Experimental and numerical investigations are performed to provide an assessment of the transport behavior of an ultrasonic oscillatory two-phase flow in a microchannel. The work is inspired by the flow observed in an innovative ultrasonic fabric drying device using a piezoelectric bimorph transducer with microchannels, where a water-air two-phase flow is transported by harmonically oscillating microchannels. The flow exhibits highly unsteady behavior as the water and air interact with each other during the vibration cycles, making it significantly different from the well-studied steady flow in microchannels. The computational fluid dynamics (CFD) modeling is realized by combing the turbulence Reynolds-averaged Navier-Stokes (RANS) š ā š model with the phase-field method to resolve the dynamics of the two-phase flow. The numerical results are qualitatively validated by the experiment. Through parametric studies, we specifically examined the effects of vibration conditions (i.e., frequency and amplitude), microchannel taper angle, and wall surface contact angle (i.e., wettability) on the flow rate through the microchannel. The results will advance the potential applications where oscillatory or general unsteady microchannel two-phase flows may be present.
The prediction of surface ship response to a far-field underwater explosion (UNDEX) requires the simulation of shock wave propagation in the fluid, cavitation, fluid-structure interaction, and structural response. Effective approaches to model the fluid include cavitating acoustic finite element (CAFE) and cavitating acoustic spectral element (CASE) methods. Although the spectral element method offers the potential for greater accuracy at lower computational cost, it also generates more spurious oscillations around discontinuities which are difficult to avoid in shock-related problems. Thus, the advantage of CASE remains unproven. In this paper, we present a 3D-partitioned FSI framework and investigate the application of CAFE and CASE to a surface ship early-time far-field UNDEX problem to determine which method has the best computational efficiency for this problem. We also associate the accuracy of the structural response with the modeling of cavitation distribution. A further contribution of this work is the examination of different nonmatching mesh information exchange schemes to demonstrate how they affect the structural response and improve the CAFE/CASE methodologies.
Solution of near-field underwater explosion (UNDEX) problems frequently require the modeling of two-way coupled fluid-structure interaction (FSI). This paper describes the addition of an embedded boundary method to an UNDEX modeling framework for multiphase, compressible and inviscid fluid using the combined algorithms of Runge-Kutta, discontinuous-Galerkin, level-set and direct ghost-fluid methods. A computational fluid dynamics (CFD) solver based on these algorithms has been developed as described in previous work. A fluid-structure coupling approach was required to perform FSI simulation interfacing with an external structural mechanics solver. Large structural deformation and possible rupture and cracking characterize the FSI phenomenon in an UNDEX, so the embedded boundary method (EBM) is more appealing for this application in comparison to dynamic mesh methods such as the arbitrary Lagrangian-Eulerian (ALE) method to enable the fluid-structure coupling algorithm in the fluid. Its limitation requiring a closed interface that is fully submerged in the fluid domain is relaxed by an adjustment described in this paper so that its applicability is extended. Two methods of implementing the fluid-structure wall boundary condition are also compared. The first solves a local 1D fluid-structure Riemann problem at each intersecting point between the wetted elements and fluid mesh. In this method, iterations are required when the Tait equation of state is utilized. A second method that does not require the Riemann solution and iterations is also implemented and the results are compared.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citationsācitations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.