Modeling And Optimization For Ethylene Cracking Furnace Systems Feed Scheduling

An ethylene plant employs multiple cracking furnaces in parallel to convert various hydrocarbon feedstocks to smaller hydrocarbon molecules.The continuous operational performance of cracking furnaces gradually decays because of coke formation in the reaction coils, which requires each furnace to be periodically shut down for decoking.Decoking process will generate a lot of CO and dust and other contaminants.For the problem of ethylene cracking furnace systems feed scheduling,we should not only consider the complexity of cracking、the process of coking reaction,but also consider the distribution of a variety of materials、the operating of cracking furnace systems and the corresponding costs of decoking,so this is a complex optimization problem.In this research, I combine practical and theoretical to solve the problem of modeling and optimization for ethylene cracking furnace systems feed scheduling.Considering the mechanistic models can not be applicated online effectively,to solve the problem of yield modeling,first of all a yield model was developed for ethylene&propylene based on PSO-SVM, and the decaying performance of ethylen yield is combined.Finally the yield modeling in the cycle time was obtained. And a new MINLP (mixed-integer nonlinear programming) model has been developed to obtain feed scheduling strategies for cracking furnace systems. At the same time, a new emission constraint was added to solve the serious problem of too much CO and dust and other pollutants produced during the coke cleaning process. The optimization results indicate that by limiting the coke cleaning time to a in a specified range, the pollutants are much less and the social benefit is much improved.Finally, the yield model is applied to the objective function of ethylene cracking furnace systems feed scheduling model, solving the problem of multiple feeds cracked in multiple furnaces feed scheduling and optimize the total decoking time. Case studies demonstrate the efficacy of the developed methodology.