Improving data-driven machine digital twins by including engineering insights

Europe should push to reduced energy consumption dramatically in the upcoming decade, as outlined in the recently proposed European Green Deal. A more efficient exploitation of the existing machines in industry can have a major impact here. To enable this effective allocation without disrupting the current production cycles, a digital twin framework allows to optimize the machine settings with respect to production quality and energy consumption. However, in practice only limited data is available to feed these digital twins, which typically leads to poor accuracy. In this project we will explore a novel framework which employs data-driven methods in combination with common engineering insights and models to ensure more robust and predictive digital twins for these production machines. This framework will allow production facilities to map the performance of their machines with respect to the various settings in order to optimize the overall production lines with multiple interacting assets.

Please follow and like us: