Tuesday, April 12, 2016: 11:21 AM
336B (Hilton Americas - Houston)
Olefins are the most important product with the largest bulk productivity in the petrochemical industry. They include ethylene, propylene, and butadiene, which are produced from large hydrocarbon molecules through heat cracking in ethylene plants. There are two ways to increase the efficiency and profit of an ethylene plant. One is by technology and facility improvement. The other is by scheduling optimization, since an ethylene plant employs multiple cracking furnaces to realize above cracking reactions and how to schedule the production of an olefin furnace system is a complex problem. In past studies, scheduling optimization mainly focuses on how to maximize the overall profit of the furnace system with consideration of various constrains such as non-simultaneous decoking, ethane recycling, recycling ethane relay cracking, etc. The object of the optimization does not consider the product ratios between ethylene, propylene and butadiene. However, when an ethylene plant is established, the product ratios are already fixed with little variation. The downstream facilities cannot separate C1 to C4 by a random ratio. Thus, there is ratio requirement in the cracking products. In this study, a MINLP model is developed for an olefin furnace system scheduling with the consideration of downstream separation. The object is still to maximize the total profit of the furnace system, but limit the products within a range of ratio. The efficacy of the model is demonstrated by case studies.