石油炼制与化工 ›› 2019, Vol. 50 ›› Issue (5): 87-92.

• 控制与优化 • 上一篇    下一篇

重油催化裂化十二集总动力学模型研究

汪伟1,王智峰2,欧阳福生1,李盾1,侯凯军2,阳斯拯1   

  1. 1. 华东理工大学化工学院石油加工所
    2. 中国石油重质油加工重点实验室
  • 收稿日期:2018-09-05 修回日期:2018-11-07 出版日期:2019-05-12 发布日期:2019-05-28
  • 通讯作者: 欧阳福生 E-mail:ouyfsh@ecust.edu.cn
  • 基金资助:
     

TWELVE-LUMP KINETIC MODEL FOR HEAVY OIL FLUID CATALYTIC CRACKING

    

  1.  
  • Received:2018-09-05 Revised:2018-11-07 Online:2019-05-12 Published:2019-05-28
  • Supported by:
     

摘要: 根据催化裂化反应机理和产物分布特点,建立了包含54条虚拟反应路径的重油催化裂化12集总反应网络。以Davison Circulating Riser(DCR)试验装置数据为基础,基于Python平台,将模型数学方程转化为程序语言,采用四阶Runge-Kutta 法求解模型微分方程、BFGS法优化目标函数,求取了模型的动力学参数。采用小型实验数据验证模型动力学参数,结果表明主要产品产率的计算值与实验值之间的相对误差均小于5%。说明所建模型的动力学参数是可靠的,能较好地反映重油催化裂化的反应规律,可用于对实际生产过程进行模拟优化。

关键词: 重油催化裂化, 集总, 动力学模型, 参数估算

Abstract: Based on the catalytic cracking reaction mechanism and product distribution characteristics, a 12-lump reaction network for heavy oil catalytic cracking including 54 virtual reaction paths was established. The mathematical equations of the model are transformed into a programming language with the Python platform on the basis of experiment data in Davison Circulating Riser (DCR) apparatus. The kinetic parameters of the model were calculated by the fourth order Runge-Kutta algorithm applied to sovle the model differential equations and the BFGS algorithm applied to optimize the objective function. Validation of model parameters showed that the relative errors between the calculated and experimental values for the main products were all below 5%. This indicates that the kinetic parameters of the model are reliable, and can better reflect the reaction rules for heavy oil catalytic cracking. The model can be further used to simulate and optimize practical production processes.

Key words: RFCC, lump, kinetic model, parameter estimation

中图分类号: