computation about flow around a yawed cone

Tylor Xie

Dear Sir or Madam,

I have just finished my dissertation Numerical Simulation of the Evolution of Flow Patterns on the Leeward Side of a Slender Circular Cone for MSc's degree. And there is an abstract below.

My work concerned of the computation of the complex vortex patterns of a yawed cone in a supersonic flow, using the conical flow assumption. Would anyone kindly tell me the latest advance in this field, and the advance in 3-D computation of the problem, and the comparison of the two results?

Thanks. And please contact me if you have any interests.

Yours sincerely,

Tylor

ABSTRACT

The complex vortex patterns on the leeward side of a 5-deg semiangle cone over a wide range of angles of attack in a M¡Þ=1.8,Re=100 000 flow are studied, using an explicit second-order accuracy NND finite difference scheme and the conical flow model.

The effects of the grid resolution and the normal scale of the grid near the wall have been taken into account. The test computations leaded to the fact that asymmetric perturbations play a crucial role in inducing the flow asymmetries. So that an initial 2-deg or 5-deg side slip as a perturbation was introduced into the flow field in the later studies. Also, some variables, such as maxdifD ( that is, make a subtract of density at the two symmetric flow field points at left and right sides of the cone, and choose the biggest one ), and side force Cy, were set to indicate the occurrence of flow asymmetries.

The evolution path of the vortical flow was obtained, and was thoroughly depicted by crossflow contours and streamline patterns. The flow is attached at 4-deg, separated and symmetric at 8-deg, symmetric until 12-deg, becomes asymmetric at around 13-deg, more and more asymmetric, abruptly back to symmetry at 27-deg, and becomes unsteady and shedding vortices like Karman vortex street from that point on. The results are in good agreement with the topological rule and the analyses of structure stability. And they were also compared with the works of Siclari & Marconi, Kandil & etc., Thomas, Dusing & Orkwis. The sameness and differences were shown.

It was also encountered that, at the angles of attack where the flow is changing from steady state to unsteady state, the computations must be dealt with very carefully. Since two different unsteady solutions with or without vortex shedding might appear, due to the fineness of the grid or the initial perturbations. In the end, the results and the conical flow model were reviewed, and expectations for three-dimensional computations were made.