Two-dimensional linear elasticity by the boundary node method

This paper presents a further development of the Boundary Node Method (BNM) for 2-D linear elasticity. In this work, the Boundary Integral Equations (BIE) for linear elasticity have been coupled with Moving Least Square (MLS) interpolants; this procedure exploits the mesh-less attributes of the MLS and the dimensionality advantages of the BIE. As a result, the BNM requires only a nodal data structure on the bounding surface of a body. A cell structure is employed only on the boundary in order to carry out numerical integration. In addition, the MLS interpolants have been suitably truncated at corners in order to avoid some of the oscillations observed while solving potential problems by the BNM (Mukherjee and Mukherjee, 1997a). Numerical results presented in this paper, including those for the solution of the Lame and Kirsch problems, show good agreement with analytical solutions. (C) 1998 Elsevier Science Ltd. All rights reserved.