A Ghost Cell Method On Adaptive Cartesian Grids For Moving Boundary

In this thesis, the Ghost Cell Method on Adaptive Cartesian Grids is developed to solve the inviscidcompressible unsteady flows involving moving boundary. The research work includes the followingparts:First, the Ghost Cell Method on Cartesian Grids for static boundary is studied, including Symmetricaltechnique and Ghost Body-Cell Method (GBCM). Furthermore, two ghost cell techniques arepresented to treat the cells along moving boundary——moving boundary symmetrical technique andmoving boundary pressure gradient method. In the moving boundary pressure gradient method, theflow variables of ghost cells can be obtained from a mirror flow extrapolation combined with gradsextrapolation of a given solution over a moving reflecting boundary.Then, a computer code based on finite volume scheme and Adaptive Cartesian Grids is developed tosolve two-dimensional inviscid unsteady Euler equation, by using HLLC Riemann Solver, togetherwith MUSCL reconstruction and MINMOD Limiter. The dual time-stepping for explicit multistageschemes is employed, in which four-stage Runge-Kutta method is used in the pseudo-time.Finally, the examples of one-dimensional and two-dimensional moving wall are used to validate thesecond order accuracy and conservation of proposed method. Then, the problem of cylinder lift-off bya strong shock wave is simulated. The problems of static wedge in supersonic flow and supersonicmoving wedge are also simulated and the comparison is made. Finally, the simulation of NACA0012oscillating airfoil is carried out. All the results obtained clearly demonstrate the robustness,conservation and high accuracy of the present Cartesian Ghost Cell Method.