A feature-aware SPH for isotropic unstructured mesh generation

In this paper, we present a feature-aware SPH method for the concurrent and automated isotropic unstructured mesh generation. Two additional objectives are achieved with proposed compared to original SPH-based generator (Fu et al., 2019). First, feature boundary correction term is introduced address issue of incomplete kernel support at vicinity. The generation curves, surfaces volumes can be handled concurrently without explicitly following dimensional sequence. Second, two-phase model characterize mesh-generation procedure by feature-size-adaptation phase mesh-quality-optimization phase. By proposing new error measurement criterion an adaptive control system two sets simulation parameters, faster feature-size adaptation local mesh-quality improvement merged into consistent framework. validated set 2D 3D numerical tests different complexities scales. results demonstrate that high-quality meshes generated significant speedup convergence.