Computational Techniques for Multiphase Fluid Flows. A High Performance Computing Approach

      Аннотация: Computer simulation of fluid flows requires numerical solution of Partial Differential Equations (PDEs) which involves discretization methods such as Finite Volume Method (FVM). These methods produce a large system of linear equations which are difficult to solve. Iterative (Krylov Subspace) solvers are one of the best ways to deal with such large systems. These systems need huge amount of memory as well as they are computationally expensive. Modern supercomputers facilitate to run the simulations on multiprocessors by writing efficient parallel algorithms for iterative solvers. This book provides good background of Krylov Subspace solvers and computational methods for multiphase fluid flows. Furthermore, detailed discussions on developing parallel algorithms for these solvers have been provided using the special data structures facilitating inter-processor communications. It is a very useful book for advanced undergraduate and graduate students, who want to gain knowledge about parallel numerical algorithms applicable to Computation Fluid Dynamics (CFD) problems.

Ключевые слова: load balancing, Multiphase Flow, Volume of Fluid Method, Analytic Relation, Diagonal Format, Parallel Iterative Solvers, Sparse Matrices, Parallel Preconditioners, CFD