| Full text | |
| Author(s): |
Total Authors: 4
|
| Affiliation: | [1] Universidade Federal do Paraná. Programa de Pós-Graduação em Engenharia Ambiental - Brasil
[2] Universidade Federal do Paraná. Departamento de Engenharia Ambiental - Brasil
[3] Universidade de São Paulo. Instituto de Ciências Matemáticas e de Computação - Brasil
[4] Universidade de São Paulo. Instituto de Ciências Matemáticas e de Computação - Brasil
Total Affiliations: 4
|
| Document type: | Journal article |
| Source: | Trends in Computational and Applied Mathematics; v. 26, 2025-02-10. |
| Abstract | |
ABSTRACT In this work, three implementation options for the ghost cell immersed boundary method are compared. These options are alternatives to a rather common implementation of the method that is susceptible to numerical instability in the calculation of the bilinear interpolation in some cases. The method is implemented for a second-order spatial discretization of the heat equation in a non-rectangular domain and the errors for each option are analyzed in terms of the order of accuracy and the way they are distributed in the domain. The best option, which was the only one to maintain the second order of convergence of the discretization, is to consider non-symmetric extrapolation with bilinear interpolation, instead of using inverse distance weighted interpolation with symmetric or non-symmetric extrapolation. (AU) | |
| FAPESP's process: | 18/24284-1 - Study on exchanges between Earth's surface and atmosphere using Large-Eddy Simulation |
| Grantee: | Livia Souza Freire Grion |
| Support Opportunities: | Research Grants - Young Investigators Grants |