Numerical modelling and experimental validation of squeezing flows in the automobile production

Abstract

This paper presents a study that combines experimental and numerical approaches to investigate hybrid joining methods involving riveting and adhesive bonding, aimed at enhancing mechanical properties by leveraging their advantages while addressing their limitations. The primary focus of this research was to address the issue of reducing adhesive pockets that commonly occur in such processes. To address this, an experimental setup was developed, enabling precise and reproducible measurements of deflections using digital image correlation (DIC). This allowed for accurate assessment of the extent of adhesive pockets. The device was designed, fabricated, and effectively validated through testing. The findings revealed that employing thicker sheets led to a reduction in adhesive pockets due to minimized sheet deformation and improved adhesive flow. Conversely, increased lengths resulted in the opposite effect. A comparison of materials with the same thickness but varying yield strength indicated minor differences in deflections. Subsequently, the experimental results were subjected to numerical modelling using two distinct methodologies. The first approach employed a Coupled Euler–Lagrange (CEL) framework, while the second employed Smoothed Particle Hydrodynamics (SPH). Prior to application, both methods were validated through squeeze-flow experiments. The numerical modelling utilizing the CEL approach outperformed the SPH method, offering a more accurate estimation of deflections. Finally, predictions of deflections were generated using a simplified analytical approach that incorporated the intricate rheological behaviour of the adhesive, accounting for loading-rate effects. This yielded satisfactory agreement between the predicted and experimental data, particularly at the centre of the specimen, with some deviation towards the edges.