Residual-Based Grid Adaptation for Meshless LSFD-U Solver

The Paper projects the meshless LSFD-U methodology as a Cartesian grid method. A point distribution obtained from a Cartesian grid is employed in a solution update, leading to an automated computational fluid dynamics process for point adaptation. An estimate of the local truncation error obtained as a higher-order residual is employed for effecting a solution-based point adaptation. Asalient aspect of this methodology is the definition of a local length scale derived from an estimate of the truncation error, which not only provides a criterion for local refinement or coarsening but also can help in evolving an automatic termination criterion for the adaptive cycle. Toward point adaptation for flows with discontinuities, a hybrid adaptive strategy employing both the error estimator and indicator is employed. Results presented demonstrate the efficacy of the proposed methodology