An implicit ghost-cell immersed boundary method for simulations of moving body problems with control of spurious force oscillations
Highlights
► An immersed boundary method for accurate and stable solutions of moving body problems is developed. ► A mass source/sink algorithm and a fully-implicit time-integration scheme are combined. ► New immersed boundary method provides enhanced control of spurious force oscillations. ► A backward time-integration scheme for multiple fresh cell treatment is developed.
Introduction
In recent years, immersed boundary methods have received special attention for simulations of fluid flow in complex configurations or over moving bodies [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [18], [19], [20]. Compared to methods using body-conforming grids, immersed boundary methods usually utilize a simple Cartesian grid and impose necessary boundary conditions on non-grid conforming surfaces of immersed bodies by assigning proper body forces. The use of an immersed boundary method significantly reduces the computational cost and time by avoiding repeated generation of body-fitted grids during moving-body simulations.
Immersed boundary methods are categorized into two major groups [21]. Methods in the first group utilize continuous body forces [2], [3], [4], [5], [6], [7], [8], which are added to the non-discretized Navier–Stokes equations. The continuous forcing immersed boundary methods have been popular for simulations of the interaction between fluid flow and elastic immersed structures. However, it is reported that representation of sharp boundaries and rigid objects are difficult for immersed boundary methods based on continuous forcing because of the restriction imposed by numerical stability [21]. In order to maintain the numerical stability, the continuous forcing methods model interfaces between the fluid and objects with diffuse boundaries and allow a relatively small time-step size for integration of the governing equations.
Methods in the second group correspond to discrete forcing immersed boundary methods and impose discrete body forces on the computational cells close to the surfaces of immersed bodies to realize necessary boundary conditions [1], [9], [10], [11], [13], [14], [15], [16], [18], [19], [20]. Immersed boundary methods in this category allow to use a larger time-step size than that allowed by continuous forcing immersed boundary methods and, with a proper treatment of discrete forcing, permit sharp representation of immersed boundaries [18], [20].
However, discrete forcing immersed boundary methods are generally known to suffer from the generation of spurious force or pressure oscillations on the surface of a moving immersed body [1], [15], [16], [20]. Uhlmann [15] attempted to reduce spurious force oscillations by combining a direct forcing method of Fadlun et al. [10] and discrete delta functions of Peskin [2]. To control the spurious force oscillations during a moving body simulation, Yang and Balaras [16] proposed an elaborate extrapolation technique, which assigns the velocity and pressure to grid cells where solid becomes fluid due to a body motion. Later, Yang and Balaras’ extrapolation procedure was simplified by Yang and Stern [17]. Recently, Lee et al. [1] identified two major sources of spurious force oscillations through a numerical experiment. One source is related to the spatial discontinuity in the pressure field across the immersed boundary where grid cells, which were previously inside an immersed body, become fluid cells due to a body motion. The other source of spurious force oscillations is associated with the temporal discontinuity in the velocity field where fluid cells turn into solid cells with a body motion. Lee et al. also found that spurious force oscillations tend to be reduced by decreasing the grid spacing and increasing the time-step size. Seo and Mittal [20] identified the primary cause of spurious force oscillations as the non-conservation of fluid mass near the immersed boundary, which is a result of the violation of geometric conservation across the immersed boundary. Seo and Mittal reported that a combination of a sharp-interface immersed boundary method and a cut cell method significantly improves the geometric conservation, thereby reducing spurious force oscillations.
Motivated by the recent understanding of sources of spurious force oscillations, in the present study, a new immersed boundary method which provides highly enhanced capability for controlling the generation of spurious force oscillations in moving-body simulations, is developed. The present method is based on a ghost-cell method [18] coupled with a mass source/sink algorithm [11]. A mass source/sink method is used to better conserve the fluid mass in the vicinity of immersed boundaries, thereby reducing one of the sources of spurious force oscillations [1], [20].
Unlike the original ghost-cell method [18], in the present method, the governing equations are integrated in time using a fully implicit method. A fully-implicit time integration of the governing equations allows to use a broad range of ratios of the time-step size to the grid spacing, thereby providing enhanced control of spurious force oscillations. However, a special treatment for multiple layers of fresh cells, due to the use of a large time-step size, is required. To effectively and accurately deal with multiple layers of fresh cells, a novel fresh-cell treatment algorithm based on a backward time-integration scheme is also developed. As will be shown in the following sections, the present fully-implicit ghost-cell immersed boundary method coupled with a mass source/sink algorithm is found to effectively suppress spurious force oscillations during simulations of flow over moving bodies.
In the following section, the present numerical methods including a fully-implicit time-integration method, a ghost-cell immersed boundary method with mass sources/sinks, and a backward time-integration method for assigning velocity vectors to fresh cells, are described in detail. In Section 3, results from test simulations for demonstrating the spatial and temporal accuracy of the method and the effectiveness of the new methodology for reducing spurious force oscillations, are presented. Finally, concluding remarks are followed in Section 4.
Section snippets
Fully-implicit time-integration method
The magnitude of spurious force oscillations on immersed boundaries is reported to be proportional to , where and are the grid spacing and the time-step size, respectively [1]. Using explicit or semi-implicit time-integration schemes limits the permissible range of the time-step size to the grid spacing or the Courant–Friedrichs–Lewy (CFL) number due to a stability requirement. In the present study, to extend the operational range of the time-step size to the grid spacing, thereby
Decaying vortices
To verify the spatial and temporal accuracy of the present numerical method, two-dimensional decaying vortices is simulated. The flow field consists of an array of vortices and the analytical solution for those vortices is as follows:The Reynolds number of this flow is 100 based on the initial maximum velocity and the size of a vortex L. The computational domain size is . Periodic boundary conditions
Concluding remarks
A novel computational methodology for simulations of flow over moving bodies on a Cartesian grid has been developed. The new method is based on a fully-implicit time integration of the discretized incompressible Navier–Stokes equations coupled with a ghost-cell immersed boundary method [18] and a mass source and sink method [11]. The present immersed boundary method allows to realize sharp interfaces between the fluid and bodies, and most importantly, provides a capability for controlling the
Acknowledgments
The authors acknowledge the support of the Office of Naval Research under Grant No. N000141110652, with Dr. Ki-Han Kim as the program manager.
References (29)
- et al.
Sources of spurious force oscillations from an immersed boundary method for moving-body problems
Journal of Computational Physics
(2011) Flow patterns around heart valves: a numerical method
Journal of Computational Physics
(1972)Numerical analysis of blood flow in the heart
Journal of Computational Physics
(1977)- et al.
Modeling a no-slip flow boundary with an external force field
Journal of Computational Physics
(1993) - et al.
An immersed boundary method with formal second-order accuracy and reduced numerical viscosity
Journal of Computational Physics
(2000) - et al.
The immersed interface method for the Navier-Stokes equations with singular forces
Journal of Computational Physics
(2001) - et al.
Combined immersed-boundary finite-difference methods for three dimensional complex flow simulations
Journal of Computational Physics
(2000) - et al.
An immersed-boundary finite-volume method for simulations of flow in complex geometries
Journal of Computational Physics
(2001) - et al.
A ghost-cell immersed boundary method for flow in complex geometry
Journal of Computational Physics
(2003) Modeling complex boundaries using an external force field on fixed Cartesian grids in large-eddy simulations
Computers and Fluids
(2004)
An immersed boundary method with direct forcing for the simulation of particulate flows
Journal of Computational Physics
An embedded-boundary formulation for large-eddy simulation of turbulent flows interacting with moving boundaries
Journal of Computational Physics
A simple and efficient direct forcing immersed boundary framework for fluid–structure interactions
Journal of Computational Physics
A versatile sharp interface immersed boundary method for incompressible flows with complex boundaries
Journal of Computational Physics
Cited by (90)
A versatile sharp boundary ghost-node method for moving rigid boundary fluid flow with meshless nodes distribution
2024, Engineering Analysis with Boundary ElementsA GPU accelerated three-dimensional ghost cell method with an improved implicit surface representation for complex rigid or flexible boundary flows
2024, Computer Physics CommunicationscDFIB: A convolutional direct forcing immersed boundary method for solving incompressible flows with time-varying geometries
2023, Journal of Computational PhysicsThe immersed boundary method: A SIMPLE approach
2023, Journal of Computational PhysicsA GPU-accelerated sharp interface immersed boundary method for versatile geometries
2023, Journal of Computational Physics