Prediction of the Form of a Hardened Metal Workpiece during the Straightening Process

In industry, metal workpieces are often heat-treated to improve their mechanical properties, which leads unwanted deformations and changes in geometry. Due high hardness (60 HRC or more), conventional bending rolling straightening approaches not effective, as a failure of the material occurs. The aim research was develop predictive model that predicts change form hardened workpiece function arbitrary set strikes deform surface plastically. A large-scale laboratory experiment carried out database 3063 samples prepared, based on controlled application plastic high-resolution capture different types input data, describing, one hand, performed workpieces, other hand point cloud geometry, were combined appropriately into is suitable for U-Net convolutional neural network. model’s performance investigated using three statistical indicators. These indicators were: relative absolute error (RAE), root mean squared (RMSE), (RSE). results showed had excellent prediction performance, with values RMSE less than 0.013, RAE 0.05, RSE 0.004 test data. Based results, we concluded proposed could be useful tool designing an optimal strategy high-hardness workpieces. Our will open doors implementing digital sustainability techniques, since more efficient handling result fewer subsequent heat treatments shorter times. An important goal reduce electricity consumption production, this approach certainly do.