基于金属塑性成形数值模拟结果的创新知识挖掘方法研究

Research on Methodology in Discovering Innovative Knowledges from the Numerical Simulation Results of Metal Forming Process

提出了从金属塑性成形数值模拟结果中挖掘创新型知识的方法,分析了实现的步骤。论述了建立有限元数值模拟结果数据模型的集成方法、粗糙集和主成分分析法等实现技术,利用Matlab语言建立了知识发现原型系统。给出了分析的实例,验证了提出的方法。基于金属塑性成形数值模拟结果的知识挖掘方法拓展了获取新知识的途径。

A new methodology was proposed to dis-cover the knowledges from the numerical simulation re-sults of metal forming process, and the implementationprocedures were also analyzed in detail. The paper main-ly elaborates the integrated method of the data model forthe FEM numerical simulation results, and the specificapplications of Rough Set Theory (RST) and PrincipalComponent Analysis (PCA). The knowledge discoverysystem was also developed using Matlab tool. Finally,the case study of an automotive oil tank forming was per-formed to testify the proposed methods. The methodolo-gy of discovering knowledges from numerical simulationresults of metal forming extends the ways to capture theknowledges.