Browsing by Author "Yan, Yongliang"
Now showing 1 - 13 of 13
Results Per Page
Sort Options
Item Open Access Applying machine learning algorithms in estimating the performance of heterogeneous, multi-component materials as oxygen carriers for chemical-looping processes(Elsevier, 2020-01-09) Yan, Yongliang; Mattisson, Tobias; Moldenhauer, Patrick; Anthony, Edward J.; Clough, Peter T.Heterogeneous, multi-component materials such as industrial tailings or by-products, along with naturally occurring materials, such as ores, have been intensively investigated as candidate oxygen carriers for chemical-looping processes. However, these materials have highly variable compositions, and this strongly influences their chemical-looping performance. Here, using machine learning techniques, we estimate the performance of heterogeneous, multi-component materials as oxygen carriers for chemical-looping. Experimental data for 19 manganese ores chosen as potential chemical-looping oxygen carriers were used to create a so-called training database. This database has been used to train several supervised artificial neural network models (ANN), which were used to predict the reactivity of the oxygen carriers with different fuels and the oxygen transfer capacity with only the knowledge of reactor bed temperature, elemental composition, and mechanical properties of the manganese ores. This novel approach explores ways of dealing with the training dataset, learning algorithms and topology of ANN models to achieve enhanced prediction precision. Stacked neural networks with a bootstrap resampling technique have been applied to achieve high precision and robustness on new input data, and the confidence intervals were used to assess the precision of these predictions. The current results indicate that the best trained ANNs can produce highly accurate predictions for both the training database and the unseen data with the high coefficient of determination (R2 = 0.94) and low mean absolute error (MAE = 0.057). We envision that the application of these ANNs and other machine learning algorithms will accelerate the development of oxygen carrying materials for a range of chemical-looping applications and offer a rapid screening tool for new potential oxygen carriers.Item Open Access Dataset for Applying machine learning algorithms in estimating the performance of heterogeneous, multi-component materials as oxygen carriers for chemical-looping processes(Cranfield University, 2020-02-05 16:24) Yan, Yongliang; Clough, Peter; Anthony, BenThe training data for 19 manganese ores as potential oxygen carriers in the chemical-looping process from a fluidised-bed reactor. This database has been used to train several supervised artificial neural network models (ANN), which were used to predict the reactivity of the oxygen carriers with different fuels and the oxygen transfer capacity with only the knowledge of reactor bed temperature, elemental composition, and mechanical properties of the manganese ores. This novel approach explores ways of dealing with the training dataset, learning algorithms and topology of the ANN models to achieve enhanced prediction precision.Item Open Access Developments in calcium/chemical looping and metal oxide redox cycles for high-temperature thermochemical energy storage: A review(Elsevier, 2019-11-27) Yan, Yongliang; Wang, Ke; Clough, Peter T.; Anthony, Edward J.Energy storage is one of the most critical factors for maximising the availability of renewable energy systems while delivering firm capacity on an as- and when-required basis, thus improving the balance of grid energy. Chemical and calcium looping are two technologies, which are promising from both the point of view of minimising greenhouse gas emissions and because of their suitability for integrating with energy storage. A particularly promising route is to combine these technologies with solar heating, thus minimising the use of fossil fuels during the materials regeneration steps. For chemical looping, the development of mixed oxide carrier systems remains the highest impact research and development goal, and for calcium looping, minimising the decay in CO2 carrying capacity with natural sorbents appears to be the most economical option. In particular, sorbent stabilisers such as those based on Mg are particularly promising. In both cases, energy can be stored thermally as hot solids or chemically as unreacted materials, but there is a need to build suitable pilot plant demonstration units if the technology is to advance.Item Open Access Dynamic transformations of metals in the burning solid matter during combustion of heavy metal-contaminated biomass(American Chemical Society, 2021-05-10) Zha, Jianrui; Huang, Yaji; Zhu, Zhicheng; Yu, Mengzhu; Clough, Peter T.; Yan, Yongliang; Dong, Lu; Cheng, HaoqiangCombustion as an efficient and reliable method is widely used for metal-enriched biomass to achieve energy and metal recoveries, but there are emission risks of heavy metals in the flue gas and bottom ash that can give rise to secondary pollutions. To optimize such combustion processes, this work investigated the combustion characteristics of a kind of hyperaccumulator biomass and focused on the intermediate states and dynamic transformations of metals for the first time. A pseudo-in situ sampling method was used to collect the burning solid residues at different time intervals before further analysis. The conversions between elemental forms were revealed, and their conversion rates were also calculated. It was found that the transformation of metals was determined by their elemental natures, species distributions, and combustion progress where there was not a consecutive process but separated by several stages, which were related to (1) the release of volatile matters, (2) the formation and consumption of the char, and (3) the fixation by silicates. Based on the information of dynamic metal characteristics, a new strategy was proposed to optimize metal distribution by adjusting the combustion time of operations. The methodology introduced in this work will also help emission control and metal recovery for other metal-rich fuels.Item Open Access Green production of a novel sorbent from kaolin for capturing gaseous PbCl2 in a furnace(Elsevier, 2020-09-22) Zha, Jianrui; Huang, Yaji; Clough, Peter T.; Xia, Zhipeng; Zhu, Zhicheng; Fan, Conghui; Yu, Mengzhu; Yan, Yongliang; Cheng, HaoqiangThe pollution of semi-volatile heavy metals is one of the key environmental risks for municipal solid waste incineration, and in-situ adsorption of metals within the furnace by mineral sorbents such as kaolin has been demonstrated as a promising emission control method. To lessen the consumption of sorbent, a novel material of amorphous silicate was produced from kaolin through pressurised hydrothermal treatment. Its performance of gaseous PbCl2 capture was tested in a fixed bed furnace and compared with unmodified kaolin and metakaolin. With increasing temperature, the adsorption rates for all sorbents declined due to higher saturated vapour pressure, while the partitions of residual form lead increased which indicated higher stability of heavy metals in the sorbent because of melting effect. The new sorbent with a larger surface area and reformed structure presented 26% more adsorption efficiency than raw kaolin at 900 °C, and increasing the modification pressure improved these properties. Additionally, the production of this high-temperature sorbent was relatively inexpensive, required little thermal energy and no chemicals to produce and no waste effluent was generated, thus being much cleaner than other modification methods.Item Open Access Harnessing the power of machine learning for carbon capture, utilisation, and storage (CCUS) – A state-of-the-art review(Royal Society of Chemistry, 2021-11-01) Yan, Yongliang; Borhani, Tohid N.; Subraveti, Sai Gokul; Pai, Kasturi Nagesh; Prasad, Vinay; Rajendran, Arvind; Nkulikiyinka, Paula; Asibor, Jude Odianosen; Zhang, Zhien; Shao, Ding; Wang, Lijuan; Zhang, Wenbiao; Yan, Yong; Ampomah, William; You, Junyu; Wang, Meihong; Anthony, Edward J.; Manovic, Vasilije; Clough, Peter T.Carbon capture, utilisation and storage (CCUS) will play a critical role in future decarbonisation efforts to meet the Paris Agreement targets and mitigate the worst effects of climate change. Whilst there are many well developed CCUS technologies there is the potential for improvement that can encourage CCUS deployment. A time and cost-efficient way of advancing CCUS is through the application of machine learning (ML). ML is a collective term for high-level statistical tools and algorithms that can be used to classify, predict, optimise, and cluster data. Within this review we address the main steps of the CCUS value chain (CO2 capture, transport, utilisation, storage) and explore how ML is playing a leading role in expanding the knowledge across all fields of CCUS. We finish with a set of recommendations for further work and research that will develop the role that ML plays in CCUS and enable greater deployment of the technologies.Item Open Access Investigation of the apparent kinetics of air and oxy-fuel biomass combustion in a spouted fluidised-bed reactor(Elsevier, 2019-11-06) Yan, Yongliang; Clough, Peter T.; Anthony, Edward J.A bench-scale spouted fluidised-bed reactor was used to investigate the combustion kinetics of pulverised woody biomass under air and oxy-fuel atmospheres. Bed temperatures were in the range of 923-1073 K and O2 concentrations were varied from 20-35 vol%. The activation energies and apparent orders of reaction were calculated for air and oxy-fuel combustion by means of an nth order Arrhenius equation approach. Results indicated that the apparent order of reaction for both air and oxy-fuel combustion was approximately zero. The activation energies were calculated assuming a zero-order reaction mechanism and were averaged over all oxygen concentrations for air and oxy-fuel combustion and found to be 18.95 kJ/mol and 26.93 kJ/mol, respectively. The rate of combustion under oxy-fuel conditions was, on average, 37.5% higher compared to air combustion. The shrinking core model with a reaction-controlled step was found to accurately represent the biomass combustion reactions under both air and oxy-fuel conditions.Item Open Access Oxy-fuel and chemical-looping combustion for a low-carbon future.(Cranfield University, 2020-09) Yan, Yongliang; Clough, Peter T.; Manovic, Vasilije; Anthony, Edward J.This thesis is focused on investigating the potential of oxy-fuel and chemical- looping combustion (CLC) for carbon capture, and their integration with sorbent enhanced steam methane reforming (SE-SMR) for low-carbon hydrogen production. Oxy-fuel combustion converts a fuel within a mixture of O₂/CO₂ instead of air, while CLC converts a fuel by reduction of a metal oxide. In both cases, the resulting flue gas is free of N₂, and consist of only CO₂ and steam, and the steam can easily be condensed out. With the use of biomass as the fuel feedstock for the oxy-fuel combustion and CLC, negative CO₂ emissions can be achieved for power and heat generation. Oxy-fuel combustion is also a likely route to decarbonise the calcination of limestone, as used in the calcium looping and SE-SMR processes. SE-SMR combines the conventional steam methane reforming with calcium looping (CaL), which utilises CO₂ sorbents (e.g. CaO) to capture the CO₂ produced during the SMR process and shifts the equilibrium of the reforming and water-gas shift reactions in favour of more H₂ production according to Le Chatelier’s principle. Three main areas of work were conducted within this thesis, which includes 1) a detailed investigation into the effects of various parameters on the reaction kinetics of air and oxy-fuel combustion of woody biomass in a lab-scale fluidised- bed reactor; 2) applying machine learning in estimating the performance of oxygen carriers in chemical-looping processes; and 3) thermodynamic and techno-economic assessment of the integration of SE-SMR with oxy-fuel and chemical-looping combustion for low-carbon hydrogen production. Firstly, combustion rates of the biomass and its char were measured by a lab- scale fluidised-bed reactor. The shirking core model was used to simulate the char conversion during the experiments and under combustion mechanisms. Then, a novel approach that uses machine learning to efficiently screen the suitable oxygen carrier materials for CLC has been proposed. Lastly, the integration of oxy-fuel and CLC within the calciner of SE-SMR has been simulated in the Aspen Plus to understand their thermodynamic limitations and optimal operating conditions. Moreover, a detailed techno-economic analysis of the proposed configurations has been conducted to investigate their feasibility for a large-scare low-carbon hydrogen production. The obtained combustion kinetics and characteristics of air- and oxy-fuel combustion of biomass can provide useful information for retrofit and design of boilers. The framework of applying machine learning in oxygen carriers is expected to accelerate the finding and designing cost-effective oxygen carriers for large-scale CLC. The results of techno-economic analysis of the integration of oxy-fuel combustion and CLC with SE-SMR indicate that it is competitive with conventional steam methane reforming (SMR) with carbon capture and storage (CCS).Item Open Access Prediction of sorption enhanced steam methane reforming products from machine learning based soft-sensor models(Elsevier, 2020-11-11) Nkulikiyinka, Paula; Yan, Yongliang; Güleç, Fatih; Manovic, Vasilije; Clough, Peter T.Carbon dioxide-abated hydrogen can be synthesised via various processes, one of which is sorption enhanced steam methane reforming (SE-SMR), which produces separated streams of high purity H2 and CO2. Properties of hydrogen and the sorbent material hinder the ability to rapidly upscale SE-SMR, therefore the use of artificial intelligence models is useful in order to assist scale up. Advantages of a data driven soft-sensor model over thermodynamic simulations, is the ability to obtain real time information dependent on actual process conditions. In this study, two soft sensor models have been developed and used to predict and estimate variables that would otherwise be difficult direct measured. Both artificial neural networks and the random forest models were developed as soft sensor prediction models. They were shown to provide good predictions for gas concentrations in the reformer and regenerator reactors of the SE-SMR process using temperature, pressure, steam to carbon ratio and sorbent to carbon ratio as input process features. Both models were very accurate with high R2 values, all above 98%. However, the random forest model was more precise in the predictions, with consistently higher R2 values and lower mean absolute error (0.002-0.014) compared to the neural network model (0.005-0.024).Item Open Access Process simulation of blue hydrogen production by upgraded sorption enhanced steam methane reforming (SE-SMR) processes(Elsevier, 2020-07-21) Yan, Yongliang; Thanganadar, Dhinesh; Clough, Peter T.; Mukherjee, Sanjay; Patchigolla, Kumar; Manovic, Vasilije; Anthony, Edward J.Clean and carbon-free hydrogen production is expected to play a vital role in future global energy transitions. In this work, six process arrangements for sorption enhanced steam methane reforming (SE-SMR) are proposed for blue H2 production: 1) SE-SMR with an air fired calciner, 2) SE-SMR with a Pressure Swing Adsorption (PSA) unit, 3) SE-SMR thermally coupled with Chemical-Looping Combustion (CLC), 4) SE-SMR+PSA+CLC, 5) SE-SMR+PSA with an oxy-fired calciner, 6) SE-SMR+PSA and indirect firing H2 combustion from the product stream recycle. The proposed process models with rigorous heat exchanger network design were simulated in Aspen Plus to understand the thermodynamic limitations in achieving the maximum CH4 conversion, H2 purity, CO2 capture efficiency, cold gas efficiency and net operating efficiency. A sensitivity study was also performed for changes in the reformer temperature, pressure, and steam to carbon (S/C) ratio to explore the optimal operating space for each case. The SE-SMR+PSA+H2 (Case 6) recycle process can achieve a maximum of 94.2% carbon capture with a trade-off in cold gas efficiency (51.3%), while a near 100% carbon capture with the maximum net efficiency of up to 76.3% is realisable by integrating CLC and PSA (Case 4) at 25 bar. Integration of oxy-fuel combustion lowered the net efficiency by 2.7% points due to the need for an air separation unit. In addition, the SE-SMR with the PSAOG process can be designed as a self-sustaining process without any additional fuel required to meet the process heat utility when the S/C ratio is ~3-3.5Item Open Access Process simulations of high-purity and renewable clean H2 production by sorption enhanced steam reforming of biogas(American Chemical Society, 2023-03-11) Capa, Alma; Yan, Yongliang; Rubiera, Fernando; Pevida, Covadonga; Gil, María Victoria; Clough, Peter T.Renewable clean H2 has a very promising potential for the decarbonization of energy systems. Sorption enhanced steam reforming (SESR) is a novel process that combines the steam reforming reaction and the simultaneous CO2 removal by a solid sorbent, such as CaO, which significantly enhances hydrogen generation, enabling high-purity H2 production. The CO2 sorption reaction (carbonation) is exothermic, but the sorbent regeneration by calcination is highly endothermic, which requires extra energy. Biogas is one of the available carbon-neutral renewable H2 production sources. It can be especially relevant for the energy integration of the SESR process since, due to the exothermic sorption reaction, the CO2 contained in the biogas provides extra heat to the system, which can help to balance the energy requirements of the process. This work studies different process configurations for the energy integration of the SESR process of biogas for high-purity renewable H2 production: (1) SESR with sorbent regeneration using a portion of the produced H2 (SESR+REG_H2), (2) SESR with sorbent regeneration using biogas (SESR+REG_BG), and (3) SESR with sorbent regeneration using biogas and adding a pressure swing adsorption (PSA) unit for hydrogen purification (SESR+REG_BG+PSA). When using biogas as fuel (Cases 2 and 3), these configurations were studied using air and oxy-fuel combustion atmospheres in the sorbent regeneration step, resulting in five case studies. A thermodynamic approach for process modeling can provide the optimal process operating conditions and configurations that maximize the energy efficiency of the process, which are the basis for subsequent optimization of the process at the practical level needed to scale up this technology. For this purpose, process simulations were performed using a steady-state plant model developed in Aspen Plus, incorporating a complex heat exchanger network (HEN) to optimize heat integration. A comprehensive parametric study assessed the effects of biogas composition, temperature, pressure, and steam to methane (S/CH4) ratio on the process performance represented by the selected key performance indicators, i.e., H2 purity, H2 yield, CH4 conversion, cold gas efficiency (CGE), net efficiency (NE), fuel consumption for the sorbent regeneration step, and CO2 capture efficiency. H2 with a purity of 98.5 vol % and a CGE of 75.7% with zero carbon emissions can be achieved. When adding a PSA unit, nearly 100% H2 purity and CO2 capture efficiency were achieved with a CGE of 77.3%. The use of oxy-fuel combustion during regeneration lowered the net efficiency of the process by 2.3% points (since it requires an air separation unit) but allowed the process to achieve negative carbon emissions.Item Open Access Sorption-enhanced steam methane reforming for combined CO2 capture and hydrogen production: a state-of-the-art review(Elsevier, 2021-10-02) Masoudi Soltani, Salman; Lahiri, Abhishek; Bahzad, Husain; Clough, Peter T.; Gorbounov, Mikhail; Yan, YongliangThe European Commission have just stated that hydrogen would play a major role in the economic recovery of post-COVID-19 EU countries. Hydrogen is recognised as one of the key players in a fossil fuel-free world in decades to come. However, commercially practiced pathways to hydrogen production todays, are associated with a considerable amount of carbon emissions. The Paris Climate Change Agreement has set out plans for an international commitment to reduce carbon emissions within the forthcoming decades. A sustainable hydrogen future would only be achievable if hydrogen production is “designed” to capture such emissions. Today, nearly 98% of global hydrogen production relies on the utilisation of fossil fuels. Among these, steam methane reforming (SMR) boasts the biggest share of nearly 50% of the global generation. SMR processes correspond to a significant amount of carbon emissions at various points throughout the process. Despite the dark side of the SMR processes, they are projected to play a major role in hydrogen production by the first half of this century. This that a sustainable, yet clean short/medium-term hydrogen production is only possible by devising a plan to efficiently capture this co-produced carbon as stated in the latest International Energy Agency (IEA) reports. Here, we have carried out an in-depth technical review of the processes employed in sorption-enhanced steam methane reforming (SE-SMR), an emerging technology in low-carbon SMR, for combined carbon capture and hydrogen production. This paper aims to provide an in-depth review on two key challenging elements of SE-SMR i.e. the advancements in catalysts/adsorbents preparation, and current approaches in process synthesis and optimisation including the employment of artificial intelligence in SE-SMR processes. To the best of the authors’ knowledge, there is a clear gap in the literature where the above areas have been scrutinised in a systematic and coherent fashion. The gap is even more pronounced in the application of AI in SE-SMR technologies. As a result, this work aims to fill this gap within the scientific literature.Item Open Access Techno-economic analysis of low-carbon hydrogen production by sorption enhanced steam methane reforming (SE-SMR) processes(Elsevier, 2020-10-30) Yan, Yongliang; Manovic, Vasilije; Anthony, Edward J.; Clough, Peter T.Hydrogen is an attractive energy carrier that will play a key role in future global energy transitions. This work investigates the techno-economic performance of six different sorption enhanced steam methane reforming (SE-SMR) configurations integrated with an indirect natural gas or biomass-fired calciner, oxy-fuel combustion and chemical-looping combustion for large-scale blue and carbon-negative hydrogen production. The techno-economic performance of the proposed cases was evaluated by their net efficiency, CO2 capture efficiency, levelised cost of hydrogen (LCOH), and costs of CO2 avoided and removal. A sensitivity analysis was also conducted to evaluate the key parameters and explore existing uncertainties that can affect the economic performance of the proposed SE-SMR processes. The results revealed that the proposed systems were comparable with conventional steam methane reforming (SMR) with carbon capture and storage (CCS). The LCOH of the proposed SE-SMR plants ranged from £1.90–2.80/kg, and the costs of CO2 avoided and removal ranged from £33-69/tonne and £58-107/tonne, respectively. By applying a carbon price (£16/tonne CO2), the costs of CO2 avoided and removal for the proposed SE-SMR processes could be significantly reduced. The results of cumulative discounted cash flow of SE-SMR plants at a hydrogen selling price of £3.00/kg indicated that all the investment of the proposed cases could be paid back after eight years, even if the carbon tax is zero.