| Abstract Scope |
The autonomy revolution in manufacturing requires that traditional mass production can be transformed into personalized production. Personalization involves low volume production of high-value, smart parts with embedded sensors and software, and requires highly-flexible manufacturing processes. Thus, flexible forming such as incremental forming gained interest in the last decade due to its dominant advantages over the conventional processes in customization and personalization. This variability is driven by novel parameterization methods of the designs resulted from utilization of topology optimizations. Incremental sheet forming is a suitable technology for adapting these continuous changes with a fast and cost-effective turn around. Moreover, there is a growing interest in data-driven methods to assist forming processes for in-situ corrections of errors. Highly accurate methods requires large data-sets training through either physical testing or high-fidelity simulations. In this context, we will discuss data-driven hybrid algorithms for defect predictions in incremental forming and their significance in attaining autonomy. |