Loading...

Plantwide Control Structure Design for Steam Methane Reforming Process Integrated with Carbon Capturing Unit

Mohammadnejad, Pouria | 2025

0 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 58228 (06)
  4. University: Sharif University of Technology
  5. Department: Chemical and Petroleum Engineering
  6. Advisor(s): Vafa, Ehsan; Panahi, Mehdi; Arjomandzadeh, Alireza
  7. Abstract:
  8. In this research, we present a process in which a CO₂ capture unit is optimally integrated into a steam methane reforming unit. Because the reforming unit economically depends on the downstream process, it is studied in the context of methanol production. Furthermore, the energy required for the capture unit is supplied by this unit. A plantwide control structure is also designed to restore near-optimal conditions during anticipated disturbances. This study focuses on an industrial methanol production unit in southern Iran. A wide range of criteria are used to achieve optimal operating conditions, but in practice, most control structures are designed primarily for stability and disturbance rejection. We investigated various objective functions, such as maximizing methanol production, carbon efficiency, and variable income (with and without carbon tax). We obtained optimal operating conditions and the control structure design through steady-state optimization. Therefore, determining the best objective function and designing self-optimizing control required a comprehensive steady-state simulation. The large-scale integrated process and its multiple tear streams resulted in time-consuming simulations with numerous convergence issues. Consequently, machine learning-based techniques were employed. Results showed that maximizing methanol production is the least effective optimization objective, causing profitability reductions of 13.51% and 18.53% in the Iranian and global markets, respectively. Similarly, carbon efficiency led to profit reductions of 11.95% and 11.04%, demonstrating the superiority of variable income as an objective. Furthermore, the methanol maximization objective resulted in further profit losses of 3.01% and 9.30% from excessive hydrogen recycle to the hydrogenation reactor inlet. Conversely, carbon efficiency and variable income functions yielded higher profitability. Ultimately, using the variable income objective, a decentralized comprehensive control structure with 94 loops was designed. This structure incorporates self-optimizing control to ensure continuous optimality and stability against predicted disturbances. We also examined how using combined measured variables as controlled variables influenced the maximum optimality deviation. Results indicate that using two process variables as controlled variables instead of one reduces this deviation by about 83%
  9. Keywords:
  10. Carbon Tax ; Self-Optimizing Control ; Machine Learning ; Control Structure ; Carbon Dioxide Absorption ; Plantwide Control Design ; Steam Reforming

 Digital Object List

 Bookmark

  • part one
  • pdf without controllers
  • pdf with controllers
  • part four