This content is not included in your SAE MOBILUS subscription, or you are not logged in.
Parameter Sensitivity Study of Self-piercing Rivet Insertion Process using Finite Element and Machine Learning Method
ISSN: 0148-7191, e-ISSN: 2688-3627
To be published on April 14, 2020 by SAE International in United States
Self-piercing rivets (SPR) are efficient and economical joining methods for lightweight automotive body structure manufacturing. Finite element method (FEM) is a potential effective way to assess joining process while some uncertain parameters can be employed in the simulation based on the prior knowledge, which could lead to significant mismatches between CAE predictions and physical tests. Thus, a sensitivity study on critical CAE parameters is important to guide the high-fidelity modeling of SPR insertion Process. In this paper, a 2-D symmetrical CAE model is constructed to simulate the insertion process of the SPR using LS-DYNA/explicit. Then, several surrogate models are trained using machine learning methods to build the linkage between selected inputs (e.g. material properties, interfacial frictions, 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 CAE simulations with a limited sampling volume. Based on trained surrogate models, an extensive sensitivity study is conducted to thoroughly understand the impact of a collection of CAE parameters. This research provides a solid foundation for meta-modelling and the CAE model calibration for SPR insertion process.