Abstract. Two dynamic mesh methods-Radial Basis Function (RBF) method and Delaunay method which have been used widely in fluid-structure interaction recently were compared. Primarily, the basic theory of these two methods and the dynamic mesh process were presented. Furthermore, using the variable bandwidth sparse matrix storage technique, the computational efficiency of the RBF method can be improved. Taking ONERA M6 wing as an example, the mesh quality and computation time of the mesh motion after the airfoil rotation and wing deformation were checked. The results show that RBF method generate a better quality mesh in both structure and unstructured mesh while Delaunay method performed high efficiency, especially when the node number of the wing surface is large.

A reduce points algorithm was presented and applied to radial basis function dynamic mesh method. According to the given error, the algorithm had chosen some points form all the points on the CFD surface mesh. By interpolating an approximately surface on the selected points, the algorithm fill the drawback of radial basis function dynamic mesh method by reducing the computation time and storage. Taking m6 wing as an example, the simulation results show the initial point's insensitivity and the high efficiency of this algorithm. The algorithm can reduce the computation time of the dynamic mesh several times or even more. This property would be helpful in reducing the nonlinear aero elasticity computation time and storage.