Optimization of thick sheet clinching tools using principal component analysis

A new optimization method based on the principal component analysis (PCA) is presented for the identification of geometry parameters in tool design for mechanical clinching process of thick sheets. By applying PCA in joining, numerous dependent variables relevant for quality can be approximated with a very low number of statistical eigenmodes. Subsequently, the eigenmodes are used instead of the actual dependent variable for mathematically characterising the entire joint. This method offers the possibility of forming a direct functional relation between the tool parameters and the adherend geometries. Thus, the quality of a joint can be optimized by a generic algorithm (GA).