(414f) Dynamic Modeling and Soft Sensor Based Predictive Control of a Moving Bed Process for CO2 Capture Using a Micro-Encapsulated Solvent

Microencapsulation of carbon sorbents (MECS) is a novel technique where a solvent for CO₂ capture is encapsulated inside a CO₂-permeable polymer material [1]. MECS can enable use of solvents that have excellent CO₂ capture properties but suffer from undesirable characteristics like high viscosity, or solids formation upon CO₂ capture, that make transport difficult and use of conventional adsorption towers impractical. Using microcapsule sizes ranging from 100 – 600 , MECS can provide very high specific surface area thus offering excellent mass and heat transfer. One of the key difficulties of using the MECS is water management. Water is also known to permeate through polymers that can be used for encapsulation. Ineffective water management can lead to undesirable solvent concentrations, excess energy penalties, and varying capsule sizes, all of which penalize performance. Therefore, efficient water management is one of the critical requirements for commercial use of the MECS technology. While there are a few studies that have investigated different reactor configurations for MECS [2-3], no studies have been conducted on water management for MECS. While control of water content within the MECS capsule is required to ensure optimal performance, it is not practical to measure water concentration inside the microcapsules. In this work, we propose a computationally based soft sensor for estimating water concentration in the solvent.

Moving bed (MB) reactors have been reported to exhibit strong potential for solid-sorbent based CO₂ capture [4-5]. Therefore, in this work, the MB technology is evaluated for MECS. A detailed first-principles dynamic model of the MB absorber and regenerator is developed using Na₂CO₃ as the encapsulated solvent. The capsule model used in the moving bed reactor is validated with experimental data from our previous study [3]. The soft sensor for estimating water concentration in the solvent is developed to infer difficult to measure water content by using the data of the easily measured variables [6]. State space models for both absorber and desorber are used to estimate the capsule H₂O concentration.

Finally, linear and nonlinear model predictive controllers for maintaining CO₂ capture, desorber temperature and minimizing water loss are developed and evaluated by simulating a number of disturbances. The performance of the predictive controllers are compared with conventional PID controllers. It is observed that while model predictive controllers without the soft sensor exhibit excellent tracking and disturbance rejection performance, they can lead to high energy loss. Predictive controllers with the soft sensor for estimating core concentration at the outlet of the absorber and desorber results in much lower energy usage mainly due to reduced water evaporation in the desorber.

References

1. Stolaroff, J. K. et al., Microencapsulation of advanced solvents for carbon capture. Faraday Discussions. DOI: 10.1039/c6fd00049e (2016).
2. Hornbostel, et al, Packed and fluidized bed absorber modeling for carbon capture with micro-encapsulated sodium carbonate solution, Applied Energy 1192-1204, 2019.
3. Kotamreddy, et al., Process Modeling and Techno-Economic Analysis of a CO2 Capture Process Using Fixed Bed Reactors with a Microencapsulated Solvent, Energy & Fuels 2019 33 (8), 7534-7549 DOI: 10.1021/acs.energyfuels.9b01255.
4. Knaebel, K.S., Temperature swing adsorption system. 2009, Google Patents.
5. Mondino, Giorgia et al. “Effect of Gas Recycling on the Performance of a Moving Bed Temperature-Swing (MBTSA) Process for CO₂ Capture in a Coal Fired Power Plant Context.” Energies, 10(6), 745. 2017.
6. Fortuna, L. et al., Soft sensors for monitoring and control of industrial processes. Springer. 2007.