An Adaptive Fuzzy Logic Finite Element Analysis

Winter Meeting, American Society of Agricultural Engineers, ASAE, 1994

MESHING
RESEARCH
CORNER

Agricultural Engineering Department
Purdue University
West Lafayette, Indiana 47907-1146 USA

Abstract
An adaptive finite element analysis with fuzzy knowledge processing techniques is proposed to solve 2-D elasticity problems. Based on heuristic knowledge, experience and ad hoc methods, the fuzzy expert system makes an expert decision about the initial mesh design and its density distribution in the problem domain by considering the geometric information, as well as the boundary and loading conditions. According to the mesh density distribution, a high quality initial mesh, that is more refined around the critical points/regions, is created with an automatic mesh generator. When the error in the solution is high, the h- version adaptive refinement is used to reduce the error. The proposed approach results in lower levels and more accurate estimates of error along with faster convergence of the solution to the desired accuracy in less time and at less cost.