Optimization of Semi-Lean Amine Design for Post-Combustion CO2 Capture Processes By Applying Simulation and Neural Network

CO₂ is the major greenhouse gas that is one of the root cause of the global climate change. The emission of CO₂ from fossil fuel combustion and industry processes contribute 78% of total greenhouse gas emission due to the globally economic and population growth. Therefore, the continuous technology advancement for CO₂ capture are crucial to reduce global greenhouse gas emissions as well as global climate changes. Amine-based post combustion CO₂ capture process has a long application history and has been recognized as a mature technology for coal-based power plant.

However, this technology is always complained by its high energy cost. The design of a side draw semi-lean amine from the stripper to absorber is a promising way to reduce the energy consumption and thus to improve the performance of the post combustion CO₂ capture process. So far, few studies provide a systematic method for the relevant optimal design. This study employs rigorous steady-state model,neural network model and optimization technology to find the optimal tray position, temperature, and flowrate of the side-draw design for semi-lean amine . First, by investigating the solvent on each stage of stripper and the liquid/gas composition on each stage of absorber, the possible selection range of the design parameters for side-draw design could be roughly identified. Second, multiple simulations will be conducted to obtain the process simulation data within the range, then train a neural network model by those data. Third, optimize the neural network model and get the optimal design parameters. Finally, by fine tuning those parameters in the rigorous simulation, the optimal design will be identified.