Demonstration of non-linear model predictive control of post-combustion CO2 capture processes (2018)

S.O. Hauger a, N. Enaasen Fløb, H. Kvamsdal c, F. Gjertsen a, T. Mejdell c, M. Hillestad d,

aCybernetica AS, Leirfossv. 27, 7038 Trondheim, Norway bTechnology Centre Mongstad (TCM), Mongstad 71, 5954 Mongstad, Norway cSintef AS, P.O. Box 4760 Torgarden, 7465 Trondheim, Norway dNorwegian University of Science and Technology (NTNU), N-7494 Trondheim, Norway *Corresponding  author
0098-1354/© 2018 Published by Elsevier Ltd.

Nonlinear model predictive control applications have been deployed on two large pilot plants for post combustion CO2 capture. The control objective is formulated in such a way  that the CO2 capture ratio is controlled at a desired value, while the reboiler duty is formulated as an unreachable maximum constraint. With a correct tuning, it is demonstrated that the controllers automatically compensate for disturbances in flue gas rates and compositions to obtain the desired capture ratio while the reboiler duty is minimized. The applications are able to minimize the transient periods between two different capture  rates with the use of minimum reboiler duty.

This article is behind a paywall. For futher information:

Demonstrating flexible operation of the Technology Centre Mongstad (TCM) CO2 capture plant (2019)

Mai Buia,b, Nina E. Fløc, Thomas de Cazenovec, Niall Mac Dowella,b,*

aCentre for Process Systems Engineering, Imperial College London, South Kensington, London SW7 2AZ UK bCentre for Environmental Policy, Imperial College London, South Kensington, London SW7 1NA UK cTechnology Centre Mongstad (TCM), 5954 Mongstad Norway *Corresponding author
International Journal of Greenhouse Gas Control 93 (2020) 102879
1750-5836/ © 2019 Elsevier Ltd. All rights reserved.

This study demonstrates the feasibility of flexible operation of CO2 capture plants with dynamic modelling and experimental testing at the Technology Centre Mongstad (TCM) CO2 capture facility in Norway. This paper presents three flexible operation scenarios: (i) effect of steam flow rate, (ii) time-varying solvent regeneration,  and (iii) variable ramp rate. The dynamic model of the TCM CO2 capture plant developed in gCCS provides further insights into the process dynamics. As the steam flow rate decreases, lean CO2 loading increases, thereby reducing CO2 capture rate and decreasing absorber temperature. The time-varying solvent regeneration scenario  is demonstrated successfully. During “off-peak” mode (periods of low electricity price), solvent is regenerated, reducing lean CO2 loading to 0.16 molCO2/molMEA and increasing CO2 capture rate to 89–97%. The “peak” mode (period of high electricity price) stores CO2 within the solvent by reducing the reboiler heat supply and increasing solvent flow rate.

During peak mode, lean CO2 loading increases to 0.48 molCO2/molMEA, reducing CO2 capture rate to 14.5%, which in turn decreases the absorber temperature profile. The variable ramp rate scenario demonstrates that different ramp rates can be applied successively to a CO2 capture plant. By maintaining constant liquid-to-gas (L/G) ratio during the changes, the CO2 capture performance will remain the same, i.e., constant lean CO2 loading (0.14–0.16 molCO2/molMEA) and CO2 capture rate (87–89%).

We show that flexible operation in a demonstration scale absorption CO2 capture process is technically feasible. The deviation between the gCCS model and dynamic experimental data demonstrates further research is needed to improve existing dynamic modelling software. Continual development in our understanding of process dynamics during flexible operation of CO2 capture plants will be essential. This paper provides additional value by presenting a comprehensive dynamic experimental dataset, which will enable others to build upon this work.

This article is behind a paywall. For futher information:

Experimental results of transient testing at the amine plant at Technology Centre Mongstad: Open-loop responses and performance of decentralized control structures for load changes (2018)

Rubén M. Montañésa,, Nina E. Fløb, Lars O. Norda

aDepartment of Energy and Process Engineering, NTNU Norwegian University of Science and Technology, Kolbjørn Hejes v. 1B, 7491, Trondheim, Norway b Technology Centre Mongstad, 5954, Mongstad, Norway Corresponding author
International Journal of Greenhouse Gas Control 73 (2018) 42–59
Available online 10 April 2018
1750-5836/ © 2018 Elsevier Ltd. All rights reserved.

Flexible operation of combined cycle thermal power plants with chemical absorption post combustion CO2 capture is a key aspect for the development of the technology. Several studies have assessed the performance of decentralized control structures applied to the post combustion CO2 capture process via dynamic process si- mulation, however there is a lack of published data from demonstration or pilot plants. In this work, experiments on transient testing were conducted at the amine plant at Technology Centre Mongstad, for flue gas from a combined cycle combined heat and power plant (3.7–4.1 CO2 vol%). The experiments include six tests on open- loop responses and eight tests on transient performance of decentralized control structures for fast power plant load change scenarios.

The transient response of key process variables to changes in flue gas volumetric flow rate, solvent flow rate  and reboiler duty were analyzed. In general the process stabilizes within 1 h for 20% step changes in process inputs, being the absorber column absorption rates the slowest process variable to stabilize to changes in reboiler duty and solvent flow rate. Tests on fast load changes (10%/min) in flue gas flow rate representing realistic load changes in an upstream power plant showed that decentralized control structures could be employed in order to bring the process to desired off-design steady-state operating conditions within (< 60 min). However, oscillations and instabilities in absorption and desorption rates driven by interactions of the capture rate and stripper temperature feedback control loops can occur when the rich solvent flow rate is changed significantly and fast as   a control action to reject the flue gas volumetric flow rate disturbance and keeping liquid to gas ratio or capture rate constant.

This article is behind a paywall. For futher information:

Scale-up and Transient Operation of CO2 Capture Plants at CO2 Technology Centre Mongstad (2014)

G.M. de Koeijer, Statoil ASA, K.I. Aasen, Statoil ASA, E.S. Hamborg, TCM DA

The CO2 Technology Centre Mongstad (TCM) is the world’s largest facility for testing and improving technologies for CO2 capture. The knowledge gained will prepare the ground for full scale CO2 capture initiatives to combat climate change. TCM is a joint venture between the Gassnova, Statoil, Shell and Sasol. It is located at the West coast of Norway, north of the city Bergen. This paper will discuss the scale-up and transient operation of amine based post-combustion CO2 capture plants in general, and presents some typical results. Scale-up and transient operation are typically among the last topics to be assessed in the technology development process because it requires bigger plants.

Results from the monoethanolamine (MEA) campaign that was executed in fall/winter 2013/2014 were used. Normalized transient data were presented for 7 important variables during a plant stop and restart and a sudden stop case. Stable CO2 product flow could be obtained after 3-4 hours, while stable emissions and CO2 product temperature took 1-2 hours more. NH3 emissions showed a peak after restart due to accumulation in the solvent during the stop. It was concluded that amine based CO2 capture plants should be able to follow their power plants without significant additional CO2 emissions. Furthermore, the discussion on scale-up showed that the process of upscaling is ongoing and that emissions, material choice, construction method, vapour/liquid distribution and reclaiming are important technical aspects of this process. The main non-technical learning for efficient upscaling is to systematically learn from previous projects on how to build and operate cheaper.

This article is behind a paywall. For futher information:

Collection overview: Research for more than 10 years