Improved Test Correlation by Optimization Technique 2005-01-0830

Virtual product development by applying simulation-based design has dramatically impacted the product development process resulting in design quality improvements and reduction of cost and delivery time. Although it has been widely accepted as a standard product development tool by the engineering community, Computer Aided Engineering (CAE) still faces many challenges in improving simulation process efficiency through process integration and automation, and simulation accuracy by analytical model/physical testing correlation. To obtain a meaningful analysis result, analytical models such as finite element (FE) model should accurately represent the real world application. This is normally achieved by experimental correlation. In the current study, a simplified boundary constraint model is proposed. The analytical model update process is automated by applying optimization techniques. Certain critical issues related to selecting targets and methods are discussed and guidelines are recommended. Very good correlation results have been achieved. The method can be applied in many real world applications.