Valuable insights through virtual plant representation
The digital twin consists of multiple elements which are virtual representations of various plant sub-units or single devices. It will be continuously (24 hours/day, 365 days/year) updated with measured data from the plant and will process this data in near real-time and run in parallel to the actual plant providing virtual monitoring, prediction and optimization functions. If a process cannot be simulated directly (white box approach), neural networks can be used for an accurate prediction.