An Improved RBF Based Differential Quadrature Method

Xu Liang,Tao Wang,Decai Huang,Zhen Liu,Ronghua Zhu,Chizhong Wang

DOI: https://doi.org/10.1016/j.enganabound.2021.11.023

IF: 3.25

2021-01-01

Engineering Analysis with Boundary Elements

Abstract:Differential quadrature (DQ) based on radial basis function (RBF) was provided by Shu [1] as a meshless method which discretizes any derivative by a weighted linear summation of functional values at randomly distributed particles. However, in the RBF-DQ technique by Shu, the particles are treated as fixed interpolation points and these points cannot move with fluid. Therefore, the conventional RBF-DQ technique is an Euler mesh-free method and lack of the Lagrangian characteristics. In this paper, RBF-DQ is improved to achieve the movement of particles with fluid. Inspired by smoothed particle hydrodynamics (SPH) technique, the particle shifting technology is applied into RBF-DQ to prevent scarce discretization error in the stretching direction and particle clustering within the compressing direction. Boundary and virtual particles are created to remedy particle deficiency near the boundary. An appropriate interpolation method is developed to solve the information of virtual particles and boundary in the Lagrangian frame. The findings indicated that the improved RBF-DQ is able to achieve Lagrangian characteristics, and has good numerical characteristics in terms of accuracy and convergence.