2-D Adaptive Weighting in Collocated Discrete Least Square Meshless Method

Adaptive analysis is an important research subject in numerical methods. In recent decades, many research activities in this field have been performed. The ultimate goal of this analysis is to obtain the numerical solution with the desired accuracy and minimum computational cost. The adaptive refinement process consists of two main sections: the first one is error estimation and the second one is refinement procedure. One of the benefits of Meshless techniques is network flexibility in the adaptive refinement process. Common methods in refinement include moving and adding the new nodes. This paper presents a low-cost technique called adaptive weighing to raise the accuracy of the answers on the Meshless Collocation Discrete Least Square (CDLS). In the presented method, the remains are employed in order to estimate error and also used in definition of weighting function.