09:30 〜 09:45
[SVC29-03] Development of multiphase CFD model based on Eulerian approach for gas-solid simulations: Applications in Volcanology
キーワード：Numerical simulation, Multiphase model, gas-solid simulation, Volcanic eruption, Pyroclastic density current
The flow dynamics of volcanic eruptions (volcanic buoyant plume and pyroclastic density current) consisting of disperse solid in continuum gas phase is highly complex due to its multiphase and multicomponent nature and transient process. Multiphase numerical models (Neri et. al. 2003; Carcano et. al., 2013; Carcano et. al., 2014; Valentine GA, 2020) have been successfully employed for the study of this flow dynamics. In this approach the gas and solid are treated as separate interpenetrating continuum phases (two-fluid model) with constitutive empirical coupling equations for momentum and energy exchanges between the two phases, respective pressure for individual phases and gas pressure as the interface pressure in the solid phase equations (Gidaspow, 1994; Mao et. al. 2003; Hudson and Harris 2006; Houim and Oran 2016;). However, previous studies either did not consider a concrete definition of the solid pressure (Neri et. al., 2003; Carcano et. al., 2013; Carcano et. al., 2014) or neglected the effect of the compressibility of the gas in the simulations (Valentine GA, 2020). So, the current study envisions the development of a compressible multiphase CFD numerical tool including the solid pressure and the effect of the compressibility of the gas in the formulations to simulate flow dynamics associated with volcanic eruptions.
This study report discusses the development of the numerical tool and results from 1D simulation (Sod shock tube test) and 2D simulation (Mach disk formation in an under-expanded supersonic jet) conducted for the purpose of validation. The current numerical setup adopted the AUSM+up for both the gas and the solid phases suitable for two-fluid model compressible multiphase flow (Kitamura et. al., 2014), a simplified model for solid pressure and solid sound speed (Mao et. al. 2003), second order MUSCL interpolation with slope limiters and first-order interpolation for the gas and solid phase respectively, third order TVD Runge-Kutta for the time integration. The numerical tool is used for simulation of 1D and 2D test problem in a fully explicit finite difference framework and cartesian coordinate system.
In the 1D test case the distribution of pressure and velocity for the dusty gas (Figure 1) achieved good agreement with the benchmark numerical solution of Saito (2002) which validates the 1D development. In the 2D test case, accurate capturing of the Mach disk formation, barrel shock and associated flow structures, increase of the Mach disk height with the increase in pressure ratio and the decrease of the maximum Mach number and Mach disk height with the increase in solid volume fraction at a fixed pressure ratio (Figure 1) shows qualitative agreement with Carcano et. al. (2013) except the Mach disk height is overpredicted compared to previous literature. However, considering the good agreement of the numerical results in 1D and flow structures captured in 2D simulations, it seems that the simulations are strongly dependent on choice of boundary conditions in 2D simulation. Thus, the developed numerical tool produces promising results in the preliminary validation of 1D and 2D test simulations and thus encourages further development and application specific flow simulations associated with volcanic eruptions.
This study report discusses the development of the numerical tool and results from 1D simulation (Sod shock tube test) and 2D simulation (Mach disk formation in an under-expanded supersonic jet) conducted for the purpose of validation. The current numerical setup adopted the AUSM+up for both the gas and the solid phases suitable for two-fluid model compressible multiphase flow (Kitamura et. al., 2014), a simplified model for solid pressure and solid sound speed (Mao et. al. 2003), second order MUSCL interpolation with slope limiters and first-order interpolation for the gas and solid phase respectively, third order TVD Runge-Kutta for the time integration. The numerical tool is used for simulation of 1D and 2D test problem in a fully explicit finite difference framework and cartesian coordinate system.
In the 1D test case the distribution of pressure and velocity for the dusty gas (Figure 1) achieved good agreement with the benchmark numerical solution of Saito (2002) which validates the 1D development. In the 2D test case, accurate capturing of the Mach disk formation, barrel shock and associated flow structures, increase of the Mach disk height with the increase in pressure ratio and the decrease of the maximum Mach number and Mach disk height with the increase in solid volume fraction at a fixed pressure ratio (Figure 1) shows qualitative agreement with Carcano et. al. (2013) except the Mach disk height is overpredicted compared to previous literature. However, considering the good agreement of the numerical results in 1D and flow structures captured in 2D simulations, it seems that the simulations are strongly dependent on choice of boundary conditions in 2D simulation. Thus, the developed numerical tool produces promising results in the preliminary validation of 1D and 2D test simulations and thus encourages further development and application specific flow simulations associated with volcanic eruptions.