Knowledge discovery from finite element simulation data

Knowledge-based engineering (KBE) and finite element analysis (FEA) have been used widely in sheet metal forming area. However, the acquisition of knowledge keeps bottleneck when building knowledge base in KBE. To properly understand the results of the FEA and consequently choose the appropriate design, a lot of knowledge and experience are needed. FEA can generate massive data, in which large amounts of usefully implicit knowledge is hidden. Thus, knowledge acquisition from them is prospective to ease the above difficulties by applying knowledge discovery in databases (KDD) technology. In this study, the characteristics of the FEA data are discussed firstly. Then a framework of knowledge discovery from FEA data is proposed. Correspondingly, a data-mining algorithm named fuzzy-rough algorithm is developed to deal with the FEA simulation data. Finally, the stamping process of a square-cup part was studied as an example. The proposed knowledge discovery process is applied to obtain some useful, implicit production rules with efficiency measure. The result shows that knowledge discovery from FEA simulation data is valuable.