In the era of Industry 4.0, substantial research has been devoted to the field of fault detection and diagnosis (FDD), which plays a critical role in preventive maintenance of large chemical processes. However, the existing studies are primarily focused on few‐shot samples of process data and without considering the role of activation functions in temporal diagnostic tasks. In this paper, an end‐to‐end chemical fault diagnosis framework that combines bidirectional long short‐term memory (LSTM) with attention mechanism is proposed. In the preprocessing stage, a special sliding time window function is developed to integrate multivariate samples containing complex temporal information via operation such as subset extraction. Afterwards, the bidirectional LSTM is constructed to address dynamic and temporal relationship on longer series observation, and the attention mechanism is adopted to highlight key fault features by assigning different attention weights. A case application is performed on the enriched Tennessee Eastman process (TEP), which reduces the bias between sample statistics and larger population parameters compared to existing few‐shot sample studies. The metric evaluation experiments for six activations show that the model configured with tanh function can achieve the optimal tradeoff in chemical process tasks, providing a strong benchmark for subsequent fault diagnosis research.
{"title":"A novel fault diagnosis framework empowered by LSTM and attention: A case study on the Tennessee Eastman process","authors":"Shuaiyu Zhao, Yiling Duan, Nitin Roy, Bin Zhang","doi":"10.1002/cjce.25460","DOIUrl":"https://doi.org/10.1002/cjce.25460","url":null,"abstract":"In the era of Industry 4.0, substantial research has been devoted to the field of fault detection and diagnosis (FDD), which plays a critical role in preventive maintenance of large chemical processes. However, the existing studies are primarily focused on few‐shot samples of process data and without considering the role of activation functions in temporal diagnostic tasks. In this paper, an end‐to‐end chemical fault diagnosis framework that combines bidirectional long short‐term memory (LSTM) with attention mechanism is proposed. In the preprocessing stage, a special sliding time window function is developed to integrate multivariate samples containing complex temporal information via operation such as subset extraction. Afterwards, the bidirectional LSTM is constructed to address dynamic and temporal relationship on longer series observation, and the attention mechanism is adopted to highlight key fault features by assigning different attention weights. A case application is performed on the enriched Tennessee Eastman process (TEP), which reduces the bias between sample statistics and larger population parameters compared to existing few‐shot sample studies. The metric evaluation experiments for six activations show that the model configured with tanh function can achieve the optimal tradeoff in chemical process tasks, providing a strong benchmark for subsequent fault diagnosis research.","PeriodicalId":501204,"journal":{"name":"The Canadian Journal of Chemical Engineering","volume":"127 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142195304","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Ahmet Alp Zembat, Elifnur Gezmis‐Yavuz, Derya Y. Koseoglu‐Imer, C. Elif Cansoy
The global challenge of providing clean water at an affordable cost has led to the need for the development of low‐cost and non‐toxic materials for the treatment and recycling of waste water. Nanofibres have emerged as a promising solution due to their superior properties. To this end, composite polyamide‐6 (PA‐6) nanofibres embedded with graphene nanoplatelets (GNPs) were prepared by electrospinning. The study investigated the effect of the ratio of GNPs, which ranged from 0.1 to 1.0 wt.%, on the mechanical properties of nanofibres and the removal of turbidity. The results showed that PA‐6 nanofibres with 0.5 wt.% GNP exhibited enhanced mechanical properties, and increasing the GNP ratio led to lower turbidity values. To the best of our knowledge, GNP‐embedded PA‐6 nanofibres have not been used for turbidity removal before, and these filter materials are promising due to their excellent fibre structure, mechanical strength, and high level of turbidity removal.
{"title":"The use of graphene nanoplatelet‐embedded PA‐6 nanofibres to remove turbidity from water","authors":"Ahmet Alp Zembat, Elifnur Gezmis‐Yavuz, Derya Y. Koseoglu‐Imer, C. Elif Cansoy","doi":"10.1002/cjce.25469","DOIUrl":"https://doi.org/10.1002/cjce.25469","url":null,"abstract":"The global challenge of providing clean water at an affordable cost has led to the need for the development of low‐cost and non‐toxic materials for the treatment and recycling of waste water. Nanofibres have emerged as a promising solution due to their superior properties. To this end, composite polyamide‐6 (PA‐6) nanofibres embedded with graphene nanoplatelets (GNPs) were prepared by electrospinning. The study investigated the effect of the ratio of GNPs, which ranged from 0.1 to 1.0 wt.%, on the mechanical properties of nanofibres and the removal of turbidity. The results showed that PA‐6 nanofibres with 0.5 wt.% GNP exhibited enhanced mechanical properties, and increasing the GNP ratio led to lower turbidity values. To the best of our knowledge, GNP‐embedded PA‐6 nanofibres have not been used for turbidity removal before, and these filter materials are promising due to their excellent fibre structure, mechanical strength, and high level of turbidity removal.","PeriodicalId":501204,"journal":{"name":"The Canadian Journal of Chemical Engineering","volume":"109 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142195295","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhenhua Yu, Wenjing Wang, Xueting Wang, Qingchao Jiang, Guan Wang
In industrial processes, complex nonlinearity and dynamics generally exist, making it challenging to achieve good results using conventional process monitoring methods. In this paper, a latent mapping embedding neural network method (LMNN) is proposed for efficient monitoring of nonlinear dynamic processes. First, a deep neural network (DNN) is employed to acquire features of state variables from nonlinear process data and expand them along with the input to a new feature subspace. Second, a latent mapping (LM) method is used to map the high‐dimensional feature subspace to a low‐dimensional subspace that includes the most beneficial time series information. Then the entire neural network and regression parameters are obtained through an end‐to‐end learning manner, through which the nonlinearity and process dynamics are well characterized. Subsequently, prediction error‐based residual is generated and the monitoring model is established. The performance of the proposed method is verified through a simulation of penicillin production process and an actual fermentation process of penicillin. Comparisons with state‐of‐the‐art methods are carried out, and results validate the effectiveness and superiority of the proposed method.
{"title":"Nonlinear dynamic process monitoring based on latent mapping embedding deep neural networks","authors":"Zhenhua Yu, Wenjing Wang, Xueting Wang, Qingchao Jiang, Guan Wang","doi":"10.1002/cjce.25461","DOIUrl":"https://doi.org/10.1002/cjce.25461","url":null,"abstract":"In industrial processes, complex nonlinearity and dynamics generally exist, making it challenging to achieve good results using conventional process monitoring methods. In this paper, a latent mapping embedding neural network method (LMNN) is proposed for efficient monitoring of nonlinear dynamic processes. First, a deep neural network (DNN) is employed to acquire features of state variables from nonlinear process data and expand them along with the input to a new feature subspace. Second, a latent mapping (LM) method is used to map the high‐dimensional feature subspace to a low‐dimensional subspace that includes the most beneficial time series information. Then the entire neural network and regression parameters are obtained through an end‐to‐end learning manner, through which the nonlinearity and process dynamics are well characterized. Subsequently, prediction error‐based residual is generated and the monitoring model is established. The performance of the proposed method is verified through a simulation of penicillin production process and an actual fermentation process of penicillin. Comparisons with state‐of‐the‐art methods are carried out, and results validate the effectiveness and superiority of the proposed method.","PeriodicalId":501204,"journal":{"name":"The Canadian Journal of Chemical Engineering","volume":"22 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142195299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
wPressure loss measurements are presented for packed beds of multi‐component mixtures of thin angular parallelepipeds and of random wood chips for a Reynolds number range of 50 to 500. For flat particles like these, the degree to which the particles overlap is an essential factor in pressure loss, and this was measured using two different methods, including a novel technique involving progressive dismantling and photography of the bed. The experimental friction factors were found to be well represented by the Nemec and Levec pressure loss correlation, an Ergun‐type equation with an explicit dependence of the parameters on particle sphericity, with the equation expanded to include the effects of particle overlap and of packing anomalies at the wall. The friction losses of the mixtures were found to be somewhat higher than those of the individual component particles, requiring a minor change in the correlation parameters. Estimates of the tortuosity of the bed channels showed that the greater losses of the mixtures correspond to an increase in tortuosity.
w 本文介绍了在雷诺数为 50 到 500 的范围内，对多组分薄角平行椭圆形混合物和随机木屑的填料床进行的压力损失测量。对于像这样的扁平颗粒，颗粒的重叠程度是压力损失的一个重要因素，我们使用两种不同的方法测量了这一因素，包括一种涉及床层逐步拆卸和拍照的新技术。实验发现，Nemec 和 Levec 压力损失相关性很好地反映了实验摩擦因数，这是一个厄尔贡式方程，其参数与颗粒球度有明确的关系，方程扩展后包括了颗粒重叠和壁面填料异常的影响。结果发现，混合物的摩擦损耗略高于单个成分颗粒的摩擦损耗，因此需要对相关参数稍作修改。对床层通道曲折度的估算表明，混合物的更大损失与曲折度的增加相对应。
{"title":"Pressure loss in packed beds of multicomponent mixtures of flat particles with particle overlap, including random chips","authors":"Evangelina Schonfeldt, William L. H. Hallett","doi":"10.1002/cjce.25471","DOIUrl":"https://doi.org/10.1002/cjce.25471","url":null,"abstract":"wPressure loss measurements are presented for packed beds of multi‐component mixtures of thin angular parallelepipeds and of random wood chips for a Reynolds number range of 50 to 500. For flat particles like these, the degree to which the particles overlap is an essential factor in pressure loss, and this was measured using two different methods, including a novel technique involving progressive dismantling and photography of the bed. The experimental friction factors were found to be well represented by the Nemec and Levec pressure loss correlation, an Ergun‐type equation with an explicit dependence of the parameters on particle sphericity, with the equation expanded to include the effects of particle overlap and of packing anomalies at the wall. The friction losses of the mixtures were found to be somewhat higher than those of the individual component particles, requiring a minor change in the correlation parameters. Estimates of the tortuosity of the bed channels showed that the greater losses of the mixtures correspond to an increase in tortuosity.","PeriodicalId":501204,"journal":{"name":"The Canadian Journal of Chemical Engineering","volume":"88 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142195303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the development of technology, the chemical production process is becoming increasingly complex and large‐scale, making fault detection particularly important. However, current detection methods struggle to address the complexities of large‐scale production processes. In this paper, we integrate the strengths of deep learning and machine learning technologies, combining the advantages of bidirectional long‐ and short‐term memory neural networks, fully connected neural networks, and the extra trees algorithm to propose a novel fault detection model named three‐layer deep learning network random trees (TDLN‐trees). First, the deep learning component extracts temporal features from industrial data, combining and transforming them into a higher‐level data representation. Second, the machine learning component processes and classifies the features extracted in the first step. An experimental analysis based on the Tennessee Eastman process verifies the superiority of the proposed method.
{"title":"Three‐layer deep learning network random trees for fault detection in chemical production process","authors":"Ming Lu, Zhen Gao, Ying Zou, Zuguo Chen, Pei Li","doi":"10.1002/cjce.25465","DOIUrl":"https://doi.org/10.1002/cjce.25465","url":null,"abstract":"With the development of technology, the chemical production process is becoming increasingly complex and large‐scale, making fault detection particularly important. However, current detection methods struggle to address the complexities of large‐scale production processes. In this paper, we integrate the strengths of deep learning and machine learning technologies, combining the advantages of bidirectional long‐ and short‐term memory neural networks, fully connected neural networks, and the extra trees algorithm to propose a novel fault detection model named three‐layer deep learning network random trees (TDLN‐trees). First, the deep learning component extracts temporal features from industrial data, combining and transforming them into a higher‐level data representation. Second, the machine learning component processes and classifies the features extracted in the first step. An experimental analysis based on the Tennessee Eastman process verifies the superiority of the proposed method.","PeriodicalId":501204,"journal":{"name":"The Canadian Journal of Chemical Engineering","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142195300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mojtaba Mohammadi, Mohammadreza Nofar, Pierre J. Carreau
Blend nanocomposites of amorphous polylactide (aPLA) and semicrystalline PLA (scPLA)‐multiwalled carbon nanotubes (MWCNTs) were prepared by a twin‐screw extruder below the melting temperature of the scPLA. The maximum weight percent of MWCNTs in the blends was 0.9 wt.%. The extrudates were either pelletized immediately or after drawing at a drawing ratio of about 10. According to small amplitude oscillatory shear rheological analysis, the rheological properties of the aPLA/scPLA (85/15 wt.%) drawn sample were significantly increased compared to the undrawn samples. With the presence of MWCNTs, more crystallites could develop in the scPLA, and the electrical conductivity of the aPLA/scPLA nanocomposites was reduced due to the encapsulation of MWCNTs within the crystallites of scPLA. Increasing the temperature during compression moulding to 190°C, which is above the melting temperature of the scPLA, effectively removed this obstacle and the electrical conductivity was increased by a factor of up to 106 compared to the samples moulded at 150°C.
{"title":"Properties of blends of amorphous and semicrystalline PLAs containing multiwalled carbon nanotubes","authors":"Mojtaba Mohammadi, Mohammadreza Nofar, Pierre J. Carreau","doi":"10.1002/cjce.25463","DOIUrl":"https://doi.org/10.1002/cjce.25463","url":null,"abstract":"Blend nanocomposites of amorphous polylactide (aPLA) and semicrystalline PLA (scPLA)‐multiwalled carbon nanotubes (MWCNTs) were prepared by a twin‐screw extruder below the melting temperature of the scPLA. The maximum weight percent of MWCNTs in the blends was 0.9 wt.%. The extrudates were either pelletized immediately or after drawing at a drawing ratio of about 10. According to small amplitude oscillatory shear rheological analysis, the rheological properties of the aPLA/scPLA (85/15 wt.%) drawn sample were significantly increased compared to the undrawn samples. With the presence of MWCNTs, more crystallites could develop in the scPLA, and the electrical conductivity of the aPLA/scPLA nanocomposites was reduced due to the encapsulation of MWCNTs within the crystallites of scPLA. Increasing the temperature during compression moulding to 190°C, which is above the melting temperature of the scPLA, effectively removed this obstacle and the electrical conductivity was increased by a factor of up to 10<jats:sup>6</jats:sup> compared to the samples moulded at 150°C.","PeriodicalId":501204,"journal":{"name":"The Canadian Journal of Chemical Engineering","volume":"24 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142195307","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhong Tang, Zhenzhong Li, Shanglong Huang, Chen Yang
The existing researches lack the comprehensive comparison of the performance of two‐fluid model (TFM) and computational fluid dynamics‐discrete element model (CFD‐DEM) using a cylindrical fluidized bed as a research object. In addition, the applicability of rotational periodic boundary conditions in CFD‐DEM simulations of cylindrical fluidized beds is still unclear. Therefore, taking cylindrical fluidized bed as the object and studying the performance of different simulation methods can provide guidance for the selection of simulation methods in subsequent related studies. In the present study, TFM and coarse‐grained CFD‐DEM were used in simulations of the fluidized bed to evaluate the performance of different numerical methods. Furthermore, the applicability of rotating periodic boundary conditions in CFD‐DEM simulations was investigated. The results show that TFM and coarse‐grained CFD‐DEM perform in general agreement in predicting macro variables (e.g., overall pressure drop and bed height). However, radial void fraction distribution and void fraction probability density function (PDF) distribution of CFD‐DEM agreed better with the experimental data. CFD‐DEM simulations with rotational periodic boundary conditions applied showed lower predicted void fraction PDF peaks at packed bed heights and poorly modelling particle mixing in the central of cylindrical fluidized bed due to changes in the boundary conditions as well as the number of particle parcels. Therefore, both TFM and CFD‐DEM can obtain reasonable macro variables, but CFD‐DEM predicted more accurate gas–solid two‐phase distribution. The CFD‐DEM with rotating periodic boundary conditions could not reasonably predict the pressure drop and gas–solid two‐phase distribution inside the cylindrical fluidized bed.
{"title":"Numerical study of gas–solid flow characteristics of cylindrical fluidized beds based on coarse‐grained CFD‐DEM method","authors":"Zhong Tang, Zhenzhong Li, Shanglong Huang, Chen Yang","doi":"10.1002/cjce.25455","DOIUrl":"https://doi.org/10.1002/cjce.25455","url":null,"abstract":"The existing researches lack the comprehensive comparison of the performance of two‐fluid model (TFM) and computational fluid dynamics‐discrete element model (CFD‐DEM) using a cylindrical fluidized bed as a research object. In addition, the applicability of rotational periodic boundary conditions in CFD‐DEM simulations of cylindrical fluidized beds is still unclear. Therefore, taking cylindrical fluidized bed as the object and studying the performance of different simulation methods can provide guidance for the selection of simulation methods in subsequent related studies. In the present study, TFM and coarse‐grained CFD‐DEM were used in simulations of the fluidized bed to evaluate the performance of different numerical methods. Furthermore, the applicability of rotating periodic boundary conditions in CFD‐DEM simulations was investigated. The results show that TFM and coarse‐grained CFD‐DEM perform in general agreement in predicting macro variables (e.g., overall pressure drop and bed height). However, radial void fraction distribution and void fraction probability density function (PDF) distribution of CFD‐DEM agreed better with the experimental data. CFD‐DEM simulations with rotational periodic boundary conditions applied showed lower predicted void fraction PDF peaks at packed bed heights and poorly modelling particle mixing in the central of cylindrical fluidized bed due to changes in the boundary conditions as well as the number of particle parcels. Therefore, both TFM and CFD‐DEM can obtain reasonable macro variables, but CFD‐DEM predicted more accurate gas–solid two‐phase distribution. The CFD‐DEM with rotating periodic boundary conditions could not reasonably predict the pressure drop and gas–solid two‐phase distribution inside the cylindrical fluidized bed.","PeriodicalId":501204,"journal":{"name":"The Canadian Journal of Chemical Engineering","volume":"109 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142195338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Kanta Nakano, Numan Luthfi, Takashi Fukushima, Kenji Takisawa
Recently, the depletion of fossil fuels has become an issue, prompting the search for sustainable alternatives. Algal biomass has gained considerable attention as a promising renewable energy source because of its high production efficiency and adaptability to external environment. However, its high‐moisture content escalates the energy requirement during the thermal drying process in algal biomass production. Thus, we proposed a new energy production system using hydrothermal carbonization, which requires no pretreatment even for high moisture content biomass, making it compatible with such materials. Herein, we investigated the decrease in moisture content of algal biomass through hydrothermal carbonization and its effect on the energy production and energy balance of algal biomass. The results showed that hydrothermal carbonization at 240°C for 3 h produced hydrochar with a moisture content of 34.6%. It was found that it was due to changes in surface structures, such as CH, CO, and OH functional groups, using scanning electron microscopy (SEM) and Fourier transform infrared (FT‐IR) analysis. However, the greatest reduction in production energy, 45%, was achieved at 240°C for 4 h. The optimal energy balance was obtained for hydrothermal carbonization at 220°C for 4 h, for which energy production was 2.7 times more efficient than that achieved by conventional methods.
{"title":"Optimizing hydrothermal carbonization for enhanced energy production from algal biomass with high moisture content","authors":"Kanta Nakano, Numan Luthfi, Takashi Fukushima, Kenji Takisawa","doi":"10.1002/cjce.25457","DOIUrl":"https://doi.org/10.1002/cjce.25457","url":null,"abstract":"Recently, the depletion of fossil fuels has become an issue, prompting the search for sustainable alternatives. Algal biomass has gained considerable attention as a promising renewable energy source because of its high production efficiency and adaptability to external environment. However, its high‐moisture content escalates the energy requirement during the thermal drying process in algal biomass production. Thus, we proposed a new energy production system using hydrothermal carbonization, which requires no pretreatment even for high moisture content biomass, making it compatible with such materials. Herein, we investigated the decrease in moisture content of algal biomass through hydrothermal carbonization and its effect on the energy production and energy balance of algal biomass. The results showed that hydrothermal carbonization at 240°C for 3 h produced hydrochar with a moisture content of 34.6%. It was found that it was due to changes in surface structures, such as CH, CO, and OH functional groups, using scanning electron microscopy (SEM) and Fourier transform infrared (FT‐IR) analysis. However, the greatest reduction in production energy, 45%, was achieved at 240°C for 4 h. The optimal energy balance was obtained for hydrothermal carbonization at 220°C for 4 h, for which energy production was 2.7 times more efficient than that achieved by conventional methods.","PeriodicalId":501204,"journal":{"name":"The Canadian Journal of Chemical Engineering","volume":"26 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142195343","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mark Werner Hlawitschka, Andreas Schleiffer, Jonas Schurr, Stephan Winkler, Daniel Danner
A novel micro‐channel technique for analyzing the coalescence of bubbles and obtaining relevant information for the creation of a coalescence database is presented. The micro‐channel improves the coalescence investigations by a continuously operated setup, reduces the accumulation of impurities and increases the amount of recorded data. To introduce the new setup, studies with alcoholic, electrolytic aqueous systems and liquid silicone oil as a second liquid are presented, showing the influence of different concentrations. Artificial intelligence has been successfully developed to automate data generation. This approach improves the understanding of bubble coalescence by introducing a reproducible setup. Furthermore, it facilitates the transition to a predictive column design through data‐based decisions and modelling.
{"title":"Coalescence investigations in a small‐scale continuously operated setup for bubble column design","authors":"Mark Werner Hlawitschka, Andreas Schleiffer, Jonas Schurr, Stephan Winkler, Daniel Danner","doi":"10.1002/cjce.25458","DOIUrl":"https://doi.org/10.1002/cjce.25458","url":null,"abstract":"A novel micro‐channel technique for analyzing the coalescence of bubbles and obtaining relevant information for the creation of a coalescence database is presented. The micro‐channel improves the coalescence investigations by a continuously operated setup, reduces the accumulation of impurities and increases the amount of recorded data. To introduce the new setup, studies with alcoholic, electrolytic aqueous systems and liquid silicone oil as a second liquid are presented, showing the influence of different concentrations. Artificial intelligence has been successfully developed to automate data generation. This approach improves the understanding of bubble coalescence by introducing a reproducible setup. Furthermore, it facilitates the transition to a predictive column design through data‐based decisions and modelling.","PeriodicalId":501204,"journal":{"name":"The Canadian Journal of Chemical Engineering","volume":"77 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142195344","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Fluidized bed technology has a 100‐year history of delivering energy solutions to the world. Examples include fluid catalytic cracking, coal combustion and gasification, and fluid coking. Moving forward, fluidization technology has the potential to underpin the development of entirely new sustainable processes in the energy transition and the circular economy and many of these will be advanced by small‐and‐medium enterprises (SMEs) and start‐ups. Focused, low‐cost, and time‐bound research outcomes will be needed to support these SMEs as they bring their new technologies to market as quickly as possible. This paper first summarizes some of the fluidized bed technologies that will play a key role in the energy transition and then considers how the strategic concept of discovery driven growth can lead to focused, rapid, and low‐cost information. The experimental data can then be used to develop hybrid models using machine learning methods that will be more robust, accurate, and reliable models. With focused, interdisciplinary research, fluidization models may be developed that would allow fluidized beds to go directly from lab or pilot scale directly to commercial. This would reduce development costs and timelines dramatically, hence bringing these new technologies to market more quickly. Early commercialization will allow the environmental benefits to begin to accrue earlier and will improve returns on investment.
{"title":"Fluidized bed applications and modern scale‐up tools for the energy transition","authors":"Todd Pugsley","doi":"10.1002/cjce.25420","DOIUrl":"https://doi.org/10.1002/cjce.25420","url":null,"abstract":"Fluidized bed technology has a 100‐year history of delivering energy solutions to the world. Examples include fluid catalytic cracking, coal combustion and gasification, and fluid coking. Moving forward, fluidization technology has the potential to underpin the development of entirely new sustainable processes in the energy transition and the circular economy and many of these will be advanced by small‐and‐medium enterprises (SMEs) and start‐ups. Focused, low‐cost, and time‐bound research outcomes will be needed to support these SMEs as they bring their new technologies to market as quickly as possible. This paper first summarizes some of the fluidized bed technologies that will play a key role in the energy transition and then considers how the strategic concept of discovery driven growth can lead to focused, rapid, and low‐cost information. The experimental data can then be used to develop hybrid models using machine learning methods that will be more robust, accurate, and reliable models. With focused, interdisciplinary research, fluidization models may be developed that would allow fluidized beds to go directly from lab or pilot scale directly to commercial. This would reduce development costs and timelines dramatically, hence bringing these new technologies to market more quickly. Early commercialization will allow the environmental benefits to begin to accrue earlier and will improve returns on investment.","PeriodicalId":501204,"journal":{"name":"The Canadian Journal of Chemical Engineering","volume":"74 23","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141643219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：