Effect Analysis for the Uncertain Parameters on Self-Piercing Riveting Simulation Model Using Machine Learning Model

Self-piercing rivets (SPR) are efficient and economical joining methods used in the manufacturing of lightweight automotive bodies. The finite element method (FEM) is a potentially effective way to assess the joining process of SPRs. However, uncertain parameters could lead to significant mismatches between the FEM predictions and physical tests. Thus, a sensitivity study on critical model parameters is important to guide the high-fidelity modeling of the SPR insertion process. In this paper, an axisymmetric FEM model is constructed to simulate the insertion process of the SPR using LS-DYNA/explicit. Then, several surrogate models are evaluated and trained using machine learning methods to represent the relations between selected inputs (e.g., material properties, interfacial frictions, and clamping force) and outputs (cross-section dimensions). It is found that it is feasible to train surrogate models with high accuracy to replace the time-consuming and computationally expensive CAE simulations with a limited sampling volume. Based on trained surrogate models, an extensive sensitivity study is conducted to thoroughly understand the effect of a set of model parameters. This work provides a solid foundation for data-modelling and CAE model calibration for the SPR insertion process.