Mathematical modelling of coke formation in thermal cracking units

点击翻译按钮获取中文摘要

Thermal cracking of ethane is an important industrial process for making ethylene and other light olefins from natural gas. These olefins are used to make polyethylene and other commercially important plastics. The thermal cracking process is performed by reacting ethane and steam in metal tubes that pass through a furnace. Heat generated in the combustion chamber outside the tubes makes the cracking reactions inside the tubes occur at high rates. An undesirable side product of the thermal cracking process is coke. Coke is solid material (primarily carbon) that deposits on the inner surface of the tubes, influencing rates of chemical reactions and the distribution of chemical products that emerge from the reactor. NOVA Chemicals and other companies who design and operate thermal cracking furnaces want to have a thorough understanding of the chemical reactions and operating conditions that influence ethane cracking, coke formation and coke removal. The objective of this research is to develop and validate a model (i.e., a set of mathematical equations inside a computer program) that NOVA will use to predict coke formation and removal in industrial ethane cracking furnaces. This model will account for a detailed array of chemical reactions and will be able to predict coke accumulation in response to changes in reactor operating conditions and furnace design. Data from the literature and data generated by NOVA will be used to estimate model parameters (numbers to use in the equations) so that the model will give accurate predictions. The resulting dynamic tubular reactor model will be used by NOVA scientists and engineers to simulate and improve the operation of industrial cracking reactors, resulting in more economical production of ethylene. In addition, enhanced knowledge of the importance of the various chemical reactions that lead to coke formation will be obtained. The model will provide a valuable test problem for statistical methods developed at Queen's, leading to new insights about parameter selection and estimation methodology for complex models. A Master's student will obtain experience at NOVA and will develop modeling and statistical skills that will be of benefit to future employers.