Modeling And Optimization Of Scheduling For Ethylene Cracking Furnace System

The cracking furnace is a core equipment during ethylene production. Industrial production of ethylene usually uses multiple cracking furnaces to process hydrocarbon feedstock in parallel. The hydrocarbon feedstock is broken down into small molecule hydrocarbons during the cracking process, such as ethylene and so on. During the operation of the cracking furnace, coke will be inevitably formed in the inside wall of furnace tubes, which will lead to a decrease in ethylene yield. Therefore, regular shutdowns of cracking furnace for decoking operation are needed for the industrial production. When multiple cracking furnaces can process various types of raw materials wtih the changing of operation cost and product yield, how to schedule the operation of the furnace system to maximize the profit is an important issue. On the other hand, ethylene cracking furnace consumes large amount of energy and massive resources and it is also the key target of environmental protection, especially in decoking stage, the furnace will emit large quantities of carbon dioxide, carbon monoxide and dust. As energy conservation and emission reduction is more and more important in China, how to use the new methods and technologies to achieve emission reduction in scheduling for ethylene cracking furnace system is a significant issue in industrial production.In order to solve the problem of profit optimization, an ethylene yield model of different kinds of feeds cracked in various furnaces under different operation conditions is firstly established. Based on this relationship model, with consideration of raw material balance constraints, a new model is raised, which combines feedstock and the load changing of each feed. Aiming at maximizing the average profit of cracking furnace system, the new MINLP model is solved by DICOPT solver in GAMS. Then the type of raw materials cracked in each cracking furnace and the flowrate of each raw material are obtained. Compared with the former scheduling model with fixed furnace load, the new model overcomes the shortage that only raw material selection is considered and that furnace load decision only depends on the operation experience. As a result, the new model improves the profit of ethylene production.On the problem of giving consideration to both the profit and pollutants emission, on the basis of cracking process mechanism, a model of furnace tube coking amount changes with processing time is established. Then, based on the coking amount model, a new scheduling model for furnace system aiming at minimizing the emmison of pollutants is raised. The new model takes into account the problems of furnace tube coking amount changing and pollutants emmison, which were ignored in previous studies. The optimization results provide the scheduling schem with minimum average amount of coke, compared with the result of model with the goal of maximum profit, the new method is able to effectively reduce the amount of coke in each ton of ethylene products, with a small amount of profit cost as sacrifice. In this way, pollutant emissions during decoking stage is reduced and substantial environmental benefits are obtained.