Study on Dynamic Decoking Policy for Ethylene Cracking Furnace System

Industrial ethylene cracking process employs multiple cracking furnaces in parallel to convert various hydrocarbon feedstocks to multiple smaller-molecule hydrocarbon products.  The operation is also a typical semi-continuous dynamic process, where a cracking furnace has to be periodically shut down for decoking.  Given the data of multiple feed characteristics, different furnace performances, various manufacturing costs, and specified operational constraints, an optimal decoking policy for the entire cracking system has to be determined to achieve the best economic performance of an ethylene cracking furnace system.  In practice, because of the feed supply and maintenance uncertainties, the furnace decoking activities are better to be conducted in a dynamic and reactive way, by which the furnace system operations can be timely rescheduled once an uncertainty is identified.  Thus, the feeds from the new delivery and the leftover inventories can be timely, feasibly, and optimally allocated to different furnaces for processing to obtain the maximum average net profit per day. 

Facing this challenge, this paper develops a novel reactive scheduling model to help generating dynamic decoking strategy for the entire furnace system.  It can optimize multiple-feed assignments, running length of each batch operation, and decoking sequence for every furnace in the system based on the new feed deliveries, the leftover feeds, and current furnace operating conditions.  It also simultaneously addresses major scheduling issues for real cracking furnace operations, such as non-simultaneous decoking, secondary ethane cracking, and seamless rescheduling.  The developed MINLP (mixed-integer nonlinear programming) model is solved by GAMS.  The efficacy of the study and its significant economic potential are demonstrated by comprehensive case studies.