Fuzzy pattern modeling of self-pierce riveting for data from experiments and computer simulations

Abstract Modeling forms the basis for optimal control of complex technical processes in context industry 4.0 development and, hence, high product quality as well efficient production. For mechanical joining process self-pierce riveting with 11 input and 5 output variables, two modeling approaches based on (1) experimental data (2) FEM computer simulation are outlined performed. A physical approach is ruled out due to problem dimensionality nonlinear dynamic relationships between variables. Alternatively, data-based lead Artificial Intelligence (AI) model designs. The cost- resource-consuming; therefore, only a relatively small set can be collected. Here, we present results from trials that serve representatives generalized by description high-dimensional parametric membership functions (fuzzification). fuzzification procedure also applied results. In principle, it provide an arbitrarily large database. However, consequently, time- computational effort increase considerably. Both sets form parallel building using AI method local fuzzy pattern models, which used describe highly input-output error-minimizing partitioning. Finally, comparison outlined. coupled strategy future adaptation proposed.