
The webinar will discuss the implementation of a deep learning-based optimization framework on a four-tower natural gas liquid (NGL) fractionation system. In the NGL application, total front-end charge could be limited by flooding of any of the four towers in series. The interaction of feed effects with those of reflux and duty on the tower flooding and product qualities allowed us to demonstrate dynamic control on a nonlinear and highly coupled system with over 50 total variables. By sharing deployment insights and learned lessons, speakers will demonstrate how AI-based advance processed controls (APC) and real-time optimization solutions can add value above and beyond traditional APC methods.
Presenters
Rob Christenson, CEO, oPRO.ai
Tim Spinner, VP, Product Management and APC, oPRO.ai
Roberto Linares, VP, AI Solutions, oPRO.ai
Sriram Ramaganesan, Director, APC, Blending and Optimization, Phillips 66
Moderator
Atique Malik, CEO and Founder, AIControl LLC