A local radial basis function-compact finite difference method for Sobolev equation arising from fluid dynamics

Abstract

In this article, a new high-order, local meshless technique is presented for numerically solving multi-dimensional Sobolev equation arising from fluid dynamics. In the proposed method, Hermite radial basis function (RBF) interpolation technique is applied to approximate the operators of the model over local stencils. This leads to compact RBF generated finite difference (RBF-FD) formula, which provides a significant improvement in the accuracy and computational efficiency. In the first stage of the proposed method, the time discretization is performed by Crank–Nicolson finite difference scheme along with temporal Richardson extrapolation technique. In the second stage, the space dimension is discretized by applying the local radial basis function-compact finite difference (RBF-CFD) method. By performing some numerical simulations and comparing the results with existing methods, the high accuracy and computational efficiency of the proposed method are clearly demonstrated. The numerical results show that the presented method has fourth-order accuracy in both space and time dimensions. Finally, it can be concluded that the proposed method is a suitable alternative to the existing numerical techniques for the Sobolev model.