作者把它用于吸附过程数据的拟合,获得了满意的结果。

Least squares polynomial spline was proposed to fit scientific and engineering datawith given conditions.Normal equations for least squares polynomial spline was derivedand solved by Conjugate Gradient method.An algorithm for adaptively determining thenumber of node points and optimizing the distribution of them was developed.A spline fit-ting software package coded in FORTRAN-77 was programmed and structured to beflexible and convenient for general usage.As an practical example,the software packagewas employed to fit the adsorption data of selectivity and breakthrough curves satisfactori-ly.