一维大气污染模型的特征-块中心差分法

Numerical methods based on characteristic block-centered finite difference method for one dimensional air pollution model

将特征-块中心差分法应用于一维大气污染模型中,求得非均匀网格上污染物浓度及其对空间变量的一阶导数项的差分解和误差估计,此法的计算量与基于线性插值的特征差分法相当,其近似解与基于二次插值的特征差分法的近似解有相同阶的误差估计。最后,通过数值实验说明了该方法的可行性和有效性。

In this paper, we propose the characteristic block-centered finite difference method on nonuniform grids to solve the problem of one dimensional air pollution model. Approximate solutions and error estimates of air pollution concentrations and their first derivatives for space variable are obtained . The computational labour of the method is the same as it of the characteristic difference method based on lin-ear interpolation. The error order of the approximate solutions is the same as it of the characteristic difference method based on quadratic interpolation. At last, a numerical example is given to illustrate feasibility efficiency of this method.