Abstract
One of the major advantages of adaptive mesh generation is the optimum fitting of the mesh to the problem, leading to good memory efficiency. By suitably choosing the refinement parameters, the adaptive mesh generation is also time efficient. As the best refinement parameters depend on the problem to be solved and can also change during the adaptive solution process, they must be chosen dynamically. The choice is based on the error distribution, that is available from the local error estimation. Therefore no additional computational cost arises due to the dynamic parameter setting.
Get full access to this article
View all access options for this article.