近似移动最小二乘(AMLS)方法和其应用.pdfVIP

Abstract AMLS Method And Application Approximate Moving Least Square Approximation Method And Application Abstract Approximate Moving Least Square Approximation Method(AMLS) is a new Meshless Method,it has the advantage of high accuracy, easy programming ,etc. In this paper, we introduced the theory of AMLS methods and given two AMLS methods for solving some differential equations. We extends the two methods to the elliptic variational inequality problems,which namely lubrication problem of cylindrical manifold and the static elastic-plastic torsion problems. The numerical results verify the feasibility and accuracy of the methods. The main work is as follows: 1. The basic principle and the construction of high-order generating functions for AMLS approximation methods is introduced. Then a more accuracy methods, named Iterated AMLS Methods is introduced. Finally,numerical examples show the efficiency of the two AMLS Methods. 2. Combined the AMLS Methods with Collocation Methods,Least-squares Methods and Galerkin Methods.,it yields AMLS Collocation Methods, AMLS-LS Methods and AMLS-Galerkin Methods.Using these AMLS methods,we solved some differential problems, it reveals the effectiveness of this methods. Finally,it discusses the effect of several parameters,different basic function,the number of nodes the value of parameters ,etc.we proved the high- accuracy of this methods. 3. It described the basic model of the lubrication problem defined on the cylindrical manifold and the static elastic-plastic torsion problem