Finite Volume method for general compressible naval hydrodynamics. (15th January 2020)

• Record Type: Journal Article
• Authors: Gatin, Inno; Liu, Shengnan; Vukčević, Vuko; Jasak, Hrvoje
• Is Part Of: Ocean engineering. Volume 196(2020)
• Publication Date: 2020-01-15
• Subjects: Compressible two phase model -- Finite Volume -- Pressure based -- Wave impact -- Green sea -- Ghost Fluid Method; Ocean engineering -- Periodicals; Ocean engineering; Periodicals; 620.4162
• DOI: 10.1016/j.oceaneng.2019.106773
• Languages: English
• ISSNs: 0029-8018

Abstract

Abstract: An efficient, pressure-based segregated Finite Volume solution method for two-phase free surface flow simulations including compressibility effects is presented in this paper. Incompressible treatment of the heavier phase enables efficient long evolutions of wave fields without the need for a separate solver. Air is treated as an ideal gas undergoing isentropic compression/expansion, removing the need for an additional energy equation, and related numerical difficulties. The discontinuity in properties at the free surface between the two phases is treated using the Ghost Fluid Method, correctly accounting for the abrupt change in density and compressibility. Detailed verification and validation is conduced on three simple test cases comprising a liquid piston, free fall impact of a horizontal water column and a regular wave propagation to test the stability and accuracy for cases with and without significant compressibility effects. The present approach is compared with the incompressible formulation for industrial-grade simulations, showing that the same level of accuracy can be achieved without significant overhead in computational time. Finally, a compressible wave breaking impact from a large-scale experimental campaign is reproduced, showing that the method is capable of capturing trapped air cushioning effects with good accuracy. Highlights: Two-phase flow model with compressible gas phase is developed. The liquid phase is modelled as incompressible. Volumetric compressibility effects are correctly accounted for. No overhead in CPU resources is observed comparing to an incompressible model. The method is suitable for general industrial-grade simulations in marine hydrodynamics.