Abstract

Abstract The accuracy of numerical results obtained by using nonuniform grid distributions is strongly affected by the size of the grid spacing. The present work focuses on a detailed procedure for the accuracy tests of the numerical results obtained by using nonuniform grid distributions. It is shown that the degree of nonuniformity of a certain grid distribution with a fixed number of grid lines also has a strong effect on the numerical results. For a fixed number of grid lines there is only one optimum nonuniform grid distribution for each physical case. Furthermore, an accurate nonuniform grid for a physical situation with steep gradients presents a very low efficiency in terms of CPU time when used for simulating cases with smaller gradients.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call