Exploiting data in smart factories : Real-time state estimation and model improvement in metal forming mass production

摘要: Modern production systems have numerous sensors that produce large amounts of data. This data can be exploited in many ways, from providing insight into the manufacturing process to facilitating automated decision making. These opportunities are still underexploited metal forming industry, due complexity these processes. In this work, a probabilistic framework is proposed for simultaneous model improvement and state estimation mass production. Recursive Bayesian used simultaneously track evolution estimate deviation between physics-based real process. A sheet bending test framework. metamodel built using proper orthogonal decomposition radial basis function interpolation. The extended with order account difference Particle filtering parameters simultaneously. approach tested analysed number simulations, based on pseudo-data obtained numerical model.