Search

Trona, a low-cost and commonly available evaporite mineral was investigated as a novel pre-treatment material for enhancing the methane yield during the anaerobic digestion (AD) of grass silages. The results from BMP tests indicated that reactors fed with hot water pretreated Guinea grass and reactors fed with the grass pretreated with 5 mg/L and 10 mg/L Trona achieved maximum methane production rates (Umax) of 28.0, 56.4 and 58 N mL CH4/d, respectively, signifying an increase in Umax from Trona-pretreated grass by 201% and 207%, compared to the control reactors. The degradation yields in the three pairs of reactors were 68%, 79% and 75%, respectively. Kinetic modelling showed that the substrate hydrolysis rates (kh-values) were 0.32, 0.35 and 0.34 d−1, respectively, indicating that Trona-pretreatment enhanced the hydrolysis of Guinea grass. A step-change in temperature significantly affected methane yield. Kinetic models also satisfactorily predicted methane productivity in the reactors with >92% prediction accuracy. These results demonstrate that Trona pretreatment and mesophilic temperatures provided accelerated hydrolysis of biomass contents which enhanced methane production.

Rotating packed bed (RPB) system has applications in CO2 removal using chemical solvents which can reduce the size about ten times compared to common packed bed (PB) system. In this study, RPB stripper using monoethanolamine (MEA) solution is modelled in gPROMS® software. The model has been validated using experimental data from literature and show good agreement. In addition to stripper modelling and validation, the process analysis is accomplished in this study by assessing the influence of four parameters namely rotor speed, reboiler temperature, flow rate of rich liquid, and pressure on desorption efficiency and desorption energy.

There has been a shift to less carbon intensive fuels such as natural gas to meet energy demand due to increasing pressure to cut CO2 emissions. This has prompted a need to assess unconventional and contaminated natural gas reserves (which contains CO2 concentration of 20 mol% or more). The CO2 capture process with MEA as the solvent is mostly adopted to treat contaminated natural gas. In this study, the option of using a blend of ionic liquids (IL) and MEA as a promising solvent in the process was investigated through modelling and simulation. A detailed rate-based model was developed for both MEA (30 wt%) solvent and IL (30 wt%)-MEA (30 wt%) blend using Aspen Plus® to assess both process and economic performances. The 1-Butylpyridinium ([bpy][BF4]) ionic liquid was selected in this study. The physiochemical properties of [bpy][BF4], predicted using Aspen Plus®, showed good accuracy compared with experimental data. The results from this study showed about 15% and 7.44% lower energy consumption in the reboiler duty and CO2 removal cost respectively with aqueous [bpy][BF4]-MEA solvent compared to 30 wt% MEA solvent. It is concluded that the aqueous [bpy][BF4]-MEA solvent is therefore a promising solvent that could replace 30 wt% MEA solvent in this process.

Post-combustion carbon capture (PCC) via chemical absorption has the potential to reduce CO2 emission significantly. The main challenges to this technology include the high energy consumption and high capital and operating costs. Alternative process configurations and new solvents are being pursued to reduce the energy consumption process. In this study, the rate-based model of the PCC process using piperazine (PZ) was developed in Aspen Plus® V8.4. The model was validated at pilot scale with experimental data. The model correctly predicted the CO2 capture level, rich loading and reboiler duty measured in the pilot plants. The validated model was scaled up to a PCC plant capable of processing flue gas from a 250 MWe natural gas combined cycle (NGCC) power plant. Three different configurations of the large-scale PCC process using 30-40 wt% PZ were simulated and the energy consumption was evaluated to determine which combination of configuration and solvent gives the best energy performance. The energy performance of the PCC process using PZ was also compared to the standard process using 30 wt% monoethanolamine (MEA) solvent. A regeneration energy of 4.97 GJ/tonneCO2 was obtained for the PCC process with 30 wt% MEA. Compared to the lowest regeneration energy obtained using the standard PCC process with PZ, the regeneration energy reduces by 35%, 41%, and 45% respectively for 30, 35 and 40 wt% PZ concentrations. The minimum regeneration energy of 2.41 GJ/tCO2 was achieved with the advanced flash stripper (AFS) process using 40 wt% PZ.

For the flexible operation of mono-ethanol-amine-based post-combustion carbon capture processes, recent studies concentrate on model-based protocols which require underline model parameters of carbon capture processes for controller design. In this paper, a novel application of the model-free adaptive control algorithm is proposed that only uses measured input-output data for carbon capture processes. Compared with proportional-integral control, the stability of the closed-loop system can be easily guaranteed by increasing a stabilizing parameter. By updating the pseudo-partial derivative vector to estimate a dynamic model of the controlled plant on-line, this new protocol is robust to plant uncertainties. Compared with model predictive control, tuning tests of the protocol can be conducted on-line without non-trivial repetitive off-line sensitivity or identification tests. Performances of the model-free adaptive control are demonstrated within a neural network carbon capture plant model, identified and validated with data generated by a first-principle carbon capture model.

This paper presents a study on using deep learning for the modelling of a post-combustion CO2 capture process. Deep learning has emerged as a very powerful tool in machine learning. Deep learning technique includes two phases: an unsupervised pre-training phase and a supervised back-propagation phase. In the unsupervised pre-training phase, a deep belief network (DBN) is pre-trained to obtain initial weights of the subsequent supervised phase. In the supervised back-propagation phase, the network weights are fine-tuned in a supervised manner. DBN with many layers of Restricted Boltzmann Machine (RBM) can extract a deep hierarchical representation of training data. In terms of the CO2 capture process, the DBN model predicts CO2 production rate and CO2 capture level using the following variables as model inputs: inlet flue gas flow rate, CO2 concentration in inlet flue gas, pressure of flue gas, temperature of flue gas, lean solvent flow rate, MEA concentration and temperature of lean solvent. A greedy layer-wise unsupervised learning algorithm is introduced to optimize DBN, which can bring better generalization than a single hidden layer neural network. The developed deep architecture network models can then be used in the optimisation of the CO2 capture process.

Application of process intensification (PI) technologies such as rotating packed beds (RPBs) to replace packed beds (PBs) in solvent-based CO 2 capture could reduce plant footprint. Concentrated monoethanolamine (MEA) solvents are generally expected to be used in RPBs. Under this circumstance, expected temperature rise during CO2 absorption should be estimated to determine whether or not intercooling is necessary for RPBs. In this study, we demonstrated that intercooling is inevitable with RPBs using 40-70 wt% monoethanolamine (MEA) solvent through liquid phase energy balance for a hypothetical scenario. Our analysis showed that liquid phase temperature rise could be as high as 80°C in some cases and this will significantly reduce absorption rate without intercooling.

The main challenges to the commercial deployment of the solvent-based post-combustion carbon capture technology include high energy consumption and high capital and operating costs. New solvents and alternative process configurations are being pursued to reduce the energy consumption and costs of the process. This paper investigates the technical and economic performance of the process using piperazine (PZ) solvent for a 250 MWe natural gas combined cycle power plant. Three different configurations of the process using PZ were evaluated and compared to the standard process using 30 wt% monoethanolamine (MEA) solvent. The technical performance of the process was evaluated using the rate-based model developed in Aspen Plus® V8.4 and the economic evaluation was carried out in Aspen Process Economic Analyzer®V8.4. The technical results showed that the total energy demand of the process reduces from 5.34 GJ/tCO2 using 30 wt% MEA to 3.56 GJ/tCO2 using 30 wt% PZ. The lowest energy demand of 2.76 GJ/tCO2 was achieved with the advanced flash stripper using 40 wt% PZ. The economic results show that the lowest total annual cost of M$26.58 per year and CO2 capture cost of $34.65 per tonneCO2 were obtained using the advanced flash stripper (AFS) with 40 wt% PZ. It was concluded that the 40 wt% PZ solvent would bring practical technical and economic benefits to the large-scale applications of the capture process compared to the current 30 wt% MEA solution. Therefore, this study will inspire policymakers and researchers towards the large-scale deployment of solvent-based PCC process.

Process intensification (PI) technologies such as rotating packed beds (RPBs) could reduce the size of absorber used in post-combustion CO 2 capture (PCC) based on chemical absorption processes by about 12 times compared to absorber with standard packed beds. However, mass transfer correlations for predicting effective interfacial area and liquid film mass transfer coefficient in RPBs are limited in literature and their prediction accuracy against experimental data is yet to be compared. This need is addressed in this study by evaluating the performances of different correlations through comparison with experimental data. Of all the correlations assessed, it is found that Lou et al . [1] and Tung and Mah [2] correlations give reliable estimate of the effective interfacial area and liquid film mass transfer coefficients respectively.

International audience; This study investigated the prospect of using aqueous mixture of 1-butylpyridinium tetrafluoroborate ([Bpy][BF4]) ionic liquid (IL) and monoethanolamine (MEA) as solvent in post-combustion CO2 capture (PCC) process. This is done by analysis of the process through modelling and simulation. In literature, reported PCC models with a mixture of IL and MEA solvent were developed using equilibrium-based mass transfer approach. In contrast, the model in this study is developed using rate-based mass transfer approach in Aspen Plus®. From the results, the mixed aqueous solvent with 5–30 wt% IL and 30 wt% MEA showed 7%–9% and 12%–27% less specific regeneration energy and solvent circulation rate respectively compared to commonly used 30 wt% MEA solvent. It is concluded that the IL concentration (wt%) in the solvent blend have significant impact on specific regeneration energy and solvent circulation rate. This study is a starting point for further research on technical and economic analysis of PCC process with aqueous blend of IL and MEA as solvent.

Solvent-based post combustion capture (PCC) is a well-developed technology for CO2 capture from power plants and industry. A reliable model that captures the dynamics of the solvent-based capture process is essential to implement suitable control design. Typically, first principle models are used, however they usually require comprehensive knowledge and deep understanding of the process. System identification approach is adopted to obtain a model that accurately describes the dynamics between key variables in the process. The nonlinear auto-regressive with exogenous (NARX) inputs model is employed to represent the relationship between the input variables and output variables as two Multiple-Input Single-Output (MISO) sub-models. The forward regression with orthogonal least squares (FROLS) algorithm is implemented to select an accurate model structure that best describes the dynamics within the process. The prediction performance of the identified NARX models is promising and shows that the models capture the underlying dynamics of the CO2 capture process.

Power generation from coal-fired power plants represents a major source of CO2 emission into the atmosphere. Efficiency improvement and integration of carbon capture and storage (CCS) facilities have been recommended for reducing the amount of CO2 emissions. The focus of this work was to evaluate the thermodynamic performance of s-CO2 Brayton cycles coupled to coal-fired furnace and integrated with 90% post-combustion CO2 capture. The modification of the s-CO2 power plant for effective utilisation of the sensible heat in the flue gas was examined. Three bottoming s-CO2 cycle layouts were investigated, which included a newly proposed single recuperator recompression cycle. The performances of the coal-fired s-CO2 power plant with and without carbon capture were compared. Results for a 290 bar and 593 °C power cycle without CO2 capture showed that the configuration with single recuperator recompression cycle as bottoming cycle has the highest plant net efficiency of 42.96% (Higher Heating Value). Without CO2 capture, the efficiencies of the coal-fired s-CO2 cycle plants were about 3.34–3.86% higher than the steam plant and about 0.68–1.31% higher with CO2 capture. The findings so far underscored the promising potential of cascaded s-CO2 power cycles for coal-fired power plant application.

The solvent-based carbon capture process is the most matured and economical route for decarbonizing the power sector. In this process, aqueous monoethanolamine (MEA) is commonly used as the solvent for CO2 scrubbing from power plant and industrial flue gases. Generally, aqueous MEA with 30 wt% (or less) concentration is considered the benchmark solvent. The CO2 solubility data in aqueous MEA solution, used for modelling of the vapour-liquid equilibria (VLE) of CO2 in MEA solutions, are widely published for 30 wt% (or less) concentration. Aqueous MEA with higher concentrations (from 40 to 100 wt%) is considered in solvent-based carbon capture designs with techniques involving process intensification (PI). PI techniques could improve the process economics and operability of solvent-based carbon capture. Developing PI for application in capture process requires CO2 solubility data for concentrated MEA solutions. These data are however limited in literature. The modelling of the vapour-liquid equilibria (VLE) of CO2 in MEA solutions for PI-based solvent capture techniques involving stronger MEA solution of about 80 wt% concentration requires solubility data at the concentration. In this study, the data for 80 wt% MEA is presented for 40,60, 100 and 120oC. The experimental technique and analytical procedure in this study were validated by comparing the measurements for 30 wt% MEA with data from the literature. The data from this study can be fitted to VLE models such as electrolyte NRTL, extended UNIQUAC etc. which is an important component of solvent-based capture model using MEA as the solvent. More accurate VLE models will improve the prediction accuracy of capture level, rich loading etc. using PI-based solvent-based capture model.

Rotating packed bed (RPB) absorber using monoethanolamine (MEA) as the solvent to capture CO2 is modelled at steady state condition in this study according to the first principles in gPROMS®. The effect of eight different kinetic reaction models and five enhancement factors is examined based on the newly developed model. Selection of kinetic model has a significant effect on the carbon capture level (CCL) but the effect of enhancement factor relation is not important. The steady state process model is validated against the experimental data and showed good agreement. The average absolute relative deviation for 12 case-runs is 3.5%. In addition, process analysis is performed to evaluate the effect of four factors namely rotor speed, MEA concentration in lean MEA solution, lean MEA solution temperature and lean MEA solution flow rate on CCL. Finally, the orthogonal array design (OAD) method is applied to analyse the simultaneous effect of the above-mentioned factors in the CCL and motor power of RPB absorber by considering 25 scenarios. The result of using OAD revealed that rotor speed has the most important effect on CCL, and after that lean MEA solution flow rate has the second importance. In addition, the OAD method is used to find the proper combination of four factors that resulted in about 90% CCL with low motor power.

Nutrient supplementation could improve the biomethane production of different biomass feedstocks during anaerobic digestion. In this study, the impact of nutrient supplementation on the anaerobic digestion of the West African Gamba and Guinea Grass for biomethane production is presented. This was undertaken in 6 separate continuous stirred tank reactors for a hydraulic residence time of 25 days under supplementation regime with trace elements (TE), cocoa pod (CP) ash-extract, and commercial cellulase from Aspergillus niger (CCA) or Trichoderma reesei ATCC 26921 (CCT). The results showed that TE inhibits the specific methane production (SMP) with about 5% lower SMP than the control. In contrast, the other supplements namely CP, CP + CCA, CCA, and CCT + TE showed about 13, 28, 18, and 12% higher SMP than the control respectively. This study is the first demonstration of the impacts of different supplements on SMP during the anaerobic digestion of the West African Gamba and Guinea grass.

The flexible operation capability of solvent-based post-combustion capture (PCC) process is vital to efficiently meet the load variation requirement in the integrated upstream power plant. This can be achieved through the deployment of an appropriate control strategy. In this paper, a nonlinear model predictive control (NMPC) system was developed and analysed for the solvent-based PCC process. The PCC process was represented as a nonlinear autoregressive with exogenous (NARX) inputs model, which was identified through the forward regression with orthogonal least squares (FROLS) algorithm. The FROLS algorithm allows the selection of an accurate model structure that best describes the dynamics of the process. The simulation results showed that the NMPC gave better performance compared with linear MPC (LMPC) with an improvement of 55.3% and 17.86% for CO2 capture level control under the scenarios considered. NMPC also gave a superior performance for reboiler temperature control with the lowest ISE values. The results from this work will support the development and implementation of NMPC strategy on the PCC process with reduced computational time and burden.

Post-combustion carbon capture (PCC) with chemical absorption has strong interactions with coal-fired power plant (CFPP). It is necessary to investigate dynamic characteristics of the integrated CFPP-PCC system to gain knowledge for flexible operation. It has been demonstrated that the integrated system exhibits large time inertial and this will incur additional challenge for controller design. Conventional PID controller cannot effectively control CFPP-PCC process. To overcome these barriers, this paper presents an improved neural network inverse control (NNIC) which can quickly operate the integrated system and handle with large time constant. Neural network (NN) is used to approximate inverse dynamic relationships of integrated CFPP-PCC system. The NN inverse model uses setpoints as model inputs and gets predictions of manipulated variables. The predicted manipulated variables are then introduced as feed-forward signals. In order to eliminate steady-state bias and to operate the integrated CFPP-PCC under different working conditions, improvements have been achieved with the addition of PID compensator. The improved NNIC is evaluated in a large-scale supercritical CFPP-PCC plant which is implemented in gCCS toolkit. Case studies are carried out considering variations in power setpoint and capture level setpoint. Simulation results reveal that proposed NNIC can track setpoints quickly and exhibit satisfactory control performances.

Solvent-based post-combustion carbon capture (PCC) with packed column is the most commercially ready CO2 capture technology. To study commercial-scale PCC processes, validated pilot scale models are often scaled up to commercial-scale using the generalized pressure drop correlation (GPDC) chart which requires assuming the column pressure drop. The GPDC method may lead to either over-estimation or under-estimation of the column diameter. In this paper, a new method for estimating the packed column diameter without assuming the pressure drop has been proposed and used for model scale-up. The method was validated by scaling between two existing pilot plant sizes. The CO2 capture process was simulated in Aspen Plus® and validated at pilot scale. The validated model was scaled up to commercial CO2 capture plant capable of serving a 250 MWe combined cycle gas turbine power plant using the new method proposed in this study. The results obtained from the scale-up study were compared to those obtained when the GPDC method was used to design the same commercial CO2 capture plant. The results showed that the GPDC method overestimated the absorber and stripper diameter by 1.6 % and 8.5 % respectively. Process simulation results for the commercial-scale plant showed about 2.12 % and 5.63 % lower solvent flow rate and reboiler duty with the proposed method. Therefore, the capital and operating costs for the process using the newly proposed scale-up method could be lower based on our estimates of the column dimensions, solvent flow rate and specific reboiler duty.

The application of rotating packed beds (RPBs) in solvent-based carbon capture processes, will greatly reduce the physical footprint, capital and operating cost of the process. However, in designing RPBs, correlations for predicting mass transfer parameters are generally limited in literature and their prediction accuracies have not been demonstrated independently. In this paper, an RPB absorber model was developed in gPROMS ModelBuilder® and used to test and compare different correlations for predicting the effective interfacial area, liquid and gas film mass transfer coefficients. Our results showed that the modified packed column mass transfer correlations where the “g” term (i.e. gravitational acceleration) is replaced with “rw2” (i.e. centrifugal acceleration) commonly used in literature for RPBs generally give poor predictions compared to using correlations developed specifically for RPBs. Also, the Tung and Mah correlation has better predictive accuracy for the liquid film mass transfer coefficient in RPBs than more complex correlations. Finally, a set of new data for the gas film mass transfer coefficient for RPBs were also derived from overall volumetric mass transfer coefficient ( K G a ) experimental data from literature. This is the first report of gas film mass transfer data for RPBs. The results in this paper will guide researchers in selecting suitable correlations for predicting mass transfer parameters in RPBs.

Post-combustion CO2 capture (PCC) is considered the most feasible and viable process for CO2 abatement in the power sector. Aqueous monoethanolamine (MEA) solvent, traditionally used in this process, brings along challenges, namely, huge energy requirement for solvent regeneration, huge solvent flow rate leading to large equipment sizes, and chemical and thermal degradability, among others. In this study, the prospects of replacing aqueous MEA solvent with a blend of ionic liquid (IL) and MEA are explored. IL is generally chemically and thermally stable among other encouraging properties but is however expensive. A blend of IL and MEA is predicted to have shared qualities of MEA and IL and therefore could hypothetically contribute to meaningful reduction in overall cost of the process.

There is increasing need for tighter controls of coal-fired plants due to more stringent regulations and addition of more renewable sources in the electricity grid. Achieving this will require better process knowledge which can be facilitated through the use of plant models. Drum-boilers, a key component of coal-fired subcritical power plants, have complicated characteristics and require highly complex routines for the dynamic characteristics to be accurately modelled. Development of such routines is laborious and due to computational requirements they are often unfit for control purposes. On the other hand, simpler lumped and semi empirical models may not represent the process well. As a result, data-driven approach based on neural networks is chosen in this study. Models derived with this approach incorporate all the complex underlying physics and performs very well so long as it is used within the range of conditions on which it was developed. The model can be used for studying plant dynamics and design of controllers. Dynamic model of the drum-boiler was developed in this study using NARX neural networks. The model predictions showed good agreement with actual outputs of the drum-boiler (drum pressure and water level).

This paper presents a study of modelling post-combustion CO 2 capture process using bootstrap aggregated neural networks. The neural network models predict CO 2 capture rate and CO 2 capture level using the following variables as model inputs: inlet flue gas flow rate, CO 2 concentration in inlet flue gas, pressure of flue gas, temperature of flue gas, lean solvent flow rate, MEA concentration and temperature of lean solvent. In order to enhance model accuracy and reliability, multiple feedforward neural network models are developed from bootstrap re-sampling replications of the original training data and are combined. Bootstrap aggregated model can offer more accurate predictions than a single neural network, as well as provide model prediction confidence bounds. Simulated CO 2 capture process operation data from gPROMS simulation are used to build and verify neural network models. Both neural network static and dynamic models are developed and they offer accurate predictions on unseen validation data. The developed neural network models can then be used in the optimisation of the CO 2 capture process.

As post-combustion CO 2 capture (PCC) technology nears commercialisation, it has become necessary for the full carbon capture and storage (CCS) chain to be studied for better understanding of its dynamic characteristics. Model-based approach is one option for economically and safely reaching this objective. However, there is need to ensure that such models are reasonably simple to avoid the requirement for high computational time when carrying out such study. In this paper, a simplification approach for a detailed rate-based model of post-combustion CO 2 capture with solvents (rate-based mass transfer and reactions assumed to be at equilibrium) is presented. The simplified model can be used in model-based control and/or full chain CCS simulation studies. With this approach, we demonstrated significant reduction in CPU time (up to 60%) with reasonable model accuracy retained in comparison with the detailed model.

Intensified regenerator/stripper using rotating packed bed (RPB) for regeneration of rich-MEA solvent in post-combustion CO2 capture with chemical absorption process was studied through modelling and simulation in this paper. This is the first systematic study of RPB regenerator through modelling as there is no such publication in the open literature. Correlations for liquid and gas mass transfer coefficients, heat transfer coefficient, liquid hold-up, interfacial area and pressure drop which are suitable for RPB regenerator were written in visual FORTRAN as subroutines and then dynamically linked with Aspen Plus® rate-based model to replace the default mass and heat transfer correlations in the Aspen Plus®. The model now represents intensified regenerator/stripper. Model validation shows good agreement between model predictions and experimental data from literature. Process analyses were performed to investigate the effect of rotor speed on the regeneration efficiency and regeneration energy (including motor power). The rotor speed was varied from 200 to 1200 rpm, which was selected to cover the validation range of rotor speed. Impact of reboiler temperature on the rate of CO2 stripping was also investigated. Effect of rich-MEA flow rate on regeneration energy and regeneration efficiency was studied. All the process analyses were done for wide range of MEA concentration (32.6 wt%, 50 wt% and 60 wt%). Comparative study between regenerator using packed column and intensified regenerator using RPB was performed and the study shows a size reduction of 9.691 times. This study indicates that RPB process has great potential in thermal regeneration application.

This paper presents modelling of a post-combustion CO2 capture process using bootstrap aggregated extreme learning machine (ELM). ELM randomly assigns the weights between input and hidden layers and obtains the weights between the hidden layer and output layer using regression type approach in one step. This feature allows an ELM model being developed very quickly. This paper proposes using principal component regression to obtain the weights between the hidden and output layers to address the collinearity issue among hidden neuron outputs. Due to the weights between input and hidden layers are randomly assigned, ELM models could have variations in performance. This paper proposes combining multiple ELM models to enhance model prediction accuracy and reliability. To predict the CO2 production rate and CO2 capture level, eight parameters in the process were utilized as model input variables: inlet gas flow rate, CO2 concentration in inlet flow gas, inlet gas temperature, inlet gas pressure, lean solvent flow rate, lean solvent temperature, lean loading and reboiler duty. The bootstrap re-sampling of training data was applied for building each single ELM and then the individual ELMs are stacked, thereby enhancing the model accuracy and reliability. The bootstrap aggregated extreme learning machine can provide fast learning speed and good generalization performance, which will be used to optimize the CO2 capture process.

For large volumes of carbon dioxide (CO 2 ) onshore and offshore transportation, pipeline is considered the preferred method. This paper presents a study of the pipeline network planned in the Humber region of the UK. Steady state process simulation models of the CO 2 transport pipeline network were developed using Aspen HYSYS®. The simulation models were integrated with Aspen Process Economic Analyser® (APEA). In this study, techno-economic evaluations for different options were conducted for the CO 2 compression train and the trunk pipelines respectively. The evaluation results were compared with other published cost models. Optimal options of compression train and trunk pipelines were applied to form an optimal case. The overall cost of CO 2 transport pipeline network was analyzed and compared between the base case and the optimal case. The results show the optimal case has an annual saving of 22.7 M€. For the optimal case, levelized energy and utilities cost is 7.62 €/t-CO 2 , levelized capital cost of trunk pipeline is about 8.11 €/t-CO 2 and levelized capital cost of collecting system is 2.62 €/t- CO 2 . The overall levelized cost of the optimal case was also compared to the result of another project to gain more insights for CO 2 pipeline network design.

Coal-fired power plants are the main source of global electricity. As environmental regulations tighten, there is need to improve the design, operation and control of existing or new built coal-fired power plants. Modelling and simulation is identified as an economic, safe and reliable approach to reach this objective. In this study, a detailed dynamic model of a 500 MWe coal-fired subcritical power plant was developed using gPROMS based on first principles. Model validations were performed against actual plant measurements and the relative error was less than 5%. The model is able to predict plant performance reasonably from 70% load level to full load. Our analysis showed that implementing load changes through ramping introduces less process disturbances than step change. The model can be useful for providing operator training and for process troubleshooting among others.

Process intensification (PI) has the potential to significantly reduce capital and operating costs in post-combustion CO 2 capture using monoethanolamine (MEA) solvent for power plants. The intensified absorber using rotating packed bed (RPB) was modelled based on Aspen Plus ® rate-based model, but some build-in correlations in Aspen Plus ® rate-based model were replaced with new correlations suitable for RPB. These correlations reflect centrifugal acceleration which is present in RPB. The new correlations were implemented in visual FORTRAN as sub-routines and were dynamically linked to Aspen Plus ® rate based model. The model for intensified absorber was validated using experimental data and showed good agreement. Process analysis carried out indicates: (a) CO 2 capture level increases with rotating speed. (b) Higher lean MEA inlet temperature leads to higher CO 2 capture level. (c) Increase in lean MEA concentration results in increase in CO 2 capture level. (d) Temperature bulge is not present in intensified absorber. Compared with conventional absorber using packed columns, the insights obtained from this study are (1) intensified absorber using RPB improves mass transfer significantly. (2) Higher flue gas temperature or lean MEA temperature will not be detrimental to the reactive separation as such cooling duty for flue gas can be saved. (3) Inter-cooling cost will not be incurred since there is no temperature bulge. A detail comparison between conventional absorber and intensified absorber using RPB was carried out and absorber volume reduction factor of 12 times was found. These insights can be useful for design and operation of intensified absorber for CO 2 capture.

This paper presents modelling of a post-combustion CO 2 capture process using bootstrap aggregated extreme learning machine. Extreme learning machine (ELM) randomly assigns the weights between input and hidden layers and obtains the weights between the hidden layer and output layer using regression type approach in one step. This paper proposes using principal component regression to obtain the weights between the hidden and output layers. Due to the weights between input and hidden layers are randomly assigned, ELM could have variations in performance. This paper proposes combining multiple ELMs to enhance model prediction accuracy and reliability. To predict the CO2 production rate and CO2 capture level, seven parameters in the process were regarded as input variables: inlet gas flow rate, CO2 concentration in inlet flow gas, inlet gas temperature, inlet gas pressure, lean solvent flowrate, lean solvent temperature, lean loading and reboiler duty. The bootstrap re-sampling of training data was applied for each single ELM and then the individual ELMs are stacked, thereby enhancing the model accuracy and reliability. The bootstrap aggregated extreme learning machine (BA-ELM) can provide fast learning speed and good generalization performance, which will be used to optimize the CO2 capture process.

This paper presents a study of modelling post-combustion CO2 capture process using bootstrap aggregated ELMs. The dynamic ELM models predict CO2 capture rate and CO2 capture level using the following variables as model inputs: inlet flue gas flow rate, CO2 concentration in inlet flue gas, pressure of flue gas, temperature of flue gas, lean solvent flow rate, MEA concentration and temperature of lean solvent. In order to enhance model accuracy and reliability, multiple ELM models are developed from bootstrap re-sampling replications of the original training data and combined. Bootstrap aggregated ELM model can offer more accurate and reliable predictions than a single ELM model, as well as provide model prediction confidence bounds. The developed models can be used in the optimisation of CO2 capture processes.

Solvent-based carbon capture is the most commercially-ready technology for economically and sustainably reaching carbon emission reduction targets in the power sector. Globally, the technology has been deployed to deal with flue gases from large scale power plants and different carbon-intensive industries. The success of the technology is due to significant R&D activities on the process development and decades of industrial experience on acid gas removal processes from gaseous mixtures. In this paper, current status of PCC based on chemical absorption—commercial deployment and demonstration projects, analysis of different solvents and process configurations—is reviewed. Although some successes have been recorded in developing this technology, its commercialization has been generally slow as evidenced in the cancellation of high profile projects across the world. This is partly due to the huge cost burden of the technology and unpredictable government policies. Different research directions, namely new process development involving process intensification, new solvent development and a combination of both, are discussed in this paper as possible pathways for reducing the huge cost of the technology.