Performance evaluation of the fast model predictive control scheme on a CO2 capture plant through absorption/stripping system
Introduction
Clean, efficient, and economical energy has been a long-standing desire to address the huge challenges posed by equipment safety, climate change, and global warming. Natural gas is a fossil fuel that comprises a significant amount of methane, as well as contaminants such as mercury, water vapours, organic sulphur, inert gases, carbon dioxide (CO2), and hydrogen sulphide (Alcheikhhamdon and Hoorfar, 2016). Due to the depletion of pure natural gas reserves, the existence of such pollutants presents a challenge. Contaminated natural gas engenders serious risks to human health and equipment safety, including lung and skin illness, corrosion, fouling, and unexpected shutdowns (Sanni et al., 2020). The high CO2 content of natural gas diminishes the heating value and burning rate of the gas mixture, resulting in decreased engine output (Ayandotun et al., 2012). Thus, efficient control strategies are paramount to ensure the sweet gas stream CO2 compositions are always maintained within the desired values (Lee and Kim, 2020, Park et al., 2021b).
The major and frequently employed CO2 capture technologies include absorption/stripping (Gutierrez et al., 2018), adsorption (Tao et al., 2019), cryogenic distillation (Babar et al., 2019), membrane separation (Harrigan et al., 2020), chemical looping combustion (CLC) (Khan et al., 2020), biological separation (Matito-Martos et al., 2020), hydrate separation (Kida et al., 2020) etc. Above all, chemical absorption with a monoethanolamine (MEA) solvent is the utmost established and commonly used CO2 capture technology at the industrial scale due to its high CO2 capture efficiency, fast reactivity, low cost, abundant solvent, uncomplicated retrofitting, and capability to handle gases having low CO2 partial pressure (Akinola et al., 2018; Tay et al., 2017). However, the complexity of the capturing system creates significant difficulties in the form of variable interactions, nonlinearities, constraints in the capture plant (Li et al., 2018). As a result, the development of an effective, adaptive, and quick responding control strategy capable of overcoming the inherent challenges in CO2 capture plants is critical for cost-effective CO2 capture.
The fundamental objectives of the control strategy are to maintain the CO2 capture rate at set points without offsets and to prevent the stripper temperature from exceeding the desired value, which might accelerate the degradation of the MEA solvent (Mechleri et al., 2017). Most of the control strategies are based on PID and MPC-based controllers to ensure the controllability and flexibility of the CO2 capture plant. The PI-based decentralized control scheme has been implemented on an MEA-based CO2 capture plant integrated with a power plant. In a study by Nittaya et al. (Nittaya et al., 2014a), the controller performance is evaluated by introducing sinusoidal changes in the gas composition and CO2 capture rate with the control objective of CO2 capture rate and purity at 87% and 95%, respectively. Another work by Cristea et al. (Cristea et al., 2020) implemented a decentralized PI controller with the control purpose of maintaining CO2 capture rate and energy performance while rejecting disturbances and tracking set points. Depending on the amount of disturbance/change, the controller acquired the new setpoints in 2–6 h. However, due to the delayed response time, high overshoot, and large offsets, a more effective control technique capable of resolving such issues is required.
The large variety of implementations of MPC-based control schemes on absorption/stripping systems has been noticed in the literature due to the benefits of handling uncertainties, variable interactions, processing with multiple-input multiple-output (MIMO) systems (Cormos et al., 2015, He et al., 2016, Rúa et al., 2021). In MPC, a model is utilized to predict the future behavior of the system. The basic schematic diagram of the MPC is illustrated in Fig. 1 (Tahir et al., 2018).
Cormos et al. (Cormos et al., 2015) implemented MPC control scheme in MATLAB® on a CO2 capture plant based on an absorption/stripping system integrated with a power plant. Step, ramp, and sinusoidal changes in inputs are introduced to control the CO2 removal efficiency. The amount of liquid flow is adjusted to maintain the CO2 removal efficiency to 87%. In comparison to a PI-based control scheme, the MPC controller has expressed better performance in reducing overshoot and controlling the constraints. He et al. (He et al., 2016) also designed an MPC control structure for a CO2 capture system to evaluate its controllability and flexibility. The performance of the MPC controller is determined under sudden changes for load and setpoint tracking scenarios. The results illustrated that the MPC scheme exhibits minimum variation from the set points (CO2 removal/product purity) as compared to the PI controller. However, the response of such conventional/classical MPC-based (CMPC) controllers is still on the slow side compared to more advanced MPC-based control strategies. It can be improved by fine-tuning the input/output weights or by using fast-responding control techniques such as the Fast MPC (FMPC) control scheme.
The control algorithm of CMPC is solved in the form of a Quadratic Program (QP). The QP comprises a system matrix with complex dynamics of the CO2 capture plant. The solution of such a complex QP requires a large settling time, which results in a slow output response especially for complicated systems having large dimensions (Qi et al., 2015). The computational burden of the controller can be reduced through an advanced control strategy termed as FMPC controller, which solves the QP fragmentally into small fractions and accelerates calculation time in control action through online optimization (Blanchard and Adegbege, 2017; Kouzoupis et al., 2015; Mönnigmann and Kastsian, 2011; Quirynen et al., 2015). The FMPC-based approaches have been widely applied to traffic, electrical and electronic systems including management of freeway traffic congestion (Han et al., 2017), energy storage system (Ni et al., 2021), turbine generator system (Hou et al., 2020), multilevel power converters (Ramirez et al., 2020), Jump Markov Systems (Tonne and Stursberg, 2017), solar systems (Pipino et al., 2020) etc. Qi et al. (Qi et al., 2015) established a FMPC control approach on a DC motor to demonstrate the MIMO feasibility of a fast dynamic system with 1-bit signal control. On another work, (Schindele and Aschemann, 2011) a nonlinear FMPC method was developed for the two primary axes of an overhead travelling crane, ensuring required trajectory tracking and active dampening of the crane load and oscillations, respectively.
However, FMPC approach has not been implemented on CO2 capture plants due to the challenges of identified model’s accuracy and presence of offsets. Since the key variables to be controlled in the system is the amount of CO2 in the outlet gas stream, which is critical to ensure that it is always at the desired specifications under any disturbances/set-point changes, a fast-acting controller is clearly beneficial. Therefore, in this paper, the implementation of the FMPC on a standard CO2 capture plant via absorption/stripping system is investigated and gauged against the CMPC. Based on the authors’ knowledge, this is the first study related to the implementation of the FMPC controller on a CO2 capture plant based on an absorption/stripping system.
Section snippets
Process Case Study
For the control analysis and comparison, a standard MEA-based CO2 capture system is considered. The CO2 capture process based on absorption/stripping system consists of mainly an absorber, stripper, heat exchanger; alongside other units such as mixers, pump, a splitter etc. The sour gas comprising natural gas and CO2 flows into the absorber from bottom to top at a temperature of 25 °C. The monoethanolamine (MEA) solvent enters from the top of the absorber at the same temperature. The CO2 gas is
Steady state results
The CO2 gas from the sour gas is absorbed into the MEA solvent through interaction between lean solvent and sour gas inside the absorber. As the rate-based modelling approach is not supported in Aspen PLUS®, the absorber and stripper are designed through the adjustment of certain parameters to attain the maximum CO2 capture rate. The absorber is designed with 5 stages that are sequentially numbered from top to bottom. The MEA rich stream, termed as lean solvent, are introduced into the absorber
Conclusions
The goal of this work is to assess the performance of the Fast Model Predictive Control (FMPC) on a dynamically complex CO2 capture plant based on an absorption/stripping system. To generate the input-output data, the steady-state and dynamic model of the capture plant is designed in Aspen PLUS®, and the optimal state-space model is identified using MATLAB® System Identification Toolbox. To examine the performance, the FMPC controller is implemented on the model in MATLAB® with the single and
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgment
The authors like to express their appreciation to the Ministry of Higher Education (MOHE), Grant Reference Code FRGS/1/2018/TK02/UTP/02/5 for the funding provided for this research. The authors would also like to extend their appreciation to Yayasan University Technology PETRONAS (YUTP) (Grant 015LC0-139) for additional funding provided for this work.
References (75)
- et al.
Natural gas quality enhancement: A review of the conventional treatment processes, and the industrial challenges facing emerging technologies
J. Nat. Gas. Sci. Eng.
(2016) - et al.
Efficient CO2 capture using NH2–MIL–101/CA composite cryogenic packed bed column
Cryogenics
(2019) - et al.
An SOR-Like Method for Fast Model Predictive Control
IFAC-Pap.
(2017) - et al.
Dynamic modelling and optimisation of flexible operation in post-combustion CO2 capture plants—A review
Comput. Chem. Eng.
(2014) - et al.
Computer-based optimization of acid gas removal unit using modified CO 2 absorption kinetic models
Int. J. Greenh. Gas. Control
(2017) - et al.
LQR based improved discrete PID controller design via optimum selection of weighting matrices using fractional order integral performance index
Appl. Math. Model.
(2013) - et al.
Observer and controller design for a methane bioconversion process
Eur. J. Control
(2021) - et al.
Model predictive control and its application in agriculture: A review
Comput. Electron. Agric.
(2018) - et al.
Controllability and flexibility analysis of CO 2 post-combustion capture using piperazine and MEA
Int. J. Greenh. Gas. Control
(2016) - et al.
System identification methods for dynamic models of brain activity
Biomed. Signal Process. Control
(2021)
Steady-state energy optimization and transition assessment in a process of CO2 absorption from natural gas
Energy
Resolving freeway jam waves by discrete first-order model-based predictive control of variable speed limits
Transp. Res. Part C: Emerg. Technol.
Development and implementation of advanced control strategies for power plant cycling with carbon capture
Comput. Chem. Eng.
Dynamic modeling and advanced control of post-combustion CO2 capture plants
Chem. Eng. Res. Des.
Dynamic modelling of a commercial-scale CO 2 capture plant integrated with a natural gas combined cycle (NGCC) power plant
Int. J. Greenh. Gas. Control
Flexible operation and simultaneous scheduling and control of a CO2 capture plant using model predictive control
Int. J. Greenh. Gas. Control
Controllability and optimal scheduling of a CO 2 capture plant using model predictive control
Int. J. Greenh. Gas. Control
Fuzzy modeling and fast model predictive control of gas turbine system
Energy
Dynamic analysis and linear model predictive control for operational flexibility of post-combustion CO2 capture processes
Comput. Chem. Eng.
Model-free adaptive control for MEA-based post-combustion carbon capture processes
Fuel
Application of piece-wise linear system identification to solvent-based post-combustion carbon capture
Fuel
Control for Carbon Dioxide Exchange Process in a Membrane Oxygenator Using Online Self-Tuning Fuzzy-PID Controller
Biomed. Signal Process. Control
Stabilized Prediction Error Method for Closed-loop Identification of Unstable Systems
IFAC-Pap.
Process control strategies for flexible operation of post-combustion CO 2 capture plants
Int. J. Greenh. Gas. Control
Fast explicit MPC with multiway trees
IFAC Proc. Vol.
Enhancing resilience of DC microgrids with model predictive control based hybrid energy storage system
Int. J. Electr. Power Energy Syst.
Dynamic modelling and control of MEA absorption processes for CO2 capture from power plants
Fuel
Process integration and design for maximizing energy efficiency of a coal-fired power plant integrated with amine-based CO2 capture process
Appl. Energy
A robust nonlinear model predictive controller for a post-combustion CO2 capture absorber unit
Fuel
A novel fast model predictive control with actuator saturation for large-scale structures
Comput. Struct.
Nonlinear temperature regulation of solar collectors with a fast adaptive polytopic LPV MPC formulation
Sol. Energy
A fast MPC algorithm for reducing computation burden of MIMO
Chin. J. Chem. Eng.
Fast Model-based Predictive Control (FMPC) for grid connected Modular Multilevel Converters (MMC)
Int. J. Electr. Power Energy Syst.
Data preprocessing & evaluation used in the microplastics identification process: A critical review & practical guide
TrAC Trends Anal. Chem.
The removal of CO2 and N2 from natural gas: A review of conventional and emerging process technologies
J. Pet. Sci. Eng.
Dynamic modelling, simulation and basic control of CO2 absorption based on high pressure pilot plant for natural gas treatment
Int. J. Greenh. Gas. Control
An overview on control strategies for CO2 capture using absorption/stripping system
Chem. Eng. Res. Des.
Cited by (13)
Performance of industrially implemented turbulent contact absorbers – A comparative study
2024, Chemical Engineering Research and DesignProcess intensification in bio-jet fuel production: Design and control of a catalytic reactive distillation column for oligomerization
2023, Chemical Engineering and Processing - Process IntensificationA hybrid receding horizon optimization and active disturbance rejection control of boiler superheated steam temperature
2023, Process Safety and Environmental ProtectionWaste oleaster seed-derived activated carbon mixed with coarse particles of fluid catalytic cracking as a highly-efficient CO<inf>2</inf> adsorbent at low temperatures
2023, Process Safety and Environmental ProtectionSaturation of the MEA solution with CO<inf>2</inf>: Absorption prototype and experimental technique
2023, Results in Engineering