Abstract
The control of polymer quality has become increasingly important as the production of polymer grades has increased in quantity and diversity. On a typical polymerization plant there is an extruder downstream of the reactor. This paper provides details on the use of the operating variables of such an extruder as an empirical sensor for the melt flow index, an important indicator of polymer quality. An empirical model was built in several stages. First a simple model was built which related the melt index with the extruder pressure. After many iterations a model which also included extruder speed and temperature compensation and a bias updating procedure was developed. The final bias updated model has been installed at the plant and detects a change in the melt index nine (9) minutes before the online instrument. The end goal is to use this soft sensor to build a data based plant model, and subsequently use this model to compute optimal grade transition trajectories in the plant.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
