Computational methods for solving coupled systems of partial differential equations can generally be divided into segregated and monolithic solvers. Monolithic solvers are often considered to be computationally expensive, and it is believed that their rewards are realized only in situations where segregated solvers have convergence problems due to the strong coupling. We give opposite empirical evidence by demonstrating that cost-effective monolithic solvers may be derived from using segregated solvers as preconditioners in the iterative solution of monolithic systems. © 2012 Springer-Verlag.
CITATION STYLE
Malinen, M. (2012). The development of fully coupled simulation software by reusing segregated solvers. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7133 LNCS, pp. 242–248). https://doi.org/10.1007/978-3-642-28151-8_24
Mendeley helps you to discover research relevant for your work.