Труды Института системного программирования РАН (Feb 2019)
Multiscale approach for simulation of complex transient processes of fluid flows in technical systems
Abstract
The paper presents a multiscale approach for simulation of the two-phase flows processes in complex technical systems. The multiscale approach is based both on the division of the computational domain into subdomains with their own system of equations, as well as on the splitting of the initial system of equations into several subsystems each is valid to the corresponding scale under consideration. The problem of the far field acoustic noise calculation during the launch of a rocket vehicle is considered as an example of the possible use of a multiscale model. The case is studied with account to noise suppression due to water supply into the gas jets of the propulsion system. Other areas of application of the multiscale model include the cases of the oil and gas industry: killing gas-producing wells located at great depth, killing oil wells with a high gas factor at the fields. The proposed multi-scale mathematical model includes 5 sub-models: 1) gas dynamics of high-speed multicomponent gas mixture flows; 2) the hydrodynamics of a two-phase mixture flow in a homogeneous approximation with the account for the compressibility of the gas phase and the mass exchange between the phases; 3) the liquid-gas interface transport; 4) the transport of a cloud of droplets and its interaction with a gas-liquid medium; 5) noise calculation in the far field using the Ffowks Williams-Hawking acoustic analogy. The model can be extended to include additional sub-models, such as the Eulerian-Lagrange Jet Atomization. The implementation of the submodels can be done on the basis of open source packages: OpenFOAM, Nektar ++, ITHACA-FV. The acoustics library and the hybrid algorithm for compressible homogeneous two-phase flow are implemented as libAcoustics and hybridCentralSolvers modules based on the OpenFOAM open package. The source code of the developed model is freely available through the GiHub project https://github.com/unicfdlab.
Keywords
