A study on robust optimization of layered plates bonding process based on inverse analysis

Optimization problem is considered for a layered plates bonding process to reduce crack failures arising due to the residual stress at the surface of the adherent, which is caused by different thermal expansion coefficients. During the analysis of bonding process, accurate knowledge of the adhesive material, which is typically low melting point metals and undergoes extreme plastic deformation, is important factor for reliable prediction of the process. A simplified process is posed by fabricating a small specimen to simulate the bonding process, and experiment is conducted to measure the deformation using the micro-Mioré interferometry. Inverse analysis technique is employed to determine the adhesive property by closely matching the FE analysis and experiment results. Optimization study is then conducted to find out the best setup which minimizes residual stress. Robust concept is employed to minimize the mean as well as its variance.