A block implicit lower-upper symmetric Gauss-Seidel (LU-SGS) approximate factorization scheme is developed and implemented for unstructured grids of arbitrary topology including viscous adaptive Cartesian grids. CPU times and memory requirements for the block LU-SGS (BLU-SGS), the original LU-SGS, and a fully implicit scheme (FIS) with a preconditioned conjugate gradient squared solver for several representative test examples are compared. Computational results showed that the block LU-SGS scheme requires about 20-30% more memory than the original LU-SGS but converges many times faster. The BLU-SGS scheme has a convergence rate competitive to and in many cases faster than the FIS while requiring much less memory.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.