Development of optimal decoking scheduling strategies for an industrial naphtha cracking furnace system

Abstract

A naphtha cracking furnace produces various petrochemical products ranging from ethylene to pitch. Continuous operation of the furnace leads to coke formation, in increasing amounts, on the inner surface of the cracking coils. The coke deposits inside the coils decrease the productivity of the furnace, and therefore, its operation is periodically stopped for decoking to restore the productivity. An industrial furnace system has many furnaces in parallel, which share a common naphtha feed flow. This leads to an optimization problem in which the decoking schedule for a set of parallel furnaces must be decided simultaneously along with the distribution of the inlet naphtha feed flow among them. The optimization problem can be formulated as an MINLP, which may be computationally intractable for an industrially sized problem. Three alternative solution strategies have been developed to circumvent the inevitable nonlinear terms in the objective function. The performance of each strategy is presented in terms of the solution quality and computational time.