Advanced process control (APC) applications require accurate inputs for stream qualities, including product compositions and distillation curve points. While online analyzers are effective in supporting model predictive controllers, they are expensive and often unavailable for important process streams. There is a more cost-effective approach for building virtual analyzers or inferentials to accurately estimate stream qualities and it starts with digital twins.
In this on-demand webinar experts presented a semi-rigorous methodology applied to examples from the refining industry. Learn how you can leverage a unit digital twin to:
- Extract chemical and physical parameters to build simplified core equations for the inferential
- Quickly deploy inferentials without significant training for initial data reconciliation
- Apply to other process use cases to build accurate inferentials, while keeping costs down
In addition, we shared customer case studies and their results.